From icculus at clutteredmind.org Thu Sep 1 01:19:53 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Thu, 01 Sep 2005 01:19:53 -0400 Subject: [quake3] Sgi IRIX port.. Message-ID: <43168F79.9030105@clutteredmind.org> > Yes, but the comment in the makefile sounded like some IRIX-specific > parts could be missing from the released q3 source. Some time ago > (around the q3 1.17 times) an Sgi employee had accesss to the Loki > codebase and ported q3 to IRIX. The binaries produced at that time > don't work anymore and I was wondering if q3 could be ported to IRIX > again. I just want to make sure that I am not missing any of the > modified sources as I wouldn't want to duplicate someone else's work.. I think this was lost, but you can ask TTimo...if he doesn't have it, it might be easier to just use the new SDL code and let it handle all the platform stuff...that'll probably Just Work once you get it compiling. --ryan. From zakk at timedoctor.org Thu Sep 1 01:30:37 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 31 Aug 2005 22:30:37 -0700 Subject: Infostring thing Message-ID: <20050901053037.GA12318@timedoctor.org> Could somebody test the fix for the infostring bug/feature/crack: https://bugzilla.icculus.org/show_bug.cgi?id=2356 I applied but can't test right now. http://bugs.gentoo.org/show_bug.cgi?id=82149 has more info -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Thu Sep 1 01:56:02 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 31 Aug 2005 22:56:02 -0700 Subject: Make clean Message-ID: <20050901055602.GC12318@timedoctor.org> Make clean doesn't seem to actually work right, if somebody has two seconds that would be a good thing to fix. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From jaq at spacepants.org Thu Sep 1 01:57:31 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Thu, 1 Sep 2005 15:57:31 +1000 Subject: [quake3] Make clean In-Reply-To: <20050901055602.GC12318@timedoctor.org> References: <20050901055602.GC12318@timedoctor.org> Message-ID: <20050901055731.GE11993@spacepants.org> This one time, at band camp, Zachary J. Slater wrote: >Make clean doesn't seem to actually work right, if somebody has two >seconds that would be a good thing to fix. Are you really using the makefile, or are you using cons to build? From zakk at timedoctor.org Thu Sep 1 02:08:17 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 31 Aug 2005 23:08:17 -0700 Subject: [quake3] Make clean In-Reply-To: <20050901055731.GE11993@spacepants.org> References: <20050901055602.GC12318@timedoctor.org> <20050901055731.GE11993@spacepants.org> Message-ID: <20050901060817.GA19026@timedoctor.org> On Thu, Sep 01, 2005 at 03:57:31PM +1000, Jamie Wilkinson wrote: > Are you really using the makefile, or are you using cons to build? Using make in code/unix -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From jaq at spacepants.org Thu Sep 1 02:25:02 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Thu, 1 Sep 2005 16:25:02 +1000 Subject: [quake3] Make clean In-Reply-To: <20050901060817.GA19026@timedoctor.org> References: <20050901055602.GC12318@timedoctor.org> <20050901055731.GE11993@spacepants.org> <20050901060817.GA19026@timedoctor.org> Message-ID: <20050901062502.GF11993@spacepants.org> This one time, at band camp, Zachary J. Slater wrote: >On Thu, Sep 01, 2005 at 03:57:31PM +1000, Jamie Wilkinson wrote: >> Are you really using the makefile, or are you using cons to build? > >Using make in code/unix Whoa. I didn't think the Makefile was up to date -- you needed to remove all the references to the freetype code, didn't you? I've been using cons, as per the README, got a clean build without needing to hack the source like the Makefile did; and if cons is anything like SCons, you shouldn't need to ever run clean to get a correct build :) From zakk at timedoctor.org Thu Sep 1 02:32:05 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 31 Aug 2005 23:32:05 -0700 Subject: [quake3] Make clean In-Reply-To: <20050901062502.GF11993@spacepants.org> References: <20050901055602.GC12318@timedoctor.org> <20050901055731.GE11993@spacepants.org> <20050901060817.GA19026@timedoctor.org> <20050901062502.GF11993@spacepants.org> Message-ID: <20050901063205.GB19026@timedoctor.org> On Thu, Sep 01, 2005 at 04:25:02PM +1000, Jamie Wilkinson wrote: > Whoa. I didn't think the Makefile was up to date -- you needed to remove > all the references to the freetype code, didn't you? > > I've been using cons, as per the README, got a clean build without needing > to hack the source like the Makefile did; and if cons is anything like SCons, > you shouldn't need to ever run clean to get a correct build :) We've only been updating the makefile, read the i_o-q3-readme, I'll rename the id one to prevent confusion. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From vincent at cojot.name Thu Sep 1 02:40:25 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Thu, 1 Sep 2005 08:40:25 +0200 (CEST) Subject: [quake3] Sgi IRIX port.. In-Reply-To: <43168F79.9030105@clutteredmind.org> References: <43168F79.9030105@clutteredmind.org> Message-ID: On Thu, 1 Sep 2005, Ryan C. Gordon wrote: > I think this was lost, but you can ask TTimo...if he doesn't have it, it > might be easier to just use the new SDL code and let it handle all the > platform stuff...that'll probably Just Work once you get it compiling. > > --ryan. Ryan, Thanks for the e-mail. Just for the record, I got in touch with TTimo (very nice guy, btw) and he gave me the e-mail of an other ex-Loki employee who worked on the IRIX port. There was one IRIX port built by Richard Hess (the 1.17 IRIX build was an internal BETA at SGI) but the project was cancelled and the sources and binaries were destroyed as per the agreement with Loki. Refs: http://www.nekochan.net/weblog/archives/000069.html http://www.siliconbunny.com/mirrors/ftp.sgi.com/sgi/quake/intro.html http://groups.google.fr/group/comp.sys.sgi.graphics/msg/a7b271d24fe72822?dmode=source&hl=en So it looks like I have no source to try to recover and I will start from the current code base... -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From zakk at timedoctor.org Thu Sep 1 02:47:10 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 31 Aug 2005 23:47:10 -0700 Subject: [quake3] Sgi IRIX port.. In-Reply-To: References: <43168F79.9030105@clutteredmind.org> Message-ID: <20050901064710.GC19026@timedoctor.org> If it is any consolation we are glad to hear you are working on this. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From ludwig.nussel at suse.de Thu Sep 1 03:09:53 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Thu, 1 Sep 2005 09:09:53 +0200 Subject: [quake3] Make clean In-Reply-To: <20050901055602.GC12318@timedoctor.org> References: <20050901055602.GC12318@timedoctor.org> Message-ID: <20050901070953.GA24199@suse.de> Zachary J. Slater wrote: > Make clean doesn't seem to actually work right, if somebody has two > seconds that would be a good thing to fix. it cleans up code/unix/{release,debug}* fine here. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From zakk at timedoctor.org Thu Sep 1 03:16:30 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 1 Sep 2005 00:16:30 -0700 Subject: [quake3] Make clean In-Reply-To: <20050901070953.GA24199@suse.de> References: <20050901055602.GC12318@timedoctor.org> <20050901070953.GA24199@suse.de> Message-ID: <20050901071630.GD19026@timedoctor.org> Ah, it deletes the files, not the directories, strange. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Thu Sep 1 04:22:02 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 1 Sep 2005 01:22:02 -0700 Subject: sound Message-ID: <20050901082202.GE19026@timedoctor.org> Hey everyone, if you're having problems with SDL sound in quake3: http://icculus.org/lgfaq/#setthatdriver and http://icculus.org/lgfaq/#ioq3sndprb which should be up on the web in less than twenty minutes. Otherwise, I'm not sure, but there may be some lag in the audio, could just be my junky laptop, we'll see. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From jaq at spacepants.org Thu Sep 1 04:24:28 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Thu, 1 Sep 2005 18:24:28 +1000 Subject: [quake3] sound In-Reply-To: <20050901082202.GE19026@timedoctor.org> References: <20050901082202.GE19026@timedoctor.org> Message-ID: <20050901082428.GG11993@spacepants.org> This one time, at band camp, Zachary J. Slater wrote: >Hey everyone, if you're having problems with SDL sound in quake3: >http://icculus.org/lgfaq/#setthatdriver >and >http://icculus.org/lgfaq/#ioq3sndprb >which should be up on the web in less than twenty minutes. Firefox really doesn't like that, looks like the page claims to be html strict but there's a stray at the end of the

element. From icculus at clutteredmind.org Thu Sep 1 10:36:23 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Thu, 01 Sep 2005 10:36:23 -0400 Subject: [quake3] Make clean In-Reply-To: <20050901062502.GF11993@spacepants.org> References: <20050901055602.GC12318@timedoctor.org> <20050901055731.GE11993@spacepants.org> <20050901060817.GA19026@timedoctor.org> <20050901062502.GF11993@spacepants.org> Message-ID: <431711E7.5090904@clutteredmind.org> > I've been using cons, as per the README, got a clean build without needing > to hack the source like the Makefile did; and if cons is anything like SCons, > you shouldn't need to ever run clean to get a correct build :) We should probably pick one or the other and remove the one we don't use. The cons system isn't building the SDL stuff, btw, since it won't define USE_SDL=1 and doesn't have sdl_snd.c/sdl_glimp.c in the build. --ryan. From jaq at spacepants.org Thu Sep 1 11:03:21 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Fri, 2 Sep 2005 01:03:21 +1000 Subject: [quake3] Make clean In-Reply-To: <431711E7.5090904@clutteredmind.org> References: <20050901055602.GC12318@timedoctor.org> <20050901055731.GE11993@spacepants.org> <20050901060817.GA19026@timedoctor.org> <20050901062502.GF11993@spacepants.org> <431711E7.5090904@clutteredmind.org> Message-ID: <20050901150321.GI11993@spacepants.org> This one time, at band camp, Ryan C. Gordon wrote: > >>I've been using cons, as per the README, got a clean build without needing >>to hack the source like the Makefile did; and if cons is anything like >>SCons, >>you shouldn't need to ever run clean to get a correct build :) > >We should probably pick one or the other and remove the one we don't use. > >The cons system isn't building the SDL stuff, btw, since it won't define >USE_SDL=1 and doesn't have sdl_snd.c/sdl_glimp.c in the build. Should be pretty trivial to add sdl_snd.c, etc. to the build in the Conscripts, and add a USE_SDL option. From icculus at clutteredmind.org Thu Sep 1 12:19:48 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Thu, 01 Sep 2005 12:19:48 -0400 Subject: [quake3] Infostring thing In-Reply-To: <20050901053037.GA12318@timedoctor.org> References: <20050901053037.GA12318@timedoctor.org> Message-ID: <43172A24.7060304@clutteredmind.org> Zachary J. Slater wrote: > Could somebody test the fix for the infostring bug/feature/crack: > https://bugzilla.icculus.org/show_bug.cgi?id=2356 > I applied but can't test right now. This is a good fix; Luigi has harrassed me to patch like four Quake titles now. :) --ryan. From zakk at timedoctor.org Thu Sep 1 14:26:41 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 1 Sep 2005 11:26:41 -0700 Subject: [pomac@vapor.com: [q3] general C info.] Message-ID: <20050901182641.GD10292@timedoctor.org> I thought this would be more appropriate for the list, though obviously replies should go to Ian. ----- Forwarded message from Ian Kumlien ----- Subject: [q3] general C info. From: Ian Kumlien To: zakk at icculus.org Date: Thu, 01 Sep 2005 19:25:45 +0200 X-Mailer: Evolution 2.2.3 X-Spambayes-Classification: ham; 0.03 Hi, I'm working on a patch that enables you to compile the project with: -Wall -Wimplicit -Wstrict-prototypes -Werror To minimize the possible errors in new code... But, how do you on a variable initialization line send the values on? ie, Init ONE variable with the right values and return them to the others of the same type. I think i have seen it done but i can't remember... Also, i have been unable to find the definition of bot_moveresult_t... If you know where it is, or find it, lemme know. -- Ian Kumlien -- http://pomac.netswarm.net ----- End forwarded message ----- -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Thu Sep 1 14:28:56 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 1 Sep 2005 11:28:56 -0700 Subject: [quake3] sound In-Reply-To: <20050901082428.GG11993@spacepants.org> References: <20050901082202.GE19026@timedoctor.org> <20050901082428.GG11993@spacepants.org> Message-ID: <20050901182856.GE10292@timedoctor.org> Well, I fixed that after I saw your message, however, now the problem is redundant and gone. The setthatdriver entry is probably still valid for most people, though. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From floam at sh.nu Thu Sep 1 18:28:18 2005 From: floam at sh.nu (Aaron Gyes) Date: Thu, 01 Sep 2005 15:28:18 -0700 Subject: OpenAL stuff Message-ID: <1125613698.24328.2.camel@localhost> Anyone played with BlackAura's OpenAL code? A tarball with the new code is here: http://files.frashii.com/~sp00nz/Doom/files/BlackAura/q3_snd_al-7.tar.gz Apparantly he never got the makefile to work or something so he didn't patch the Makefile. Some third party guy made a Makefile from the icculus.org quake3 that supposedly builds this: http://www.gsklan.de/sonstiges/Makefile Obviously, it's a work in progress. Aaron Gyes From zakk at timedoctor.org Thu Sep 1 18:46:58 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 1 Sep 2005 15:46:58 -0700 Subject: [quake3] OpenAL stuff In-Reply-To: <1125613698.24328.2.camel@localhost> References: <1125613698.24328.2.camel@localhost> Message-ID: <20050901224657.GA12268@timedoctor.org> This isn't to go into SVN until blackaura himself says that it is done, and offers it to us. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From michel at daenzer.net Thu Sep 1 19:06:00 2005 From: michel at daenzer.net (Michel =?ISO-8859-1?Q?D=E4nzer?=) Date: Thu, 01 Sep 2005 19:06:00 -0400 Subject: [quake3] Anyone run Linux/PPC? In-Reply-To: <20050831032910.GB8526@timedoctor.org> References: <20050831032910.GB8526@timedoctor.org> Message-ID: <1125615960.23262.110.camel@localhost> On Tue, 2005-08-30 at 20:29 -0700, Zachary J. Slater wrote: > If anyone runs Linux/PPC and would like to test, it would > be very helpful, thanks. Basically works nicely. The VM interpreter has a couple of issues that my own hacked up tree doesn't have, I need to see if I can come up with a patch for those. The compiler would be much nicer anyway though. :) The bigger issue is that console input doesn't work, nothing happens when typing in the console. Other than that, keyboard input works fine though, including (un)hiding the console. So, kudos for having it working this well on PPC already, keep it up, and maybe I can contribute some bits here and there. -- Earthling Michel D?nzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer From floam at sh.nu Thu Sep 1 21:18:03 2005 From: floam at sh.nu (Aaron Gyes) Date: Thu, 01 Sep 2005 18:18:03 -0700 Subject: [quake3] OpenAL stuff In-Reply-To: <20050901224657.GA12268@timedoctor.org> References: <1125613698.24328.2.camel@localhost> <20050901224657.GA12268@timedoctor.org> Message-ID: <1125623883.6858.0.camel@localhost> On Thu, 2005-09-01 at 15:46 -0700, Zachary J. Slater wrote: > This isn't to go into SVN until blackaura himself says that it is done, > and offers it to us. I wasn't asking that it be merged - clearly it's a work in progress. I was asking if anyone subscribed here has gotten it to work themeselves. I havn't. From zakk at timedoctor.org Thu Sep 1 22:41:43 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 1 Sep 2005 19:41:43 -0700 Subject: [quake3] OpenAL stuff In-Reply-To: <1125623883.6858.0.camel@localhost> References: <1125613698.24328.2.camel@localhost> <20050901224657.GA12268@timedoctor.org> <1125623883.6858.0.camel@localhost> Message-ID: <20050902024143.GB12268@timedoctor.org> Ah, you'll have to forgive me, I am rather ill today and a number of people who aren't the guy who wrote this have approached me me regarding merging it. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From icculus at clutteredmind.org Fri Sep 2 00:35:54 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Fri, 02 Sep 2005 00:35:54 -0400 Subject: [quake3] Anyone run Linux/PPC? In-Reply-To: <1125615960.23262.110.camel@localhost> References: <20050831032910.GB8526@timedoctor.org> <1125615960.23262.110.camel@localhost> Message-ID: <4317D6AA.7010706@clutteredmind.org> > The bigger issue is that console input doesn't work, nothing happens > when typing in the console. Other than that, keyboard input works fine > though, including (un)hiding the console. This was broken by a buggy patch last night; if you haven't tried the latest in Subversion, please try again and let me know if this issue is fixed. --ryan. From michel at daenzer.net Fri Sep 2 00:45:14 2005 From: michel at daenzer.net (Michel =?ISO-8859-1?Q?D=E4nzer?=) Date: Fri, 02 Sep 2005 00:45:14 -0400 Subject: [quake3] Anyone run Linux/PPC? In-Reply-To: <4317D6AA.7010706@clutteredmind.org> References: <20050831032910.GB8526@timedoctor.org> <1125615960.23262.110.camel@localhost> <4317D6AA.7010706@clutteredmind.org> Message-ID: <1125636314.23262.141.camel@localhost> On Fri, 2005-09-02 at 00:35 -0400, Ryan C. Gordon wrote: > > The bigger issue is that console input doesn't work, nothing happens > > when typing in the console. Other than that, keyboard input works fine > > though, including (un)hiding the console. > > This was broken by a buggy patch last night; if you haven't tried the > latest in Subversion, please try again and let me know if this issue is > fixed. Works fine now, thanks. -- Earthling Michel D?nzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer From zakk at timedoctor.org Sat Sep 3 01:16:01 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 2 Sep 2005 22:16:01 -0700 Subject: Ok, sound? Message-ID: <20050903051601.GB18339@timedoctor.org> I was just playing on Primer's server (running on an x86_64 box is online at newcastle.devrandom.net:1717) and then on my own with some reaction quake 3 (rq3). I /think/ the SDL stuff, or maybe our version in general, adds some sound lag. What I really need right now is for other people to check for that. Small, otherwise insignificant changes like this, could ruin the player experience. If you have good hearing and time to check, please do, platform doesn't matter as long as you have sound working at all! Just try to think if the sound seems like it takes longer to come out of your headphones/speakers than it does with the id binaries. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From valfonso at teleline.es Sat Sep 3 03:43:19 2005 From: valfonso at teleline.es (Valeriano Alfonso Rodriguez) Date: Sat, 3 Sep 2005 07:43:19 +0000 Subject: [quake3] Ok, sound? In-Reply-To: <20050903051601.GB18339@timedoctor.org> References: <20050903051601.GB18339@timedoctor.org> Message-ID: <20050903074319.4e0fa70e@whitestar> On Fri, 2 Sep 2005 22:16:01 -0700 "Zachary J. Slater" wrote: > I was just playing on Primer's server (running on an x86_64 box is > online at newcastle.devrandom.net:1717) and then on my own with some > reaction quake 3 (rq3). I /think/ the SDL stuff, or maybe our version > in general, adds some sound lag. What I really need right now is for > other people to check for that. Small, otherwise insignificant > changes like this, could ruin the player experience. > I was noticing the same problem. One possible fix is to lower the "desired.samples" value, in SNDDMA_Init:code/unix/sdl_snd.c . This will make sdl_audio_callback:code/unix/sdl_snd.c being called more frequently. > If you have good hearing and time to check, please do, platform > doesn't matter as long as you have sound working at all! > Just try to think if the sound seems like it takes longer to come out > of your headphones/speakers than it does with the id binaries. I agree. From vincent at cojot.name Sat Sep 3 09:41:46 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sat, 3 Sep 2005 15:41:46 +0200 (CEST) Subject: 'Remove Useless setuid code' ??? In-Reply-To: <20050903074319.4e0fa70e@whitestar> References: <20050903051601.GB18339@timedoctor.org> <20050903074319.4e0fa70e@whitestar> Message-ID: Just saw a bunch of 'Remove Useless setuid code' in svn. Could we know why these were removed and also more generally what kind of approval/testing does a new patch get before it's accepted in svn? Can we at least see and discuss these patches first on the mailing list? Maybe this kind of 'getuid() == saved_euid)' isn't needed on Linux but it might be useful on other platforms. Can we please take care not to change too much the actual code that worked? If the goal of the project isn't to remain stable and extended the cross-platformability of the engine, Please say so, that's fine with me: I'll just fork a new project where I'll invest time in the non-Linux ports.. When I worked on i-o-q2, many of the patches that got into CVS were tested on Linux/x86 only and immediately found their way into CVS where I had to fix them afterwards while I always backported my Solaris fixes to the generic UNIX tree. In the recent years, I ported the TGE engine to Solaris/Sparc and some things were never fixed because it was/is/will mostly x86-only 'oriented'. Thanks for reading, Please don't flame. :) Vincent From vincent at cojot.name Sat Sep 3 09:42:26 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sat, 3 Sep 2005 15:42:26 +0200 (CEST) Subject: [quake3] Make clean In-Reply-To: <20050901071630.GD19026@timedoctor.org> References: <20050901055602.GC12318@timedoctor.org> <20050901070953.GA24199@suse.de> <20050901071630.GD19026@timedoctor.org> Message-ID: On Thu, 1 Sep 2005, Zachary J. Slater wrote: > Ah, it deletes the files, not the directories, strange. It's been the same at the iD since q1 and q2. Perhaps some historical stuff... :) -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From debeuk at gmail.com Sat Sep 3 09:53:55 2005 From: debeuk at gmail.com (Sven Beukenex) Date: Sat, 3 Sep 2005 15:53:55 +0200 Subject: [quake3] Ok, sound? In-Reply-To: <20050903051601.GB18339@timedoctor.org> References: <20050903051601.GB18339@timedoctor.org> Message-ID: <9a224f8b050903065348256672@mail.gmail.com> On 9/3/05, Zachary J. Slater wrote: > > I was just playing on Primer's server (running on an x86_64 box is > online at newcastle.devrandom.net:1717) > and then on my own with some > reaction quake 3 (rq3). I /think/ the SDL stuff, or maybe our version in > general, adds some sound lag. What I really need right now is for other > people to check for that. Small, otherwise insignificant changes like > this, could ruin the player experience. > > If you have good hearing and time to check, please do, platform doesn't > matter as long as you have sound working at all! > Just try to think if the sound seems like it takes longer to come out of > your headphones/speakers than it does with the id binaries. -- > -Zachary J. Slater > zakk at timedoctor.org > zacharyslater at gmail.com > I can confirm that, more audio lag then with the original id binaries. Oh and hello list ;) Sven -- Why are the pretty ones always insane? -- J.G. Thirlwell -------------- next part -------------- An HTML attachment was scrubbed... URL: From pomac at vapor.com Sat Sep 3 10:22:17 2005 From: pomac at vapor.com (Ian Kumlien) Date: Sat, 03 Sep 2005 16:22:17 +0200 Subject: [quake3] 'Remove Useless setuid code' ??? In-Reply-To: References: <20050903051601.GB18339@timedoctor.org> <20050903074319.4e0fa70e@whitestar> Message-ID: <1125757337.17577.123.camel@localhost> On Sat, 2005-09-03 at 15:41 +0200, vincent at cojot.name wrote: > Just saw a bunch of 'Remove Useless setuid code' in svn. > > Could we know why these were removed and also more generally what kind of > approval/testing does a new patch get before it's accepted in svn? Well my patch had a small oopsie in it... But thats fixed now =) > Can we at least see and discuss these patches first on the mailing list? I assume thats what this ml is for, i mainly did a cleanup thats why i sent to zakk directly... > Maybe this kind of 'getuid() == saved_euid)' isn't needed on Linux but it > might be useful on other platforms. Can we please take care not to change > too much the actual code that worked? If the goal of the project isn't to > remain stable and extended the cross-platformability of the engine, Please > say so, that's fine with me: I'll just fork a new project where I'll > invest time in the non-Linux ports.. I thought that the setuid was to change resolution in the old days or so? I'll check it if still works on my laptop (since it can't use sdl due to sdl actually deadlocking X on debian sid). I don't have access to anything beyond x86/amd64 that could run this so, i can't test. > When I worked on i-o-q2, many of the patches that got into CVS were tested > on Linux/x86 only and immediately found their way into CVS where I had to > fix them afterwards while I always backported my Solaris fixes to the > generic UNIX tree. In the recent years, I ported the TGE engine to > Solaris/Sparc and some things were never fixed because it was/is/will > mostly x86-only 'oriented'. Well, it should be portable imho, but with C you have to make a effort to make it work on f.ex. amd64 and ia32 and still break it on others (ppl should know about endian by now imho) As a side note, perhaps zakk should change the homepage to say that patches should be sent here? PS. the first section seems to use root access to be able to access the sound, so it could really be related to the things we now use groups to access (or atleast most unixes does this). DS. -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From pomac at vapor.com Sat Sep 3 11:44:59 2005 From: pomac at vapor.com (Ian Kumlien) Date: Sat, 03 Sep 2005 17:44:59 +0200 Subject: [quake3] 'Remove Useless setuid code' ??? In-Reply-To: References: <20050903051601.GB18339@timedoctor.org> <20050903074319.4e0fa70e@whitestar> Message-ID: <1125762299.17577.130.camel@localhost> On Sat, 2005-09-03 at 15:41 +0200, vincent at cojot.name wrote: > Just saw a bunch of 'Remove Useless setuid code' in svn. It seems like the setuid is infact useless and related to that root usually owns devices like that... I also seem to remember that id discuraged ppl from running it setuid.. (i'm tired though so it might be context leakage). Anyways, if we merged that, then why not -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: q3-setuid-cleanup.diff Type: text/x-patch Size: 1084 bytes Desc: URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From ludwig.nussel at suse.de Sat Sep 3 11:54:15 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sat, 3 Sep 2005 17:54:15 +0200 Subject: [quake3] 'Remove Useless setuid code' ??? In-Reply-To: <1125762299.17577.130.camel@localhost> References: <20050903051601.GB18339@timedoctor.org> <1125762299.17577.130.camel@localhost> Message-ID: <200509031754.15841.ludwig.nussel@suse.de> Am Samstag, 3. September 2005 17:44 schrieb Ian Kumlien: > On Sat, 2005-09-03 at 15:41 +0200, vincent at cojot.name wrote: > > Just saw a bunch of 'Remove Useless setuid code' in svn. > > It seems like the setuid is infact useless and related to that root > usually owns devices like that... I also seem to remember that id > discuraged ppl from running it setuid.. (i'm tired though so it might be > context leakage). It was not only useless but also dangerous. Q3 clearly is not designed to run setuid root plus the way it tried to drop the privileges was broken. > Anyways, if we merged that, then why not I'd even remove the bit that tries to load libgl from the current directory. One can just as well set LD_LIBRARY_PATH="." to achieve that. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From ludwig.nussel at suse.de Sat Sep 3 12:00:36 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sat, 3 Sep 2005 18:00:36 +0200 Subject: [quake3] Ok, sound? In-Reply-To: <20050903051601.GB18339@timedoctor.org> References: <20050903051601.GB18339@timedoctor.org> Message-ID: <200509031800.36841.ludwig.nussel@suse.de> Am Samstag, 3. September 2005 07:16 schrieb Zachary J. Slater: > I was just playing on Primer's server (running on an x86_64 box is > online at newcastle.devrandom.net:1717) and then on my own with some > reaction quake 3 (rq3). I /think/ the SDL stuff, or maybe our version in > general, adds some sound lag. What I really need right now is for other > people to check for that. Small, otherwise insignificant changes like > this, could ruin the player experience. Are you using the alsa backend and dmix? cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Sat Sep 3 12:19:04 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sat, 3 Sep 2005 18:19:04 +0200 (CEST) Subject: [quake3] 'Remove Useless setuid code' ??? In-Reply-To: <200509031754.15841.ludwig.nussel@suse.de> References: <20050903051601.GB18339@timedoctor.org> <1125762299.17577.130.camel@localhost> <200509031754.15841.ludwig.nussel@suse.de> Message-ID: On Sat, 3 Sep 2005, Ludwig Nussel wrote: >> It seems like the setuid is infact useless and related to that root >> usually owns devices like that... I also seem to remember that id >> discuraged ppl from running it setuid.. (i'm tired though so it might be >> context leakage). > > It was not only useless but also dangerous. Q3 clearly is not > designed to run setuid root plus the way it tried to drop the > privileges was broken. I Agree. Are we sure this code was only for Linux? I don't think it's needed on Solaris or IRIX anyway either. I'm just glad to see that we are having that discussion now.. :) > I'd even remove the bit that tries to load libgl from the current > directory. One can just as well set LD_LIBRARY_PATH="." to achieve > that. Agreed with that one too.. (most systems have a system-wide libGL these days) but maybe we could comment it out in case someone needs to re-activate it in the future. Thanks for not flaming, btw.. :) -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Sat Sep 3 12:24:06 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sat, 3 Sep 2005 18:24:06 +0200 (CEST) Subject: [quake3] 'Remove Useless setuid code' ??? In-Reply-To: <1125762299.17577.130.camel@localhost> References: <20050903051601.GB18339@timedoctor.org> <20050903074319.4e0fa70e@whitestar> <1125762299.17577.130.camel@localhost> Message-ID: On Sat, 3 Sep 2005, Ian Kumlien wrote: > On Sat, 2005-09-03 at 15:41 +0200, vincent at cojot.name wrote: >> Just saw a bunch of 'Remove Useless setuid code' in svn. > > It seems like the setuid is infact useless and related to that root > usually owns devices like that... I also seem to remember that id > discuraged ppl from running it setuid.. (i'm tired though so it might be > context leakage). > > Anyways, if we merged that, then why not In q2 we had this (amazing how some code looks identical...): [...snip...] if ( ( glw_state.OpenGLLib = dlopen( dllname, RTLD_LAZY ) ) == 0 ) { char fn[MAX_OSPATH]; char *path; // try basedir next path = ri.Cvar_Get ("basedir", ".", CVAR_NOSET)->string; snprintf (fn, MAX_OSPATH, "%s/%s", path, dllname ); if ( ( glw_state.OpenGLLib = dlopen( fn, RTLD_LAZY ) ) == 0 ) { ri.Con_Printf( PRINT_ALL, "%s\n", dlerror() ); return false; } Com_Printf ("Using %s for OpenGL...", fn); } else { Com_Printf ("Using %s for OpenGL...", dllname); } [...snip...] -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From pomac at vapor.com Sat Sep 3 12:31:48 2005 From: pomac at vapor.com (Ian Kumlien) Date: Sat, 03 Sep 2005 18:31:48 +0200 Subject: [quake3] 'Remove Useless setuid code' ??? In-Reply-To: <200509031754.15841.ludwig.nussel@suse.de> References: <20050903051601.GB18339@timedoctor.org> <1125762299.17577.130.camel@localhost> <200509031754.15841.ludwig.nussel@suse.de> Message-ID: <1125765109.17577.132.camel@localhost> On Sat, 2005-09-03 at 17:54 +0200, Ludwig Nussel wrote: > Am Samstag, 3. September 2005 17:44 schrieb Ian Kumlien: > > It seems like the setuid is infact useless and related to that root > > usually owns devices like that... I also seem to remember that id > > discuraged ppl from running it setuid.. (i'm tired though so it might be > > context leakage). > > It was not only useless but also dangerous. Q3 clearly is not > designed to run setuid root plus the way it tried to drop the > privileges was broken. Heh, ok, good, someone who ahs checked the code in all contexts, i just looked at the diff.. > > Anyways, if we merged that, then why not > > I'd even remove the bit that tries to load libgl from the current > directory. One can just as well set LD_LIBRARY_PATH="." to achieve > that. Hummm thats a fall back if ld.so.conf can't find it... But something like this? -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: q3-setuid-cleanup-2.diff Type: text/x-patch Size: 1386 bytes Desc: URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From zakk at timedoctor.org Sat Sep 3 13:40:43 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sat, 3 Sep 2005 10:40:43 -0700 Subject: [quake3] Ok, sound? In-Reply-To: <200509031800.36841.ludwig.nussel@suse.de> References: <20050903051601.GB18339@timedoctor.org> <200509031800.36841.ludwig.nussel@suse.de> Message-ID: <20050903174043.GA20806@timedoctor.org> No, on my laptop I use alsa (no dmix) and on my main machine I use oss. Perhaps because we're not getting the mmap() thing (or at least I would think) with SDL, the delay or something is going too long. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Sat Sep 3 13:57:09 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sat, 3 Sep 2005 10:57:09 -0700 Subject: a side note In-Reply-To: <1125757337.17577.123.camel@localhost> References: <20050903051601.GB18339@timedoctor.org> <20050903074319.4e0fa70e@whitestar> <1125757337.17577.123.camel@localhost> Message-ID: <20050903175709.GB20806@timedoctor.org> On Sat, Sep 03, 2005 at 04:22:17PM +0200, Ian Kumlien wrote: > As a side note, perhaps zakk should change the homepage to say that > patches should be sent here? > Done. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From vincent at cojot.name Sat Sep 3 16:00:29 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sat, 3 Sep 2005 22:00:29 +0200 (CEST) Subject: [quake3] Ok, sound? In-Reply-To: <20050903051601.GB18339@timedoctor.org> References: <20050903051601.GB18339@timedoctor.org> Message-ID: On Fri, 2 Sep 2005, Zachary J. Slater wrote: > I was just playing on Primer's server (running on an x86_64 box is > online at newcastle.devrandom.net:1717) and then on my own with some > reaction quake 3 (rq3). I /think/ the SDL stuff, or maybe our version in > general, adds some sound lag. What I really need right now is for other > people to check for that. Small, otherwise insignificant changes like > this, could ruin the player experience. > > If you have good hearing and time to check, please do, platform doesn't > matter as long as you have sound working at all! > Just try to think if the sound seems like it takes longer to come out of > your headphones/speakers than it does with the id binaries. Zakk, I am running the latest build on RHEL3 (RHAS, 32bit, i686, kernel 2.4). My hearing isn't too bad and I cannot seen any difference between the iD binaries and those I compiled myself. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From brendanburns at comcast.net Sat Sep 3 16:38:25 2005 From: brendanburns at comcast.net (brendanburns at comcast.net) Date: Sat, 03 Sep 2005 20:38:25 +0000 Subject: Crash on x86_64? Message-ID: <090320052038.12148.431A09C10009CA5200002F7422058844849C029D9A0D020E0B020A9D0D@comcast.net> Hey folks, Just got around to trying this with on my x86_64 Debian box, I'm using the demo pak0.pk3 since i don't own the game. I start quake with: ./linuxquake3 +set sv_pure 0 +set vm_cgame 0 +set vm_game 0 +set vm_ui 0 Everything launches fine, but when I try to start a single player game it dies. Running it in gdb, I see that it dies on line 840 of qcommon/cvar.c with a segfault. Based on the comments, this function seems to be vm specific. Should it not be called with vm_foo set to zero? I'll play around some more, but if someone on x86_64 has this working, let me know! Thanks in advance! --brendan From pomac at vapor.com Sat Sep 3 16:52:05 2005 From: pomac at vapor.com (Ian Kumlien) Date: Sat, 03 Sep 2005 22:52:05 +0200 Subject: [quake3] Crash on x86_64? In-Reply-To: <090320052038.12148.431A09C10009CA5200002F7422058844849C029D9A0D020E0B020A9D0D@comcast.net> References: <090320052038.12148.431A09C10009CA5200002F7422058844849C029D9A0D020E0B020A9D0D@comcast.net> Message-ID: <1125780725.17577.144.camel@localhost> On Sat, 2005-09-03 at 20:38 +0000, brendanburns at comcast.net wrote: > I'll play around some more, but if someone on x86_64 has this working, > let me know! Works just fine here, i'm on tier two and i have also been playing it as a server for several hours. -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From arny at ats.s.bawue.de Sat Sep 3 17:08:03 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Sat, 3 Sep 2005 23:08:03 +0200 Subject: [quake3] Ok, sound? In-Reply-To: <20050903051601.GB18339@timedoctor.org> References: <20050903051601.GB18339@timedoctor.org> Message-ID: <200509032308.05554.arny@ats.s.bawue.de> On Saturday 03 September 2005 07:16, Zachary J. Slater wrote: > I was just playing on Primer's server (running on an x86_64 box is > online at newcastle.devrandom.net:1717) and then on my own with some > reaction quake 3 (rq3). I /think/ the SDL stuff, or maybe our version in > general, adds some sound lag. What I really need right now is for other > people to check for that. Small, otherwise insignificant changes like > this, could ruin the player experience. I found that editing the .asoundrc in your home directory alleviates the problem. for example lines like: period_size 1024 buffer_size 4096 already smell like delay. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From brendanburns at comcast.net Sat Sep 3 17:12:32 2005 From: brendanburns at comcast.net (brendanburns at comcast.net) Date: Sat, 03 Sep 2005 21:12:32 +0000 Subject: [quake3] Crash on x86_64? Message-ID: <090320052112.15363.431A11C00002C57500003C0322070215739C029D9A0D020E0B020A9D0D@comcast.net> Thanks for the note: Questions: What distro are you on? What compiler are you using? Thanks! --brendan > On Sat, 2005-09-03 at 20:38 +0000, brendanburns at comcast.net wrote: > > I'll play around some more, but if someone on x86_64 has this working, > > let me know! > > Works just fine here, i'm on tier two and i have also been playing it as > a server for several hours. > > -- > Ian Kumlien -- http://pomac.netswarm.net > -------------- next part -------------- An embedded message was scrubbed... From: Ian Kumlien Subject: Re: [quake3] Crash on x86_64? Date: Sat, 3 Sep 2005 20:46:03 +0000 Size: 681 URL: From pomac at vapor.com Sat Sep 3 17:23:01 2005 From: pomac at vapor.com (Ian Kumlien) Date: Sat, 03 Sep 2005 23:23:01 +0200 Subject: [quake3] Crash on x86_64? In-Reply-To: <090320052112.15363.431A11C00002C57500003C0322070215739C029D9A0D020E0B020A9D0D@comcast.net> References: <090320052112.15363.431A11C00002C57500003C0322070215739C029D9A0D020E0B020A9D0D@comcast.net> Message-ID: <1125782582.17577.147.camel@localhost> On Sat, 2005-09-03 at 21:12 +0000, brendanburns at comcast.net wrote: > Thanks for the note: > Questions: > What distro are you on? Gentoo, i have also run it on debian sid ia32. > What compiler are you using? gcc 4.0.1 -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From zakk at timedoctor.org Sun Sep 4 01:09:27 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sat, 3 Sep 2005 22:09:27 -0700 Subject: Coders take note!1111111111 Message-ID: <20050904050927.GA13818@timedoctor.org> inline isn't legal in c, use __inline instead. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Sun Sep 4 02:17:06 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sat, 3 Sep 2005 23:17:06 -0700 Subject: Hoorj! Message-ID: <20050904061706.GB13818@timedoctor.org> Good news, windows compiles fine after six fixed some things, she is also our fourth person with svn access. Hooray! BONZAI!!!!!!!!!!!!!!!!!1111111111111 -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From arny at ats.s.bawue.de Sun Sep 4 02:38:14 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Sun, 4 Sep 2005 08:38:14 +0200 Subject: [quake3] Hoorj! In-Reply-To: <20050904061706.GB13818@timedoctor.org> References: <20050904061706.GB13818@timedoctor.org> Message-ID: <200509040838.16971.arny@ats.s.bawue.de> On Sunday 04 September 2005 08:17, Zachary J. Slater wrote: > Good news, windows compiles fine after six fixed some things, she is > also our fourth person with svn access. She? Hooray!!!!!!!!!!!!!!!!1111oneoneone *scnr* :P -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From icculus at clutteredmind.org Sun Sep 4 03:19:11 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Sun, 04 Sep 2005 03:19:11 -0400 Subject: [quake3] Hoorj! In-Reply-To: <200509040838.16971.arny@ats.s.bawue.de> References: <20050904061706.GB13818@timedoctor.org> <200509040838.16971.arny@ats.s.bawue.de> Message-ID: <431A9FEF.2030905@clutteredmind.org> > She? Yeah, they still let women on the internet. (btw, "inline" is valid in C++ and the C99 spec for regular C. __inline is not valid C in any spec...the underscores sort of unofficially alert you that it's an extension, even though gcc and VS.NET both support it. If you find a compiler that doesn't, a simple: #define __inline Will make that code compile with more or less the same results.) --ryan. From vincent at cojot.name Sun Sep 4 03:56:16 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sun, 4 Sep 2005 09:56:16 +0200 (CEST) Subject: libdvdcss: svn commit r184 (robux4) (fwd) Message-ID: Just a Q... Would it be a good idea to set up a list for people who want Subversion notifications? See the example below... I, for one, would very much like to receive these notifications... Vincent ---------- Forwarded message ---------- Date: Sun, 04 Sep 2005 09:36:01 +0200 From: Subversion daemon Reply-To: libdvdcss-devel at videolan.org To: libdvdcss-devel at videolan.org Subject: libdvdcss: svn commit r184 (robux4) r184 | robux4 | 2005-09-04 09:36:01 +0200 (Sun, 04 Sep 2005) | 1 line Changed paths: M /trunk/src/dvdcss/dvdcss.h * libdvdcss: allow building DLL exports -- This is the libdvdcss-devel mailing-list, see http://developers.videolan.org/ To unsubscribe, go to: http://developers.videolan.org/lists.html From jaq at spacepants.org Sun Sep 4 04:03:46 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Sun, 4 Sep 2005 18:03:46 +1000 Subject: [quake3] libdvdcss: svn commit r184 (robux4) (fwd) In-Reply-To: References: Message-ID: <20050904080346.GT11993@spacepants.org> This one time, at band camp, vincent at cojot.name wrote: > >Just a Q... Would it be a good idea to set up a list for people who want >Subversion notifications? See the example below... I, for one, would very >much like to receive these notifications... Great idea, as long as zakk is a bit more verbose about what he's fixening :) From zakk at timedoctor.org Sun Sep 4 04:12:58 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sun, 4 Sep 2005 01:12:58 -0700 Subject: [quake3] libdvdcss: svn commit r184 (robux4) (fwd) In-Reply-To: References: Message-ID: <20050904081258.GB17517@timedoctor.org> Ryan has to set it up, and I have no problem with him doing it as long as he has time. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From admin at www0.org Sun Sep 4 05:15:37 2005 From: admin at www0.org (fs) Date: Sun, 4 Sep 2005 12:15:37 +0300 Subject: gpl anticheater Message-ID: <20050904091537.GA12017@www0.org> I was wondering if any one/group is developing a gpl client-server anticheat system for q3 or/and other games. I thought your group may now. k, that's all, thanks. From debeuk at gmail.com Sun Sep 4 10:16:34 2005 From: debeuk at gmail.com (Sven Beukenex) Date: Sun, 4 Sep 2005 16:16:34 +0200 Subject: [quake3] Crash on x86_64? In-Reply-To: <090320052038.12148.431A09C10009CA5200002F7422058844849C029D9A0D020E0B020A9D0D@comcast.net> References: <090320052038.12148.431A09C10009CA5200002F7422058844849C029D9A0D020E0B020A9D0D@comcast.net> Message-ID: <9a224f8b050904071625209c7f@mail.gmail.com> On 9/3/05, brendanburns at comcast.net wrote: > > Hey folks, > Just got around to trying this with on my x86_64 Debian box, I'm using the > demo pak0.pk3 since i don't own the game. > > I start quake with: > > ./linuxquake3 +set sv_pure 0 +set vm_cgame 0 +set vm_game 0 +set vm_ui 0 > > Everything launches fine, but when I try to start a single player game it > dies. Running it in gdb, > I see that it dies on line 840 of qcommon/cvar.c with a segfault. Based on > the comments, this function seems to be vm specific. Should it not be called > with vm_foo set to zero? > > I'll play around some more, but if someone on x86_64 has this working, let > me know! > > Thanks in advance! > --brendan > > Can you try (re)moving q3config.cfg? I experienced crashes in Team Arena, single player mode. Moving the configfiles fixed it for me (didn't think of making a diff to see what the problem was, sorry) Sven -- Why are the pretty ones always insane? -- J.G. Thirlwell -------------- next part -------------- An HTML attachment was scrubbed... URL: From icculus at clutteredmind.org Sun Sep 4 10:55:08 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Sun, 04 Sep 2005 10:55:08 -0400 Subject: [quake3] libdvdcss: svn commit r184 (robux4) (fwd) In-Reply-To: <20050904081258.GB17517@timedoctor.org> References: <20050904081258.GB17517@timedoctor.org> Message-ID: <431B0ACC.3000701@clutteredmind.org> > Ryan has to set it up, and I have no problem with him doing it as long > as he has time. It is done; send a blank email to quake3-commits-subscribe at icculus.org and you'll get an email every time someone makes a commit to the svn repository. This is a seperate mailing list from this one, to reduce chatter for those that don't care. The email's subject header will contain "[quake3-commits]" if you want to filter it into a subfolder. Discussion should remain here. --ryan. From tim at ngus.net Sun Sep 4 11:25:40 2005 From: tim at ngus.net (Tim Angus) Date: Sun, 4 Sep 2005 16:25:40 +0100 Subject: Coders take note!1111111111 In-Reply-To: <20050904050927.GA13818@timedoctor.org> References: <20050904050927.GA13818@timedoctor.org> Message-ID: <20050904162540.7e555b54.tim@ngus.net> On Sat, 3 Sep 2005 22:09:27 -0700 Zachary wrote: > inline isn't legal in c, use __inline instead. *pedantic mode engaged* It's legal in C99. ;) From tim at ngus.net Sun Sep 4 11:26:27 2005 From: tim at ngus.net (Tim Angus) Date: Sun, 4 Sep 2005 16:26:27 +0100 Subject: Hoorj! In-Reply-To: <431A9FEF.2030905@clutteredmind.org> References: <20050904061706.GB13818@timedoctor.org> <200509040838.16971.arny@ats.s.bawue.de> <431A9FEF.2030905@clutteredmind.org> Message-ID: <20050904162627.27d6d725.tim@ngus.net> On Sun, 04 Sep 2005 03:19:11 -0400 Ryan wrote: > (btw, "inline" is valid in C++ and the C99 spec for regular C. > __inline is not valid C in any spec...the underscores sort of > unofficially alert you that it's an extension, even though gcc and > VS.NET both support it. If you find a compiler that doesn't, a simple: Meh, some smart arse beat me too it. :( From brendanburns at comcast.net Sun Sep 4 12:09:04 2005 From: brendanburns at comcast.net (brendanburns at comcast.net) Date: Sun, 04 Sep 2005 16:09:04 +0000 Subject: [quake3] Crash on x86_64? Message-ID: <090420051609.10033.431B1C20000119E70000273122058844849C029D9A0D020E0B020A9D0D@comcast.net> Hey, I found part of the problem... if VM_DllSysCall is set to the new version of the code (as it is by default because __i386__ is not defined, then that code crashes in CVar_Register.) If I use the old code (e.g. not placing the arguments into memory) then the crash doesn't occur. Which is the appropriate thing for x86_64 to do? With that change, now when I try to run either single or multi-player when the game starts, I just get a black screen that goes entirely red occasionally as I wiggle the mouse.... Very weird. --brendan -------------- next part -------------- An embedded message was scrubbed... From: Sven Beukenex Subject: Re: [quake3] Crash on x86_64? Date: Sun, 4 Sep 2005 14:16:49 +0000 Size: 3857 URL: From ludwig.nussel at suse.de Sun Sep 4 13:04:54 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sun, 4 Sep 2005 19:04:54 +0200 Subject: [quake3] Hoorj! In-Reply-To: <20050904061706.GB13818@timedoctor.org> References: <20050904061706.GB13818@timedoctor.org> Message-ID: <20050904170454.GA19223@suse.de> Zachary J. Slater wrote: > Good news, windows compiles fine after six fixed some things, she is > also our fourth person with svn access. Can we agree to write at least some words about what got changed into the svn log, even though it would be obvious when looking at the diff? A link to bugzilla only also makes it hard to track changes later. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From ludwig.nussel at suse.de Sun Sep 4 13:06:06 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sun, 4 Sep 2005 19:06:06 +0200 Subject: [quake3] Hoorj! In-Reply-To: <431A9FEF.2030905@clutteredmind.org> References: <20050904061706.GB13818@timedoctor.org> <200509040838.16971.arny@ats.s.bawue.de> <431A9FEF.2030905@clutteredmind.org> Message-ID: <20050904170606.GA19246@suse.de> Ryan C. Gordon wrote: > (btw, "inline" is valid in C++ and the C99 spec for regular C. __inline > is not valid C in any spec...the underscores sort of unofficially alert > you that it's an extension, even though gcc and VS.NET both support it. > If you find a compiler that doesn't, a simple: > > #define __inline > > Will make that code compile with more or less the same results.) There already is a macro ID_INLINE in q_shared.h. It expands to __inline for VC, static for lcc and inline for gcc. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Sun Sep 4 13:36:18 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sun, 4 Sep 2005 19:36:18 +0200 (CEST) Subject: [quake3] libdvdcss: svn commit r184 (robux4) (fwd) In-Reply-To: <431B0ACC.3000701@clutteredmind.org> References: <20050904081258.GB17517@timedoctor.org> <431B0ACC.3000701@clutteredmind.org> Message-ID: On Sun, 4 Sep 2005, Ryan C. Gordon wrote: > It is done; send a blank email to quake3-commits-subscribe at icculus.org > and you'll get an email every time someone makes a commit to the svn > repository. This is a seperate mailing list from this one, to reduce > chatter for those that don't care. Hi Ryan, It's quite perfect. Many thanks for this.. Vincent From zakk at timedoctor.org Sun Sep 4 14:22:24 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sun, 4 Sep 2005 11:22:24 -0700 Subject: [quake3] Re: Coders take note!1111111111 In-Reply-To: <20050904162540.7e555b54.tim@ngus.net> References: <20050904050927.GA13818@timedoctor.org> <20050904162540.7e555b54.tim@ngus.net> Message-ID: <20050904182224.GC17517@timedoctor.org> On Sun, Sep 04, 2005 at 04:25:40PM +0100, Tim Angus wrote: > *pedantic mode engaged* It's legal in C99. ;) Doesn't matter, I have to compile the windows version with a compiler that gives a damn. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Sun Sep 4 14:24:45 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sun, 4 Sep 2005 11:24:45 -0700 Subject: [quake3] Hoorj! In-Reply-To: <20050904170454.GA19223@suse.de> References: <20050904061706.GB13818@timedoctor.org> <20050904170454.GA19223@suse.de> Message-ID: <20050904182445.GD17517@timedoctor.org> I thought all non-trivial commits (as in, not me editing the readme or fixing something quick in the makefile) had some words. I understand your point about the bugzilla ones though, sure, I'll be more verbose. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Sun Sep 4 15:46:45 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sun, 4 Sep 2005 12:46:45 -0700 Subject: [quake3] gpl anticheater In-Reply-To: <20050904091537.GA12017@www0.org> References: <20050904091537.GA12017@www0.org> Message-ID: <20050904194644.GF17517@timedoctor.org> I personally don't think any GPL punkbuster style program would be effective, though I would certainly welcome people to try. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Sun Sep 4 16:07:47 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sun, 4 Sep 2005 13:07:47 -0700 Subject: Sound lag fix Message-ID: <20050904200747.GG17517@timedoctor.org> Ryan just ci'd a fix for the sound lag, let us know if you still hear any. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From pomac at vapor.com Sun Sep 4 20:01:56 2005 From: pomac at vapor.com (Ian Kumlien) Date: Mon, 05 Sep 2005 02:01:56 +0200 Subject: [quake3] Crash on x86_64? In-Reply-To: <090420051609.10033.431B1C20000119E70000273122058844849C029D9A0D020E0B020A9D0D@comcast.net> References: <090420051609.10033.431B1C20000119E70000273122058844849C029D9A0D020E0B020A9D0D@comcast.net> Message-ID: <1125878517.26040.5.camel@localhost> On Sun, 2005-09-04 at 16:09 +0000, brendanburns at comcast.net wrote: > Hey, > I found part of the problem... if VM_DllSysCall is set to the new > version of the code (as it is by default because __i386__ is not > defined, then that code crashes in CVar_Register.) If I use the > old code (e.g. not placing the arguments into memory) then the crash > doesn't occur. Hummm, you mean the arguments? I put mine in the config file, and i have had no such problems... > With that change, now when I try to run either single or multi-player > when the game starts, I just get a black screen that goes entirely red > occasionally as I wiggle the mouse.... Could it be demo vs real thing? > Very weird. Yep, since i last played some today... Not much since i have a fever and that kinda makes you sluggish... -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From pomac at vapor.com Sun Sep 4 20:05:50 2005 From: pomac at vapor.com (Ian Kumlien) Date: Mon, 05 Sep 2005 02:05:50 +0200 Subject: [quake3] gpl anticheater In-Reply-To: <20050904194644.GF17517@timedoctor.org> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> Message-ID: <1125878750.26040.13.camel@localhost> On Sun, 2005-09-04 at 12:46 -0700, Zachary J. Slater wrote: > I personally don't think any GPL punkbuster style program would be > effective, though I would certainly welcome people to try. I totally agree, since all binaries would change size and md5 etc... It's really unlucky, but i doubt that there is a way around it.. =P F.ex. UT99 relied on server forcing md5 checks and sedning data to the client. Bunkbuster relies on even more magic but anything opensource can easily be hacked unless we make some way of compiling in sourcefile md5's that can't be changed etc =) (Just ignore me, i have a fever =)) -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From arny at ats.s.bawue.de Sun Sep 4 19:59:49 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Mon, 5 Sep 2005 01:59:49 +0200 Subject: [quake3] gpl anticheater In-Reply-To: <1125878750.26040.13.camel@localhost> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> <1125878750.26040.13.camel@localhost> Message-ID: <200509050159.54394.arny@ats.s.bawue.de> On Monday 05 September 2005 02:05, Ian Kumlien wrote: > On Sun, 2005-09-04 at 12:46 -0700, Zachary J. Slater wrote: > F.ex. UT99 relied on server forcing md5 checks and sedning data to the > client. Bunkbuster relies on even more magic but anything opensource can > easily be hacked unless we make some way of compiling in sourcefile > md5's that can't be changed etc =) > (Just ignore me, i have a fever =)) How can any code in the world you write prevent, that a cheater changes the source code so that the quake3 engine would send faked hashes, md5 etc...? -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pomac at vapor.com Sun Sep 4 20:45:28 2005 From: pomac at vapor.com (Ian Kumlien) Date: Mon, 05 Sep 2005 02:45:28 +0200 Subject: [quake3] gpl anticheater In-Reply-To: <200509050159.54394.arny@ats.s.bawue.de> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> <1125878750.26040.13.camel@localhost> <200509050159.54394.arny@ats.s.bawue.de> Message-ID: <1125881128.26040.15.camel@localhost> On Mon, 2005-09-05 at 01:59 +0200, Thilo Schulz wrote: > On Monday 05 September 2005 02:05, Ian Kumlien wrote: > > On Sun, 2005-09-04 at 12:46 -0700, Zachary J. Slater wrote: > > F.ex. UT99 relied on server forcing md5 checks and sedning data to the > > client. Bunkbuster relies on even more magic but anything opensource can > > easily be hacked unless we make some way of compiling in sourcefile > > md5's that can't be changed etc =) > > (Just ignore me, i have a fever =)) > > How can any code in the world you write prevent, that a cheater changes the > source code so that the quake3 engine would send faked hashes, md5 etc...? As i said, i have a fever, but that would be one requirement, etc... I really think it is impossible though... -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From tim at ngus.net Sun Sep 4 20:49:24 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 5 Sep 2005 01:49:24 +0100 Subject: gpl anticheater In-Reply-To: <20050904194644.GF17517@timedoctor.org> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> Message-ID: <20050905014924.40591e2e.tim@ngus.net> On Sun, 4 Sep 2005 12:46:45 -0700 Zachary wrote: > I personally don't think any GPL punkbuster style program would be > effective, though I would certainly welcome people to try. It's unfortunate, but yes. At a fundamental level the client needs to be trusted to some degree. Any anti-cheat mechanism therefore needs to pretty much rely on obscurity as its primary means of operation. I don't really see how this is possible with an open source program. The double whammy is that since Q3 is now GPL, it cannot be distributed with any closed source libraries. So implementing a free-as-in-beer alternative to punkbuster is hampered slightly. Correct me if I'm wrong (IANAL etc.), but there is no problem with a GPLed application linking against a non GPL library? At least, not if you dynamically load the library at runtime as Q3 does with libGL. Right? From jaq at spacepants.org Sun Sep 4 21:09:47 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Mon, 5 Sep 2005 11:09:47 +1000 Subject: viewsvn Message-ID: <20050905010947.GV11993@spacepants.org> Is there a viewsvn at icculus.org? From jaq at spacepants.org Sun Sep 4 21:14:04 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Mon, 5 Sep 2005 11:14:04 +1000 Subject: [quake3] viewsvn In-Reply-To: <20050905010947.GV11993@spacepants.org> References: <20050905010947.GV11993@spacepants.org> Message-ID: <20050905011404.GA8460@spacepants.org> This one time, at band camp, Jamie Wilkinson wrote: >Is there a viewsvn at icculus.org? > yeah, forget it, I can't read... From tfogal at cisunix.unh.edu Sun Sep 4 21:34:32 2005 From: tfogal at cisunix.unh.edu (tom fogal) Date: Sun, 04 Sep 2005 21:34:32 -0400 Subject: [quake3] Re: gpl anticheater In-Reply-To: Your message of "Mon, 05 Sep 2005 01:49:24 BST." <20050905014924.40591e2e.tim@ngus.net> Message-ID: <200509050134.j851YWEK001577@euler.unh.edu> <20050905014924.40591e2e.tim at ngus.net>Tim Angus writes: > On Sun, 4 Sep 2005 12:46:45 -0700 Zachary wrote: > > I personally don't think any GPL punkbuster style program would be > > effective, though I would certainly welcome people to try. > > The double whammy is that since Q3 is now GPL, it cannot be distributed > with any closed source libraries. So implementing a free-as-in-beer > alternative to punkbuster is hampered slightly. Correct me if I'm wrong > (IANAL etc.), but there is no problem with a GPLed application linking > against a non GPL library? I am fairly (but not 100% sure) that this is illegal as per the GPL, if done at link time... thats why the LGPL was developed. See this quote from the bottom of the GPL page on gnu.org[1]: This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Lesser General Public License instead of this License. I take this to mean that a non-GPL program cannot link in a GPL library, and to imply that the converse is true. Of course, this text isn't actually part of the license, just a little hint at the bottom. > At least, not if you dynamically load the > library at runtime as Q3 does with libGL. Right? Dynamic loading really obscures the issue... I'm not sure... -tom [1] http://www.gnu.org/copyleft/gpl.html From tim at ngus.net Sun Sep 4 22:01:04 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 5 Sep 2005 03:01:04 +0100 Subject: gpl anticheater In-Reply-To: <200509050134.j851YWEK001577@euler.unh.edu> References: <20050905014924.40591e2e.tim@ngus.net> <200509050134.j851YWEK001577@euler.unh.edu> Message-ID: <20050905030104.5d265f0a.tim@ngus.net> On Sun, 04 Sep 2005 21:34:32 -0400 tom wrote: > > At least, not if you dynamically load the > > library at runtime as Q3 does with libGL. Right? > > Dynamic loading really obscures the issue... I'm not sure... I guess the other obscurity with libGL in particular is that the legal status of the installed libGL is not even KNOWN to a GPLed program, so if it is the case that linking against it (in whatever fashion) is not legal, then the onus is really on the end user to decide what is right and wrong since they choose which libGL they use. The GPL is great in spirit, but sometimes it can be a real pain in the ass. From admin at www0.org Mon Sep 5 02:06:59 2005 From: admin at www0.org (fs) Date: Mon, 5 Sep 2005 09:06:59 +0300 Subject: [quake3] Re: gpl anticheater In-Reply-To: <20050905014924.40591e2e.tim@ngus.net> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> <20050905014924.40591e2e.tim@ngus.net> Message-ID: <20050905060659.GA30933@www0.org> On Mon, Sep 05, 2005 at 01:49:24AM +0100, Tim Angus wrote: > On Sun, 4 Sep 2005 12:46:45 -0700 Zachary wrote: > > I personally don't think any GPL punkbuster style program would be > > effective, though I would certainly welcome people to try. > > It's unfortunate, but yes. At a fundamental level the client needs to be > trusted to some degree. Any anti-cheat mechanism therefore needs to > pretty much rely on obscurity as its primary means of operation. I don't > really see how this is possible with an open source program. > > The double whammy is that since Q3 is now GPL, it cannot be distributed > with any closed source libraries. So implementing a free-as-in-beer > alternative to punkbuster is hampered slightly. Correct me if I'm wrong > (IANAL etc.), but there is no problem with a GPLed application linking > against a non GPL library? At least, not if you dynamically load the > library at runtime as Q3 does with libGL. Right? I was having that idea even before q3 was gpl'ed and I'm less of a programmer and more of a gamer so bare with me :] I was thinking about it as a team project that would work with certified builds either of its own or co-operative gpl projects that provide those "certified" builts. That would probably take out the problem of having cheats on the q3 binaries themselves. As for the anticheater itself, I said I'm not a programmer but I was hoping if there was a way to do it. If it's documented to be absolutely impossible how about certified builds of the anticheater itself? something like giving the source but data parts of it to be of the anticheater team. Or a double licence or a special gpl-like licence. And what about evenbalance's punkbuster itself, is it anhackable? And what about suse/redhat, etc. linux, they provide gpl software with non-free libraries. Sorry if I've become tiring to some of you but I'd like to see those ideas that come from the gaming community (it's not just me that thought about it) to be dismissed only if they are really impossible. thanks for listening. From admin at www0.org Mon Sep 5 02:34:20 2005 From: admin at www0.org (fs) Date: Mon, 5 Sep 2005 09:34:20 +0300 Subject: [quake3] Re: gpl anticheater In-Reply-To: <200509050134.j851YWEK001577@euler.unh.edu> References: <20050905014924.40591e2e.tim@ngus.net> <200509050134.j851YWEK001577@euler.unh.edu> Message-ID: <20050905063420.GA31277@www0.org> On Sun, Sep 04, 2005 at 09:34:32PM -0400, tom fogal wrote: > This General Public License does not permit incorporating your > program into proprietary programs. If your program is a subroutine > library, you may consider it more useful to permit linking > proprietary applications with the library. If this is what you want > to do, use the GNU Lesser General Public License instead of this > License. > > I take this to mean that a non-GPL program cannot link in a GPL > library, and to imply that the converse is true. Of course, this text > isn't actually part of the license, just a little hint at the bottom. That part says a gpl program can't be incorporated into programs of proprietary licences, not exactly that case. And I think a this point, a legal mailing list would be better. I'll check if FSF offers mailing to help in such matters. by the way, just in case that's what you meant, I find it of little importance to have such a program in a way that ships with the q3 sources. having the gamers get the anticheater afterwars seems a technicality to me that only spends some minor time to the gamers (exactly like pb does..). From admin at www0.org Mon Sep 5 02:52:17 2005 From: admin at www0.org (fs) Date: Mon, 5 Sep 2005 09:52:17 +0300 Subject: [quake3] Re: gpl anticheater In-Reply-To: <200509050134.j851YWEK001577@euler.unh.edu> References: <20050905014924.40591e2e.tim@ngus.net> <200509050134.j851YWEK001577@euler.unh.edu> Message-ID: <20050905065217.GA31332@www0.org> On Sun, Sep 04, 2005 at 09:34:32PM -0400, tom fogal wrote: > > The double whammy is that since Q3 is now GPL, it cannot be distributed > > with any closed source libraries. So implementing a free-as-in-beer > > alternative to punkbuster is hampered slightly. Correct me if I'm wrong > > (IANAL etc.), but there is no problem with a GPLed application linking > > against a non GPL library? > > I am fairly (but not 100% sure) that this is illegal as per the GPL, if > done at link time... thats why the LGPL was developed. k, found a relevant answer about it on http://www.fsf.org/licensing/licenses/gpl-faq.html#GPLIncompatibleLibs the link is pretty good about anything gpl related actually. ==== What legal issues come up if I use GPL-incompatible libraries with GPL software? If the libraries that you link with fall within the following exception in the GPL: However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. then you don't have to do anything special to use them; the requirement to distribute source code for the whole program does not include those libraries, even if you distribute a linked executable containing them. Thus, if the libraries you need come with major parts of a proprietary operating system, the GPL says people can link your program with them without any conditions. If you want your program to link against a library not covered by that exception, you need to add your own exception, wholly outside of the GPL. This copyright notice and license notice give permission to link with the program FOO: Copyright (C) yyyy This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA In addition, as a special exception, gives permission to link the code of this program with the FOO library (or with modified versions of FOO that use the same license as FOO), and distribute linked combinations including the two. You must obey the GNU General Public License in all respects for all of the code used other than FOO. If you modify this file, you may extend this exception to your version of the file, but you are not obligated to do so. If you do not wish to do so, delete this exception statement from your version. Only the copyright holders for the program can legally authorize this exception. If you wrote the whole program yourself, then assuming your employer or school does not claim the copyright, you are the copyright holder--so you can authorize the exception. But if you want to use parts of other GPL-covered programs by other authors in your code, you cannot authorize the exception for them. You have to get the approval of the copyright holders of those programs. When other people modify the program, they do not have to make the same exception for their code--it is their choice whether to do so. If the libraries you intend to link with are non-free, please also see the section on writing Free Software which uses non-free libraries. From admin at www0.org Mon Sep 5 03:10:13 2005 From: admin at www0.org (fs) Date: Mon, 5 Sep 2005 10:10:13 +0300 Subject: [quake3] Re: gpl anticheater In-Reply-To: <20050905065217.GA31332@www0.org> References: <20050905014924.40591e2e.tim@ngus.net> <200509050134.j851YWEK001577@euler.unh.edu> <20050905065217.GA31332@www0.org> Message-ID: <20050905071013.GA31410@www0.org> On Mon, Sep 05, 2005 at 09:52:17AM +0300, fs wrote: > k, found a relevant answer about it on > http://www.fsf.org/licensing/licenses/gpl-faq.html#GPLIncompatibleLibs > > the link is pretty good about anything gpl related actually. k, last message from me on this post, sorry for the spammage =] It seems the exact answer from the faq is not exactly relevant and sorry about it, but still a usefull faq listing when "IANAL" strikes. From jaq at spacepants.org Mon Sep 5 02:57:58 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Mon, 5 Sep 2005 16:57:58 +1000 Subject: [quake3] Re: gpl anticheater In-Reply-To: <20050905060659.GA30933@www0.org> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> <20050905014924.40591e2e.tim@ngus.net> <20050905060659.GA30933@www0.org> Message-ID: <20050905065758.GH8460@spacepants.org> This one time, at band camp, fs wrote: >Sorry if I've become tiring to some of you but I'd like to see those >ideas that come from the gaming community (it's not just me that >thought about it) to be dismissed only if they are really impossible. I was going to stay quiet, but I wanted to stop you before you went too far down the boring licensing path. At a fundamental level, you're talking about DRM for games. Cory Doctorow wrote a short piece about why open source DRM can't work here: http://www.boingboing.net/2005/08/24/drm_ssl.html There's some fundamental computational theory that prevents you from solving this in the general sense anyway -- enough that you can get away with it in a closed source environment but the moment you open up your protocols and algorithms you can't trust anyone. The only way to do it safely is to do it all on the server -- and then you're starting to get complicated with things like behavioural analysis. I'd much rather spend that time with people I trust on a private server :) From admin at www0.org Mon Sep 5 03:33:59 2005 From: admin at www0.org (fs) Date: Mon, 5 Sep 2005 10:33:59 +0300 Subject: [quake3] Re: gpl anticheater In-Reply-To: <20050905065758.GH8460@spacepants.org> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> <20050905014924.40591e2e.tim@ngus.net> <20050905060659.GA30933@www0.org> <20050905065758.GH8460@spacepants.org> Message-ID: <20050905073359.GA31504@www0.org> On Mon, Sep 05, 2005 at 04:57:58PM +1000, Jamie Wilkinson wrote: > This one time, at band camp, fs wrote: > >Sorry if I've become tiring to some of you but I'd like to see those > >ideas that come from the gaming community (it's not just me that > >thought about it) to be dismissed only if they are really impossible. > > I was going to stay quiet, but I wanted to stop you before you went too far > down the boring licensing path. > > At a fundamental level, you're talking about DRM for games. Cory Doctorow > wrote a short piece about why open source DRM can't work here: > http://www.boingboing.net/2005/08/24/drm_ssl.html > > There's some fundamental computational theory that prevents you from > solving this in the general sense anyway -- enough that you can get away > with it in a closed source environment but the moment you open up your > protocols and algorithms you can't trust anyone. The only way to do it > safely is to do it all on the server -- and then you're starting to get > complicated with things like behavioural analysis. I'd much rather spend > that time with people I trust on a private server :) Thanks for the informative reply Jamie, much appreciated. From jaq at spacepants.org Mon Sep 5 03:33:09 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Mon, 5 Sep 2005 17:33:09 +1000 Subject: build system Message-ID: <20050905073309.GI8460@spacepants.org> I'm pretty sure that the make-based build is the deprecated build; as such I've tried to merge in the changes that ludwig did early on for x86_64 into the Conscript; I havne't checked all the commits on makefiles yet, but with this patch it should get rolling: makes the top-level makefile call cons instead for the release and debug build, and fixes up some cflags. Also works around a bug on Ubuntu where /lib/libc.so.6 isn't executable for the config stage, which ttimo used to work out the library version linked against. My intended goal with this is to port all the makefile changes to cons, and standardise on cons (with the exception of lcc for now) leaving only a courtesy makefile at the top for people who expect to type make in the tree root. Secondly to work out whether cons supports Visual Studio project targets as SCons does, and if not, port the build system to SCons, which has the following two advantages over cons: it is actively maintained, and it's not perl :) What do you guys think, does that sound like a worthwhile goal for icculus? -------------- next part -------------- Index: code/Construct =================================================================== --- code/Construct (revision 84) +++ code/Construct (working copy) @@ -63,7 +63,11 @@ else { # libc .. do the little magic! + if ( -x '/lib/libc.so.6' ) { $libc_cmd = '/lib/libc.so.6 |grep "GNU C "|grep version|awk -F "version " \'{ print $2 }\'|cut -b -3'; + } else { + $libc_cmd = 'nm -D /lib/libc.so.6 | grep GLIBC_ | grep -v PRIVATE | tail -n 1 | cut -f2 -d_'; + } $libc = `$libc_cmd`; chop ($libc); } @@ -227,17 +231,17 @@ # build the config directory $CONFIG_DIR = $config . '-' . $cpu . '-' . $OS . '-' . $libc; -$COMMON_CFLAGS = '-pipe '; +$COMMON_CFLAGS = '-pipe -Wall '; # XXX: -Werror is good for the soul if ($config eq 'debug') { # use -Werror for better QA - $BASE_CFLAGS = $COMMON_CFLAGS . '-g -Wall -O0 '; + $BASE_CFLAGS = $COMMON_CFLAGS . '-g -O0 '; $BSPC_BASE_CFLAGS = $COMMON_CFLAGS . '-g -O0 -DLINUX -DBSPC -Dstricmp=strcasecmp '; } else { - $BASE_CFLAGS = $COMMON_CFLAGS . '-DNDEBUG -O3 -march=i686 -fomit-frame-pointer -ffast-math -malign-loops=2 -malign-jumps=2 -malign-functions=2 -fno-strict-aliasing -fstrength-reduce '; + $BASE_CFLAGS = $COMMON_CFLAGS . '-DNDEBUG -O3 -march=i686 -fomit-frame-pointer -ffast-math -falign-loops=2 -falign-jumps=2 -falign-functions=2 -fno-strict-aliasing -fstrength-reduce '; $BSPC_BASE_CFLAGS = $BASE_CFLAGS . '-DLINUX -DBSPC -Dstricmp=strcasecmp '; } Index: Makefile =================================================================== --- Makefile (revision 84) +++ Makefile (working copy) @@ -1,8 +1,11 @@ VERSION=1.33_SVN$(shell LANG=C svnversion .) -release debug clean: - $(MAKE) -C code/unix $@ +release debug: + cd code && unix/cons -- $@ +clean: + $(MAKE) -C code/unix clean + dist: rm -rf quake3-$(VERSION) svn export . quake3-$(VERSION) From ludwig.nussel at suse.de Mon Sep 5 04:12:28 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Mon, 5 Sep 2005 10:12:28 +0200 Subject: [quake3] build system In-Reply-To: <20050905073309.GI8460@spacepants.org> References: <20050905073309.GI8460@spacepants.org> Message-ID: <20050905081228.GA32719@suse.de> Jamie Wilkinson wrote: > [make -> cons...] > What do you guys think, does that sound like a worthwhile goal for icculus? As someone who also does packaging work for a distribution I can only say that I hate custom build scripts. That's usually even worse than custom Makefiles. I'd rather go for automake. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From jaq at spacepants.org Mon Sep 5 04:21:32 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Mon, 5 Sep 2005 18:21:32 +1000 Subject: [quake3] build system In-Reply-To: <20050905081228.GA32719@suse.de> References: <20050905073309.GI8460@spacepants.org> <20050905081228.GA32719@suse.de> Message-ID: <20050905082132.GJ8460@spacepants.org> This one time, at band camp, Ludwig Nussel wrote: >Jamie Wilkinson wrote: >> [make -> cons...] >> What do you guys think, does that sound like a worthwhile goal for icculus? > >As someone who also does packaging work for a distribution I can >only say that I hate custom build scripts. That's usually even worse >than custom Makefiles. I'd rather go for automake. Well, my experience with quakeforge's quake2 and automake was everything but fun. Sure, it provides some very useful make targets for distributions, but developing with it is like choking on a bucket of dicks. I'm convinced that quake3 was released using cons, the Makefiles are merely leftovers. I suppose I should contact ttimo to confirm this, but the fact that the makefiles didn't work out of the box and cons did is in my mind pretty heavy evidence of the One True Build System here. I don't think the packaging would be much of an issue. It's pretty easy to tweak SCons to do the right thing, so I expect making the install targets in the Conscripts would be easy too. There's a significant amount of work already done in the SCons community for doing automake-like distribution targets (prefixed install, tarballs, etc). From mandin.patrice at wanadoo.fr Mon Sep 5 06:03:14 2005 From: mandin.patrice at wanadoo.fr (Patrice Mandin) Date: Mon, 5 Sep 2005 12:03:14 +0200 Subject: [quake3] build system In-Reply-To: <20050905073309.GI8460@spacepants.org> References: <20050905073309.GI8460@spacepants.org> Message-ID: <20050905120314.00e3ca40.mandin.patrice@wanadoo.fr> Le Mon, 5 Sep 2005 17:33:09 +1000 Jamie Wilkinson a ?crit: > What do you guys think, does that sound like a worthwhile goal for > icculus? Hum, I don't know much about scons, but is it possible to cross-compile quake3 using it? autotools are good for that, so I hope scons can do it also. -- Patrice Mandin WWW: http://perso.wanadoo.fr/patrice.mandin/ Programmeur Linux, Atari Sp?cialit?: D?veloppement, jeux From arny at ats.s.bawue.de Mon Sep 5 06:13:16 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Mon, 5 Sep 2005 12:13:16 +0200 Subject: Adding EliteForce compatibility Message-ID: <200509051213.22386.arny@ats.s.bawue.de> Hello, I am a great fan of the EliteForce game using the quake3 engine. The game is pretty old (the binaries are from something like 2001 - but there are active players still!) and the program is split into two .exe binaries, one for multiplayer, one for single player. Unfortunately, support for it of course ceased, and there never was a Linux client version. I have started trying to alter the q3 source code, so that the eliteforce vms can be loaded (the source code for them is available) - I successfully compiled them using gcc to .so files on my Linux system already and can do full debugging on them. The User Interface mostly works (with a few glitches still), see http://thilo.kickchat.com/ef-quake3.png. I was able to load a map after doing a few dirty hacks for a quick Proof of Concept. I think, Raven has not performed big changes in the engine and I'd like to try adapting the codebase. My question now is: would you be willing to accept my patches for these modifications, if I make the changes completely seperable by using #ifdefs and stuff like that? I really like the idea of the icculus codebase - and would also like to see IPv6 support in future versions of EF. What needs to be done? - I must add non-existing functions to the API that the eliteforce vms expect - I must add mp3 support, as many sounds are only available in mp3 format. - I must add a few EF specific shader options - nothing big I think, though. - I must add support for raven's way of handling models. This will probably be the toughest task for me, and the point most likely for me to fail. I haven't had a good look at their model formats yet. Why adapt the quake3 source, and not the vm code of EF itself? Well, there are two reasons: Much stuff, like mp3 format, shaders, player models are not quite what quake3 expects. Either support for that has to be added, or all the EF specific stuff must be converted, which would probably be much more work. Secondly, if I want to retain compatibility to existing servers and unchanged windows clients, I guess I'll have to make sure that all the original data is being used for the game. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From tfogal at cisunix.unh.edu Mon Sep 5 09:23:25 2005 From: tfogal at cisunix.unh.edu (tom fogal) Date: Mon, 05 Sep 2005 09:23:25 -0400 Subject: [quake3] build system In-Reply-To: Your message of "Mon, 05 Sep 2005 18:21:32 +1000." <20050905082132.GJ8460@spacepants.org> Message-ID: <200509051323.j85DNPFV006978@euler.unh.edu> <20050905082132.GJ8460 at spacepants.org>Jamie Wilkinson writes: > This one time, at band camp, Ludwig Nussel wrote: > >Jamie Wilkinson wrote: > >> [make -> cons...] > >> What do you guys think, does that sound like a worthwhile goal for icculus > ? > > > >As someone who also does packaging work for a distribution I can > >only say that I hate custom build scripts. That's usually even worse > >than custom Makefiles. I'd rather go for automake. > > I'm convinced that quake3 was released using cons, the Makefiles are merely > leftovers. I suppose I should contact ttimo to confirm this, but the fact > that the makefiles didn't work out of the box and cons did is in my mind > pretty heavy evidence of the One True Build System here. I've had a conversation with TTimo on IRC, pretty much along the same lines as this thread. TTimo intended cons to be the build system. In any case, just because it wasn't the intended use doesn't mean it's wrong. This \emph{is} open source, after all =) > I don't think the packaging would be much of an issue. It's pretty easy to > tweak SCons to do the right thing, so I expect making the install targets in > the Conscripts would be easy too. There's a significant amount of work > already done in the SCons community for doing automake-like distribution > targets (prefixed install, tarballs, etc). Personally I'm very much in favor of the GNU autotools (since I've recently learned them and they don't feel like magic anymore). They provide some nice features (easy cross compilation was mentioned) and IMHO are 'the right way' to go about building software -- checking for certain supported features of the build environment and adapting based on those checks (instead of hardcoded, "Apple does things this way, so #ifdef __APPLE__..."). I know of at least one person that has already started on q3 autotools support [1]. Anyway I do feel the current Makefile system is fine for now. People are building with it, its working for most (all?) everyone, and every UNIX developer does/should know how to maintain a Makefile. If we do decide to go the cons route though, I think porting to scons is a good idea. I don't know anything about the technical merits of either project, but scons being actively maintained is key in my mind. An unsupported build system is just another opportunity for code rot... My $0.02 -tom [1] http://g-rave.nl/quake3.html From 421621 at ingta.unizar.es Mon Sep 5 11:16:45 2005 From: 421621 at ingta.unizar.es (Tei) Date: Mon, 05 Sep 2005 17:16:45 +0200 Subject: [quake3] Re: gpl anticheater In-Reply-To: <20050905065758.GH8460@spacepants.org> References: <20050904091537.GA12017@www0.org> <20050904194644.GF17517@timedoctor.org> <20050905014924.40591e2e.tim@ngus.net> <20050905060659.GA30933@www0.org> <20050905065758.GH8460@spacepants.org> Message-ID: <1125933405.3728.37.camel@vore> El lun, 05-09-2005 a las 16:57 +1000, Jamie Wilkinson escribi?: .. > There's some fundamental computational theory that prevents you from > solving this in the general sense anyway -- enough that you can get away > with it in a closed source environment but the moment you open up your > protocols and algorithms you can't trust anyone. The only way to do it > safely is to do it all on the server -- and then you're starting to get > complicated with things like behavioural analysis. I'd much rather spend > that time with people I trust on a private server :) He guys! Its Ok to load dinamically a optional library not GPL to a GPL engine. Because that its USE. And the GPL its about distribution. Theres no use restriction on the GPL. You only need to care about to distribute your GPL engine separate from the closed source safety module. And make these module optional. This how some quakeworld GPL engines work (fuhquake, etc..). And will work with Quake3. --Tei From icculus at clutteredmind.org Mon Sep 5 11:55:54 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Mon, 05 Sep 2005 11:55:54 -0400 Subject: [quake3] build system In-Reply-To: <200509051323.j85DNPFV006978@euler.unh.edu> References: <200509051323.j85DNPFV006978@euler.unh.edu> Message-ID: <431C6A8A.6000601@clutteredmind.org> > Personally I'm very much in favor of the GNU autotools (since I've > recently learned them and they don't feel like magic anymore). The problem is that someone always decides they are suddenly competent at autotools, converts a project, and vanihses. Then the rest of us are helpless when "configure/make/make install" stops working. The Makefile is FINE, don't make this harder than it needs to be. --ryan. From icculus at clutteredmind.org Mon Sep 5 12:22:32 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Mon, 05 Sep 2005 12:22:32 -0400 Subject: [quake3] Sound lag fix In-Reply-To: <20050904200747.GG17517@timedoctor.org> References: <20050904200747.GG17517@timedoctor.org> Message-ID: <431C70C8.6060609@clutteredmind.org> Zachary J. Slater wrote: > Ryan just ci'd a fix for the sound lag, let us know if you still hear > any. Anyone try this yet? Does it make things better or worse? --ryan. From debeuk at gmail.com Mon Sep 5 14:26:19 2005 From: debeuk at gmail.com (Sven Beukenex) Date: Mon, 5 Sep 2005 20:26:19 +0200 Subject: [quake3] Sound lag fix In-Reply-To: <431C70C8.6060609@clutteredmind.org> References: <20050904200747.GG17517@timedoctor.org> <431C70C8.6060609@clutteredmind.org> Message-ID: <9a224f8b050905112687ca647@mail.gmail.com> On 9/5/05, Ryan C. Gordon wrote: > > Zachary J. Slater wrote: > > Ryan just ci'd a fix for the sound lag, let us know if you still hear > > any. > > Anyone try this yet? Does it make things better or worse? > > --ryan. > > > No noticable lag anymore, definitely an improvement. Thanks for the great work Ryan. Sven -- Why are the pretty ones always insane? -- J.G. Thirlwell -------------- next part -------------- An HTML attachment was scrubbed... URL: From zakk at timedoctor.org Mon Sep 5 14:52:15 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 5 Sep 2005 11:52:15 -0700 Subject: [quake3] build system In-Reply-To: <20050905073309.GI8460@spacepants.org> References: <20050905073309.GI8460@spacepants.org> Message-ID: <20050905185215.GH17517@timedoctor.org> We'll be sticking with make and the Makefile as the main build system. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From davebytes at comcast.net Mon Sep 5 23:04:16 2005 From: davebytes at comcast.net (David Chait) Date: Mon, 5 Sep 2005 23:04:16 -0400 Subject: Hey all -- who here is an X config god? ;) References: <20050905073309.GI8460@spacepants.org> <20050905185215.GH17517@timedoctor.org> Message-ID: <002001c5b28f$ab0bec80$6e01a8c0@sixfour> Figure one of you knows linux X setup WAY better than I do: I had a box running SDL/GL within the last year, but haven't tried GL at all recently. Inbetween, I've switched monitors twice (and reconfigured X for different resolution support), and only used X for small things, no gaming. Most work has been via remote ssh into the box. ;) I'm trying to get Q3A base running to try porting it to some new platforms/cpus. I've just installed the latest nv drivers (have a GeFX5200), ran Xconfigurator, went back and manually adjusted to use "glx" and "nvidia" in the respective places, have even rebooted, but I'm getting: Initializing OpenGL display ...setting mode 3: 640 480 SDL_SetVideoMode failed: Couldn't find matching GLX visual ... and does that last message a dozen times or so. I ran some app that verified the GLX config (i.e., it showed me the NV 2.0 driver, etc.). I'm running X/KDE in 1280x1024, native res of my Dell 1704FP. Is there a commandline arg to start in a given res (I've scoured the internet, didn't find a thing, except lots of 'dead sites')? Not sure it'd help, but I thought trying in 1280x1024 to match the current X mode might help. Thanks all! Probably something stupid (again, it WAS working once upon a time!). Regards, - David Chait From tfogal at cisunix.unh.edu Mon Sep 5 23:37:49 2005 From: tfogal at cisunix.unh.edu (tom fogal) Date: Mon, 05 Sep 2005 23:37:49 -0400 Subject: [quake3] Hey all -- who here is an X config god? ;) In-Reply-To: Your message of "Mon, 05 Sep 2005 23:04:16 EDT." <002001c5b28f$ab0bec80$6e01a8c0@sixfour> Message-ID: <200509060337.j863bn0O004209@euler.unh.edu> <002001c5b28f$ab0bec80$6e01a8c0 at sixfour>"David Chait" writes: > I'm trying to get Q3A base running to try porting it to some new > platforms/cpus. I've just installed the latest nv drivers (have a > GeFX5200), ran Xconfigurator, went back and manually adjusted to use "glx" > and "nvidia" in the respective places, have even rebooted, but I'm getting: > > Initializing OpenGL display > ...setting mode 3: 640 480 > SDL_SetVideoMode failed: Couldn't find matching GLX visual > > ... and does that last message a dozen times or so. > > I ran some app that verified the GLX config (i.e., it showed me the NV 2.0 > driver, etc.). I'm running X/KDE in 1280x1024, native res of my Dell > 1704FP. What does 'glxinfo' give? In particular, does "direct rendering: " say "Yes"? > Is there a commandline arg to start in a given res (I've scoured the > internet, didn't find a thing, except lots of 'dead sites')? Not sure it'd > help, but I thought trying in 1280x1024 to match the current X mode might > help. Not sure, but you can try making sure 640x480 is supported. In /etc/X11/xorg.conf or /etc/X11/XFree86.conf look, for a 'Section "Screen"'. You want to make sure your "Modes" line contains 640x480, in addition to what you normally like to run X at. For example, here's mine: Section "Screen" Identifier "screen-one" Device "agp-one" Monitor "monitor" DefaultDepth 24 SubSection "Display" Modes "1280x1024" "1024x768" "800x600" "640x480" Virtual 0 0 Depth 24 EndSubSection EndSection It might take some grepping to figure out which "Screen" is the active one, and make sure you add the required Modes to the appropriate depth (based on the DefaultDepth and Depth), but you get the idea. HTH, -tom From davebytes at comcast.net Tue Sep 6 00:50:11 2005 From: davebytes at comcast.net (David Chait) Date: Tue, 6 Sep 2005 00:50:11 -0400 Subject: [quake3] Hey all -- who here is an X config god? ;) References: <200509060337.j863bn0O004209@euler.unh.edu> Message-ID: <003401c5b29e$767bf820$6e01a8c0@sixfour> Hey Tom... thanks for the pointers. I had 640x480 already, so definitely weird. glxinfo looks good on this run (think it was the same last time). however, I just did remove my 16-bit display modes from the Screen and now defaulting to 24 (to match yours), renamed the monitor and device (still should have been fine before..), and added the Virtual setting you had. Replicated the changes to both my XF86 config files (I can't figure out why there are two...). THEN, I also just switched to have Gnome as my default desktop. Lo and behold, NOW I get past the SDL error. Of course, past that I get a "Sys_Error: recursive error after: User Interface is version 3, expected 6". I'm using the Q3A demo pak0.pk3 file (assumed that'd work since I saw it mentioned somewhere regarding running the Q3A code), placed in baseq3. Anyone know if the User Interface version issue is because of the pakfile, and if so what I can do about it? I'll go pick up a copy of Q3A somewhere if I have to -- not an issue, just wanted to get up and running quickly with demo content if possible! Thanks all! -d ----- Original Message ----- >> I'm trying to get Q3A base running to try porting it to some new >> platforms/cpus. I've just installed the latest nv drivers (have a >> GeFX5200), ran Xconfigurator, went back and manually adjusted to use >> "glx" >> and "nvidia" in the respective places, have even rebooted, but I'm >> getting: >> >> Initializing OpenGL display >> ...setting mode 3: 640 480 >> SDL_SetVideoMode failed: Couldn't find matching GLX visual > > What does 'glxinfo' give? In particular, does "direct rendering: " say > "Yes"? > >> Is there a commandline arg to start in a given res (I've scoured the >> internet, didn't find a thing, except lots of 'dead sites')? Not sure >> it'd >> help, but I thought trying in 1280x1024 to match the current X mode might >> help. > > Not sure, but you can try making sure 640x480 is supported. > > In /etc/X11/xorg.conf or /etc/X11/XFree86.conf > > look, for a 'Section "Screen"'. You want to make sure your "Modes" > line contains 640x480, in addition to what you normally like to run X > at. For example, here's mine: > > Section "Screen" > Identifier "screen-one" > Device "agp-one" > Monitor "monitor" > DefaultDepth 24 > SubSection "Display" > Modes "1280x1024" "1024x768" "800x600" "640x480" > Virtual 0 0 > Depth 24 > EndSubSection > EndSection From icculus at clutteredmind.org Tue Sep 6 02:34:45 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Tue, 06 Sep 2005 02:34:45 -0400 Subject: [quake3] Hey all -- who here is an X config god? ;) In-Reply-To: <003401c5b29e$767bf820$6e01a8c0@sixfour> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> Message-ID: <431D3885.40108@clutteredmind.org> > glxinfo looks good on this run (think it was the same last time). however, > I just did remove my 16-bit display modes from the Screen and now defaulting > to 24 (to match yours), renamed the monitor and device (still should have > been fine before..), and added the Virtual setting you had. Replicated the > changes to both my XF86 config files (I can't figure out why there are > two...). THEN, I also just switched to have Gnome as my default desktop. It's possible my new SDL code (or SDL itself) is broken in cases where it has to try various visuals to find one that works. If you are really really bored, go back to your original setup where the game dies and try it with the original binaries from id...that'll tell us if it's something we should fix. > Lo and behold, NOW I get past the SDL error. Of course, past that I get a > "Sys_Error: recursive error after: User Interface is version 3, expected 6". > I'm using the Q3A demo pak0.pk3 file (assumed that'd work since I saw it > mentioned somewhere regarding running the Q3A code), placed in baseq3. > Anyone know if the User Interface version issue is because of the pakfile, > and if so what I can do about it? I'll go pick up a copy of Q3A somewhere > if I have to -- not an issue, just wanted to get up and running quickly with > demo content if possible! I had this problem, too. If anyone's bored, can you see if this is fixable for demo content? --ryan. From ludwig.nussel at suse.de Tue Sep 6 03:41:24 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Tue, 6 Sep 2005 09:41:24 +0200 Subject: [quake3] demoq3 (Was: Hey all -- who here is an X config god? ;)) In-Reply-To: <431D3885.40108@clutteredmind.org> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431D3885.40108@clutteredmind.org> Message-ID: <20050906074124.GA15063@suse.de> Ryan C. Gordon wrote: > > Lo and behold, NOW I get past the SDL error. Of course, past that I get a > > "Sys_Error: recursive error after: User Interface is version 3, expected 6". > > I'm using the Q3A demo pak0.pk3 file (assumed that'd work since I saw it > > mentioned somewhere regarding running the Q3A code), placed in baseq3. > > Anyone know if the User Interface version issue is because of the pakfile, > > and if so what I can do about it? I'll go pick up a copy of Q3A somewhere > > if I have to -- not an issue, just wanted to get up and running quickly with > > demo content if possible! > > I had this problem, too. If anyone's bored, can you see if this is > fixable for demo content? RTFM :-) Just use the shared libraries that get compiled together with the new main binary. I doubt it's worth the effort to support the old vm. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From jaq at spacepants.org Tue Sep 6 05:47:54 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Tue, 6 Sep 2005 19:47:54 +1000 Subject: -Werror patch Message-ID: <20050906094754.GP8460@spacepants.org> If you add -Werror to the CFLAGS, only one showstopper appears with GCC 3.3 on Ubuntu. This patch changes the format arg to an int from a longint, given that the argument is passed in as an int and the other operand is cast to int *, this seems reasonable. -------------- next part -------------- Index: code/qcommon/vm.c =================================================================== --- code/qcommon/vm.c (revision 84) +++ code/qcommon/vm.c (working copy) @@ -852,7 +852,7 @@ f = fopen("syscalls.log", "w" ); } callnum++; - fprintf(f, "%i: %li (%i) = %i %i %i %i\n", callnum, args - (int *)currentVM->dataBase, + fprintf(f, "%i: %i (%i) = %i %i %i %i\n", callnum, args - (int *)currentVM->dataBase, args[0], args[1], args[2], args[3], args[4] ); } From pomac at vapor.com Tue Sep 6 06:40:45 2005 From: pomac at vapor.com (Ian Kumlien) Date: Tue, 06 Sep 2005 12:40:45 +0200 Subject: [quake3] -Werror patch In-Reply-To: <20050906094754.GP8460@spacepants.org> References: <20050906094754.GP8460@spacepants.org> Message-ID: <1126003245.26040.27.camel@localhost> On Tue, 2005-09-06 at 19:47 +1000, Jamie Wilkinson wrote: > If you add -Werror to the CFLAGS, only one showstopper appears with GCC 3.3 > on Ubuntu. > > This patch changes the format arg to an int from a longint, given that the > argument is passed in as an int and the other operand is cast to int *, this > seems reasonable. Sorry but it's not, since now you broke it for all 64 bit machines... It's also kinda strange, since a long is the same size as a int on a 32 bit machine. Anyways, Here is a fix. -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: q3-32vs64bit-fix.diff Type: text/x-patch Size: 1171 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From jaq at spacepants.org Tue Sep 6 06:34:02 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Tue, 6 Sep 2005 20:34:02 +1000 Subject: [quake3] -Werror patch In-Reply-To: <1126003245.26040.27.camel@localhost> References: <20050906094754.GP8460@spacepants.org> <1126003245.26040.27.camel@localhost> Message-ID: <20050906103402.GQ8460@spacepants.org> This one time, at band camp, Ian Kumlien wrote: >On Tue, 2005-09-06 at 19:47 +1000, Jamie Wilkinson wrote: >> If you add -Werror to the CFLAGS, only one showstopper appears with GCC 3.3 >> on Ubuntu. >> >> This patch changes the format arg to an int from a longint, given that the >> argument is passed in as an int and the other operand is cast to int *, this >> seems reasonable. > >Sorry but it's not, since now you broke it for all 64 bit machines... >It's also kinda strange, since a long is the same size as a int on a 32 >bit machine. Anyways, Here is a fix. Ah, shit. Sorry. Thanks for the fix. I was wondering about the pointer math, and thinking maybe %p should have been used instead. From brendanburns at comcast.net Tue Sep 6 09:01:36 2005 From: brendanburns at comcast.net (Brendan Burns) Date: Tue, 06 Sep 2005 09:01:36 -0400 Subject: [quake3] demoq3 (Was: Hey all -- who here is an X config god? ;)) In-Reply-To: <20050906074124.GA15063@suse.de> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431D3885.40108@clutteredmind.org> <20050906074124.GA15063@suse.de> Message-ID: <1126011696.3185.2.camel@localhost.localdomain> Hey folks, I had similar problems with the demo data... See earlier posts. I got the "User Interface is ..." error. I tracked down the line of code (sorry don't have it on me right now...) and commented out the line. Seems to work, but when I start to play my display is messed up as mentioned before... I haven't tried messing with display resolutions. That might fix things, I'll try that tonight.... --brendan On Tue, 2005-09-06 at 09:41 +0200, Ludwig Nussel wrote: > Ryan C. Gordon wrote: > > > Lo and behold, NOW I get past the SDL error. Of course, past that I get a > > > "Sys_Error: recursive error after: User Interface is version 3, expected 6". > > > I'm using the Q3A demo pak0.pk3 file (assumed that'd work since I saw it > > > mentioned somewhere regarding running the Q3A code), placed in baseq3. > > > Anyone know if the User Interface version issue is because of the pakfile, > > > and if so what I can do about it? I'll go pick up a copy of Q3A somewhere > > > if I have to -- not an issue, just wanted to get up and running quickly with > > > demo content if possible! > > > > I had this problem, too. If anyone's bored, can you see if this is > > fixable for demo content? > > RTFM :-) Just use the shared libraries that get compiled together > with the new main binary. I doubt it's worth the effort to support > the old vm. > > cu > Ludwig > From davebytes at comcast.net Tue Sep 6 10:31:54 2005 From: davebytes at comcast.net (David Chait) Date: Tue, 6 Sep 2005 10:31:54 -0400 Subject: [quake3] demoq3 (Was: Hey all -- who here is an X config god? ;)) References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431D3885.40108@clutteredmind.org> <20050906074124.GA15063@suse.de> Message-ID: <001b01c5b2ef$ba6bc500$6e01a8c0@sixfour> First, I did RTFM. Saw a page discussing the source release and successful use of the demo pakfile. :-p ;-) Second, I don't know where this makes an impact but I'm targeting >unsupported< CPUs. So while I'm currently trying to run on x86, an asm-vm binary won't do me any good (and I have only vm_interpreted.c and vm_none.c in my minorly-tweaked makefile). -d ----- Original Message ----- Ludwig Nussel wrote: > Ryan C. Gordon wrote: >> > Lo and behold, NOW I get past the SDL error. Of course, past that I >> > get a >> > "Sys_Error: recursive error after: User Interface is version 3, >> > expected 6". >> > I'm using the Q3A demo pak0.pk3 file (assumed that'd work since I saw >> > it >> > mentioned somewhere regarding running the Q3A code), placed in baseq3. >> > Anyone know if the User Interface version issue is because of the >> > pakfile, >> > and if so what I can do about it? I'll go pick up a copy of Q3A >> > somewhere >> > if I have to -- not an issue, just wanted to get up and running quickly >> > with >> > demo content if possible! >> >> I had this problem, too. If anyone's bored, can you see if this is >> fixable for demo content? > > RTFM :-) Just use the shared libraries that get compiled together > with the new main binary. I doubt it's worth the effort to support > the old vm. > From admin at www0.org Tue Sep 6 11:24:53 2005 From: admin at www0.org (fs) Date: Tue, 6 Sep 2005 18:24:53 +0300 Subject: [quake3] Hey all -- who here is an X config god? ;) In-Reply-To: <431D3885.40108@clutteredmind.org> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431D3885.40108@clutteredmind.org> Message-ID: <20050906152453.GA4198@www0.org> On Tue, Sep 06, 2005 at 02:34:45AM -0400, Ryan C. Gordon wrote: > It's possible my new SDL code (or SDL itself) is broken in cases where > it has to try various visuals to find one that works. If you are really > really bored, go back to your original setup where the game dies and try > it with the original binaries from id...that'll tell us if it's > something we should fix. got to say then, (no various tests, just a single case here) with the retail's pak0.pk3, on 16 bit color depth, debian unstable current, xorg, nvidia latest, nv 5700le), icculus' q3 would stop at SDL initialization unless run at 24 bit color depth. (color depth affects performance a lot on 1600x1200). ps. btw, I got those sound skips some ppl noticed too but they were more apparent when cpu load was at its highest (e.g. looking at the mirror of q3dm0) From mail at cynapses.org Tue Sep 6 12:01:14 2005 From: mail at cynapses.org (Andreas Schneider) Date: Tue, 06 Sep 2005 18:01:14 +0200 Subject: [quake3] Sound lag fix In-Reply-To: <431C70C8.6060609@clutteredmind.org> References: <20050904200747.GG17517@timedoctor.org> <431C70C8.6060609@clutteredmind.org> Message-ID: <431DBD4A.5070706@cynapses.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 For me, alsa is broken now... -- a Ryan C. Gordon wrote: > Zachary J. Slater wrote: > >>Ryan just ci'd a fix for the sound lag, let us know if you still hear >>any. > > > Anyone try this yet? Does it make things better or worse? > > --ryan. > > - -- http://www.cynapses.org/ - cybernetic synapses -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDHb1KwDudlFFJUr4RAjviAKCIZFsgI2nFQgCqsWXR/EO4+qszkACeNpBm RsiTN3IZaSRvHPnRmOKEfW0= =HDKo -----END PGP SIGNATURE----- From icculus at clutteredmind.org Tue Sep 6 16:11:18 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Tue, 06 Sep 2005 16:11:18 -0400 Subject: [quake3] Sound lag fix In-Reply-To: <431DBD4A.5070706@cynapses.org> References: <20050904200747.GG17517@timedoctor.org> <431C70C8.6060609@clutteredmind.org> <431DBD4A.5070706@cynapses.org> Message-ID: <431DF7E6.7020306@clutteredmind.org> Andreas Schneider wrote: > For me, alsa is broken now... Run the game with "+set sdlmixsamps 32768 +set sdldevsamps 1024" sdlmixsamps can be as big as you want, it just takes more memory; that's how much space you have for the engine to mix ahead. Bigger is better, but the default is probably fine. sdldevsamps is how much SDL wants to feed the audio device at a time. Bigger numbers result in more audio latency, smaller number risk skips. The defaults were halved in latest patch, which is probably why people hear skipping now. If you have a really fast machine that can keep ahead of the audio device, you can get away with 256 samples at 44100Khz (works here)...512 to 1024 is probably more reasonable for everyone else, though. Lower frequencies (the game defaults to 22050) can work with less samples at a time. Also, please make sure you're using an SDL with ALSA if you want ALSA. If you get a console message that says "SDL audio driver is "(UNKNOWN)"", then your SDL is misconfigured. --ryan. From mail at cynapses.org Tue Sep 6 17:07:58 2005 From: mail at cynapses.org (Andreas Schneider) Date: Tue, 06 Sep 2005 23:07:58 +0200 Subject: [quake3] Sound lag fix In-Reply-To: <431DF7E6.7020306@clutteredmind.org> References: <20050904200747.GG17517@timedoctor.org> <431C70C8.6060609@clutteredmind.org> <431DBD4A.5070706@cynapses.org> <431DF7E6.7020306@clutteredmind.org> Message-ID: <431E052E.6060903@cynapses.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ryan C. Gordon wrote: > Andreas Schneider wrote: > >>For me, alsa is broken now... > > > Run the game with "+set sdlmixsamps 32768 +set sdldevsamps 1024" > Done and works, but I have the sound lag now. > sdlmixsamps can be as big as you want, it just takes more memory; that's > how much space you have for the engine to mix ahead. Bigger is better, > but the default is probably fine. > > sdldevsamps is how much SDL wants to feed the audio device at a time. > Bigger numbers result in more audio latency, smaller number risk skips. > The defaults were halved in latest patch, which is probably why people > hear skipping now. > > If you have a really fast machine that can keep ahead of the audio > device, you can get away with 256 samples at 44100Khz (works here)...512 > to 1024 is probably more reasonable for everyone else, though. Lower > frequencies (the game defaults to 22050) can work with less samples at a > time. > I've tried +set sdlmixsamps 32768 +set sdldevsamps [0-1024] +set sndspeed [22050-44100] sound is always to late... > Also, please make sure you're using an SDL with ALSA if you want ALSA. > If you get a console message that says "SDL audio driver is > "(UNKNOWN)"", then your SDL is misconfigured. The SDL audio driver is ALSA ;) Hardware: AMD64 nForce3 board with nForce3 Soundcard > > --ryan. > -- andreas P.S.: quake3 with widescreen rocks ;) - -- http://www.cynapses.org/ - cybernetic synapses -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDHgUuwDudlFFJUr4RArPIAKCE8xXf1kMJCpUladfnEz5k5qgdbACdGQGD +HSxj9qcGps6nxRXh9JIvjM= =//6C -----END PGP SIGNATURE----- From arny at ats.s.bawue.de Tue Sep 6 17:15:55 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Tue, 6 Sep 2005 23:15:55 +0200 Subject: [quake3] Sound lag fix In-Reply-To: <431E052E.6060903@cynapses.org> References: <20050904200747.GG17517@timedoctor.org> <431DF7E6.7020306@clutteredmind.org> <431E052E.6060903@cynapses.org> Message-ID: <200509062315.58121.arny@ats.s.bawue.de> On Tuesday 06 September 2005 23:07, Andreas Schneider wrote: > The SDL audio driver is ALSA ;) Do you have an .asoundrc in your home directory? Like I already replied in a prior thread to the same topic: Many settings for alsa are tuned in there. period_size 1024 buffer_size 4096 Especially those two values are to watch out for. These are mine, by the way. You may want to set it lower, or if asoundrc does not exist yet, find a suitable one (there are loads to be found for nForce2, should work with nForce3 too, works for me on a nForce4 :) -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mail at cynapses.org Tue Sep 6 17:36:11 2005 From: mail at cynapses.org (Andreas Schneider) Date: Tue, 06 Sep 2005 23:36:11 +0200 Subject: [quake3] Sound lag fix In-Reply-To: <200509062315.58121.arny@ats.s.bawue.de> References: <20050904200747.GG17517@timedoctor.org> <431DF7E6.7020306@clutteredmind.org> <431E052E.6060903@cynapses.org> <200509062315.58121.arny@ats.s.bawue.de> Message-ID: <431E0BCB.5060600@cynapses.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thilo Schulz wrote: > On Tuesday 06 September 2005 23:07, Andreas Schneider wrote: > >>The SDL audio driver is ALSA ;) > > > Do you have an .asoundrc in your home directory? > Like I already replied in a prior thread to the same topic: Many settings for > alsa are tuned in there. > > period_size 1024 > buffer_size 4096 Works, thanks! -- andreas > > Especially those two values are to watch out for. These are mine, by the way. > You may want to set it lower, or if asoundrc does not exist yet, find a > suitable one (there are loads to be found for nForce2, should work with > nForce3 too, works for me on a nForce4 :) > - -- http://www.cynapses.org/ - cybernetic synapses -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (GNU/Linux) iD8DBQFDHgvLwDudlFFJUr4RAhqwAJ4wSm/T7wSMiBEzdmn14f2/9OScIwCdFCnu 5SFbp1viutfUuQmMqbMXIAo= =fppw -----END PGP SIGNATURE----- From icculus at clutteredmind.org Wed Sep 7 02:07:44 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Wed, 07 Sep 2005 02:07:44 -0400 Subject: [quake3] Hey all -- who here is an X config god? ;) In-Reply-To: <003401c5b29e$767bf820$6e01a8c0@sixfour> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> Message-ID: <431E83B0.2080004@clutteredmind.org> > glxinfo looks good on this run (think it was the same last time). > however, I just did remove my 16-bit display modes from the Screen and > now defaulting to 24 (to match yours), renamed the monitor and device > (still should have been fine before..), and added the Virtual setting > you had. Replicated the changes to both my XF86 config files (I can't > figure out why there are two...). THEN, I also just switched to have > Gnome as my default desktop. I can't test this here, but I might have fixed the failure to run on 16-bit displays. Can someone that is able to reproduce this issue please sync to the latest in Subversion and see if the problem goes away? Thanks, --ryan. From badcdev at gmail.com Wed Sep 7 03:32:23 2005 From: badcdev at gmail.com (BlackAura) Date: Wed, 7 Sep 2005 17:32:23 +1000 Subject: OpenAL In-Reply-To: <431E83B0.2080004@clutteredmind.org> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> Message-ID: <200509071732.23646.badcdev@gmail.com> I've been working on an OpenAL sound system for Quake 3, and I'm at the stage where everything appears to work correctly. I have a preliminary patch available for testing (Linux only) from here: http://files.frashii.com/~sp00nz/Doom/files/BlackAura/q3_openal_icculus.diff It contains an OpenAL sound system which entirely replaces the existing sound system. It also adds support for Ogg Vorbis music and sound effects in both the OpenAL and original sound systems, using a separate sound decoder system. It should be portable to other systems, except for the part that loads the OpenAL library at runtime (unix/linux_qal.c). That can be disabled by defining OPENAL_STATIC, and including the appropriate libraries. From arny at ats.s.bawue.de Wed Sep 7 04:47:06 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Wed, 7 Sep 2005 10:47:06 +0200 Subject: [quake3] OpenAL In-Reply-To: <200509071732.23646.badcdev@gmail.com> References: <200509060337.j863bn0O004209@euler.unh.edu> <431E83B0.2080004@clutteredmind.org> <200509071732.23646.badcdev@gmail.com> Message-ID: <200509071047.08615.arny@ats.s.bawue.de> On Wednesday 07 September 2005 09:32, BlackAura wrote: > It contains an OpenAL sound system which entirely replaces the existing > sound system. It also adds support for Ogg Vorbis music and sound effects > in both the OpenAL and original sound systems, using a separate sound > decoder system. Will you add mp3 support too? -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From badcdev at gmail.com Wed Sep 7 06:01:09 2005 From: badcdev at gmail.com (BlackAura) Date: Wed, 7 Sep 2005 20:01:09 +1000 Subject: [quake3] OpenAL In-Reply-To: <200509071047.08615.arny@ats.s.bawue.de> References: <200509060337.j863bn0O004209@euler.unh.edu> <200509071732.23646.badcdev@gmail.com> <200509071047.08615.arny@ats.s.bawue.de> Message-ID: <200509072001.09540.badcdev@gmail.com> On Wed, 7 Sep 2005 06:47 pm, Thilo Schulz wrote: > Will you add mp3 support too? I could do. I think libmad would probably do the trick. From fatty at gsklan.de Wed Sep 7 06:32:04 2005 From: fatty at gsklan.de (fatty) Date: Wed, 07 Sep 2005 12:32:04 +0200 Subject: [quake3] OpenAL In-Reply-To: <200509071732.23646.badcdev@gmail.com> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> <200509071732.23646.badcdev@gmail.com> Message-ID: <431EC1A4.6020505@gsklan.de> BlackAura wrote: > I've been working on an OpenAL sound system for Quake 3, and I'm at the stage > where everything appears to work correctly. I can confirm that. BlackAura's work is just great and I haven't recognised any problems with the OpenAL sound system yet. Even audio lag seems to be history. At least I didn't hear any difference to the original sound system (I think it even sounds way better). fatty From davebytes at comcast.net Wed Sep 7 09:53:58 2005 From: davebytes at comcast.net (David Chait) Date: Wed, 7 Sep 2005 09:53:58 -0400 Subject: [quake3] Hey all -- who here is an X config god? ;) References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> Message-ID: <002a01c5b3b3$981b21c0$6e01a8c0@sixfour> I'll see if I can take a look today since I was the original reporter -- assuming I can get my machine back into that state! ;) I still haven't had time to get past the UI version check thing, so I'll only be able to verify up til that point in the code. -d ----- Original Message ----- From: "Ryan C. Gordon" To: Sent: Wednesday, September 07, 2005 2:07 AM Subject: Re: [quake3] Hey all -- who here is an X config god? ;) > >> glxinfo looks good on this run (think it was the same last time). >> however, I just did remove my 16-bit display modes from the Screen and >> now defaulting to 24 (to match yours), renamed the monitor and device >> (still should have been fine before..), and added the Virtual setting you >> had. Replicated the changes to both my XF86 config files (I can't figure >> out why there are two...). THEN, I also just switched to have Gnome as >> my default desktop. > > I can't test this here, but I might have fixed the failure to run on > 16-bit displays. Can someone that is able to reproduce this issue please > sync to the latest in Subversion and see if the problem goes away? > > Thanks, > --ryan. > > From ludwig.nussel at suse.de Wed Sep 7 12:46:28 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Wed, 7 Sep 2005 18:46:28 +0200 Subject: [quake3] Hey all -- who here is an X config god? ;) In-Reply-To: <431E83B0.2080004@clutteredmind.org> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> Message-ID: <20050907164628.GA30708@suse.de> Ryan C. Gordon wrote: > I can't test this here, but I might have fixed the failure to run on > 16-bit displays. Can someone that is able to reproduce this issue please > sync to the latest in Subversion and see if the problem goes away? On my Laptop with Intel 855 chip it produces wrong color in 16bit now. Looks like it installs some palette. If you move the mouse outside of the window the color is correct again and stays that way when moving the mouse back into the q3 window. The non-SDL version works fine. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From ra at ra.is Wed Sep 7 13:04:35 2005 From: ra at ra.is (Richard Allen) Date: Wed, 07 Sep 2005 17:04:35 +0000 Subject: Strange results In-Reply-To: <20050907164628.GA30708@suse.de> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> <20050907164628.GA30708@suse.de> Message-ID: <431F1DA3.8010905@ra.is> Been playing around with your version of the code. Thanks for the Makefile system. It will make life much easier :) On my system, the quake3 binary from ID gives me c.a. 362 fps with my config When I compile your code from svn it gives me 478 fps :) However, the mouse sensitivity is far out from what I'm used to. I tracked that to the SDL code. If I compile a non SDL version, I get my sensitivity back but then the resulting binary is back to the 360 fps range. So Im curious... Why is the SDL version so much faster ? (timedemo 1; demo four) And Is it planned to get the SDL version's sensitivity to par with what we are used to ? From ludwig.nussel at suse.de Thu Sep 8 03:24:09 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Thu, 8 Sep 2005 09:24:09 +0200 Subject: [quake3] Adding EliteForce compatibility In-Reply-To: <200509051213.22386.arny@ats.s.bawue.de> References: <200509051213.22386.arny@ats.s.bawue.de> Message-ID: <20050908072409.GA2762@suse.de> Thilo Schulz wrote: > [...] > I think, Raven has not performed big changes in the engine and I'd like to try > adapting the codebase. My question now is: would you be willing to accept my > patches for these modifications, if I make the changes completely seperable > by using #ifdefs and stuff like that? I really like the idea of the icculus > codebase - and would also like to see IPv6 support in future versions of EF. To keep the q3 code clean it would probably be better to create a branch for EF and then merge the head back every once in a while. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From arny at ats.s.bawue.de Thu Sep 8 06:05:07 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Thu, 8 Sep 2005 12:05:07 +0200 Subject: [quake3] Adding EliteForce compatibility In-Reply-To: <20050908072409.GA2762@suse.de> References: <200509051213.22386.arny@ats.s.bawue.de> <20050908072409.GA2762@suse.de> Message-ID: <200509081205.10524.arny@ats.s.bawue.de> On Thursday 08 September 2005 09:24, Ludwig Nussel wrote: > To keep the q3 code clean it would probably be better to create a > branch for EF and then merge the head back every once in a while. That would actually require someone to do it, once in a while. But I see what you mean. I could also try to keep EF support in seperate files as much as possible I guess. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pomac at vapor.com Thu Sep 8 22:59:02 2005 From: pomac at vapor.com (Ian Kumlien) Date: Fri, 09 Sep 2005 04:59:02 +0200 Subject: [amd64, and possibly others] The crash problem in team arena Message-ID: <1126234742.19824.16.camel@localhost> Hi, I redid my installation... I was missing the patches etc, which i'll be taring up and putting in my q3 mirror tomrrow (http://pomac.netswarm.net/mirror/ ) and i managed to get the strange crash. From my debug data: -- valgrind-- ==31775== Invalid read of size 1 [data snipped] ==31775== Address 0x11803A8E7 is not stack'd, malloc'd or (recently) free'd This made me want to try gdb, which gave me this: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 46912510151392 (LWP 32106)] 0x00002aaab489989b in Item_Bind_HandleKey (item=0x100000000, key=1, down=4292783056) at ui_shared.c:3485 3485 id = BindingIDFromName(item->cvar); to which bt full yeilded: #0 0x00002aaab489989b in Item_Bind_HandleKey (item=0x100000000, key=1, down=4292783056) at ui_shared.c:3485 id = -1263967912 i = 1 Now, id shoud correspond to a binding, and is also used to index a array... =) I never got around this for some reason.... Anyways, it's running now becouse i replaced my .q3a/missionpack/q3config.cfg with one from my laptop. I attached both the working and non working for your... *cough* reading pleasure... They are kind of a mess though, since it's a friends settings against my quick hackup. For easy diffing i did this and attached it: cat q3config.cfg |sort >test1 cat q3config.cfg_old |sort >test2 (I first did it to check the data but i know to little about cvars to make anything useful from this.) -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- bind 0 "weapon 10" bind 1 "weapon 1" bind 2 "weapon 2" bind 3 "weapon 3" bind 4 "weapon 4" bind 5 "weapon 5" bind 6 "weapon 6" bind 7 "weapon 7" bind 8 "weapon 8" bind 9 "weapon 9" bind ALT "+strafe" bind a "+moveleft" bind c "+movedown" bind CTRL "+attack" bind DEL "+lookdown" bind d "+moveright" bind DOWNARROW "+back" bind END "centerview" bind ENTER "+button2" bind ESCAPE "togglemenu" bind F11 "screenshot" bind F1 "vote yes" bind F2 "vote no" bind F3 "ui_teamorders" bind LEFTARROW "+left" bind \ "+mlook" bind MOUSE1 "+attack" bind MOUSE2 "+strafe" bind MOUSE3 "+zoom" bind MWHEELDOWN "weapnext" bind MWHEELUP "weapprev" bind PAUSE "pause" bind PGDN "+lookup" bind RIGHTARROW "+right" bind s "+back" bind SHIFT "+speed" bind _ "sizedown" bind - "sizedown" bind = "sizeup" bind + "sizeup" bind SPACE "+moveup" bind TAB "+scores" bind t "messagemode" bind ` "toggleconsole" bind ~ "toggleconsole" bind UPARROW "+forward" bind / "weapnext" bind ] "weapnext" bind [ "weapprev" bind w "+forward" // generated by quake, do not modify seta capturelimit "5" seta cg_autoswitch "1" seta cg_bobpitch "0.002" seta cg_bobroll "0.002" seta cg_bobup "0.005" seta cg_brassTime "2500" seta cg_cameraOrbitDelay "35" seta cg_crosshairHealth "1" seta cg_crosshairSize "24" seta cg_crosshairX "0" seta cg_crosshairY "0" seta cg_currentSelectedPlayer "0" seta cg_currentSelectedPlayerName "" seta cg_deferPlayers "0" seta cg_draw2D "1" seta cg_draw3dIcons "1" seta cg_drawAmmoWarning "1" seta cg_drawAttacker "1" seta cg_drawCrosshair "4" seta cg_drawCrosshairNames "1" seta cg_drawFPS "0" seta cg_drawFriend "1" seta cg_drawGun "1" seta cg_drawIcons "1" seta cg_drawRewards "1" seta cg_drawSnapshot "0" seta cg_drawStatus "1" seta cg_drawTeamOverlay "0" seta cg_drawTimer "1" seta cg_forceModel "0" seta cg_fov "90" seta cg_gibs "1" seta cg_hudFiles "ui/hud.txt" seta cg_lagometer "1" seta cg_marks "1" seta cg_noProjectileTrail "0" seta cg_noTaunt "0" seta cg_noVoiceChats "0" seta cg_noVoiceText "0" seta cg_oldPlasma "1" seta cg_oldRail "1" seta cg_oldRocket "1" seta cg_predictItems "1" seta cg_railTrailTime "400" seta cg_runpitch "0.002" seta cg_runroll "0.005" seta cg_scorePlums "1" seta cg_selectedPlayer "1" seta cg_selectedPlayerName "Lionheart" seta cg_shadows "1" seta cg_simpleItems "0" seta cg_smoothClients "0" seta cg_stereoSeparation "0.4" seta cg_teamChatHeight "0" seta cg_teamChatsOnly "0" seta cg_teamChatTime "3000" seta cg_trueLightning "0.0" seta cg_viewsize "100" seta cg_zoomfov "22.5" seta cl_allowDownload "0" seta cl_anonymous "0" seta cl_freelook "1" seta cl_maxpackets "30" seta cl_maxPing "800" seta cl_mouseAccel "0" seta cl_packetdup "1" seta cl_pitchspeed "140" seta cl_punkbuster "1" seta cl_run "1" seta cl_yawspeed "140" seta cm_playerCurveClip "1" seta color1 "4" seta color2 "5" seta com_blood "1" seta com_hunkMegs "56" seta com_introplayed "1" seta com_maxfps "85" seta com_soundMegs "8" seta com_zoneMegs "16" seta dmflags "0" seta fraglimit "10" seta g_allowVote "1" seta g_banIPs "" seta g_blueTeam "Stroggs" seta g_filterBan "1" seta g_friendlyFire "0" seta g_log "games.log" seta g_logSync "0" seta g_maxGameClients "0" seta g_redTeam "Crusaders" seta g_spAwards "\a5\3\a4\25" seta g_spScores1 "" seta g_spScores2 "\l1\1\l0\1\l24\1" seta g_spScores3 "" seta g_spScores4 "" seta g_spScores5 "" seta g_spSkill "2" seta g_spVideos "\tier1\1" seta g_teamAutoJoin "0" seta g_teamForceBalance "0" seta g_warmup "15" seta handicap "100" seta headmodel "bones/bones" seta in_dgamouse "1" seta in_joystick "0" seta in_mouse "1" seta in_subframe "1" seta joy_threshold "0.150000" seta m_filter "0" seta m_forward "0.25" seta model "bones/bones" seta m_pitch "-0.022000" seta m_side "0.25" seta m_yaw "0.022" seta name "Player" seta r_allowExtensions "1" seta rate "25000" seta r_colorbits "32" seta r_customaspect "1" seta r_customheight "1024" seta r_customwidth "1600" seta r_depthbits "24" seta r_detailtextures "1" seta r_dlightBacks "1" seta r_drawSun "0" seta r_dynamiclight "1" seta r_ext_compiled_vertex_array "1" seta r_ext_compressed_textures "0" seta r_ext_gamma_control "1" seta r_ext_multitexture "1" seta r_ext_texture_env_add "0" seta r_facePlaneCull "1" seta r_fastsky "0" seta r_finish "1" seta r_flares "0" seta r_fullscreen "1" seta r_gamma "1" seta r_glDriver "libGL.so.1" seta r_ignoreFastPath "1" seta r_ignoreGLErrors "1" seta r_ignorehwgamma "0" seta r_inGameVideo "1" seta r_lastValidRenderer "GeForce4 4200 Go/AGP/SSE2" seta r_lodbias "0" seta r_lodCurveError "250" seta r_mode "6" seta r_overBrightBits "1" seta r_picmip "0" seta r_primitives "0" seta r_railCoreWidth "6" seta r_railSegmentLength "32" seta r_railWidth "16" seta r_roundImagesDown "1" seta r_simpleMipMaps "1" seta r_smp "0" seta r_stencilbits "0" seta r_stereo "0" seta r_subdivisions "4" seta r_swapInterval "0" seta r_texturebits "32" seta r_textureMode "GL_LINEAR_MIPMAP_LINEAR" seta r_vertexLight "0" seta sdldevsamps "0" seta sdlmixsamps "0" seta s_doppler "1" seta sensitivity "5.229166" seta server1 "" seta server10 "" seta server11 "" seta server12 "" seta server13 "" seta server14 "" seta server15 "" seta server16 "" seta server2 "" seta server3 "" seta server4 "" seta server5 "" seta server6 "" seta server7 "" seta server8 "" seta server9 "" seta sex "male" seta s_khz "22" seta s_mixahead "0.2" seta s_mixPreStep "0.05" seta s_musicvolume "0.25" seta snaps "20" seta sndbits "16" seta sndchannels "2" seta snddevice "/dev/dsp" seta sndspeed "0" seta s_separation "0.5" seta sv_floodProtect "1" seta sv_hostname "noname" seta sv_lanForceRate "1" seta sv_master2 "" seta sv_master3 "" seta sv_master4 "" seta sv_master5 "" seta sv_maxclients "6" seta sv_maxPing "0" seta sv_maxRate "0" seta sv_minPing "0" seta s_volume "0.8" seta sv_punkbuster "0" seta sv_pure "0" seta sv_strictAuth "1" seta team_headmodel "*James" seta team_model "james" seta timelimit "0" seta ui_actualNetGametype "3" seta ui_bigFont "0.4" seta ui_blueteam1 "0" seta ui_blueteam2 "0" seta ui_blueteam3 "0" seta ui_blueteam4 "0" seta ui_blueteam5 "0" seta ui_blueteam "Stroggs" seta ui_browserGameType "0" seta ui_browserMaster "0" seta ui_browserShowEmpty "1" seta ui_browserShowFull "1" seta ui_browserSortKey "4" seta ui_ctf_capturelimit "8" seta ui_ctf_friendly "0" seta ui_ctf_timelimit "30" seta ui_currentMap "0" seta ui_currentNetMap "26" seta ui_currentOpponent "0" seta ui_currentTier "0" seta ui_dedicated "0" seta ui_ffa_fraglimit "20" seta ui_ffa_timelimit "0" seta ui_findPlayer "Sarge" seta ui_gametype "3" seta ui_joinGametype "0" seta ui_lastServerRefresh_0 "Sep-2, 2005 at 23:14" seta ui_lastServerRefresh_1 "Sep-2, 2005 at 23:14" seta ui_lastServerRefresh_2 "" seta ui_lastServerRefresh_3 "Sep-2, 2005 at 23:14" seta ui_mapIndex "0" seta ui_menuFiles "ui/menus.txt" seta ui_netGametype "3" seta ui_netSource "3" seta ui_opponentName "Stroggs" seta ui_q3model "0" seta ui_recordSPDemo "0" seta ui_recordSPDemoName "" seta ui_redteam1 "0" seta ui_redteam2 "0" seta ui_redteam3 "0" seta ui_redteam4 "0" seta ui_redteam5 "0" seta ui_redteam "Pagans" seta ui_scoreAccuracy "0%" seta ui_scoreAssists "0" seta ui_scoreBase "0" seta ui_scoreCaptures "0" seta ui_scoreDefends "0" seta ui_scoreExcellents "0" seta ui_scoreGauntlets "0" seta ui_scoreImpressives "0" seta ui_scorePerfect "0" seta ui_scoreScore "0" seta ui_scoreShutoutBonus "0" seta ui_scoreSkillBonus "0" seta ui_scoreTeam "0 to 0" seta ui_scoreTime "00:00" seta ui_scoreTimeBonus "0" seta ui_serverStatusTimeOut "7000" seta ui_smallFont "0.25" seta ui_teamArenaFirstRun "1" seta ui_team_fraglimit "0" seta ui_team_friendly "1" seta ui_teamName "Crusaders" seta ui_team_timelimit "20" seta ui_tourney_fraglimit "0" seta ui_tourney_timelimit "15" seta vm_cgame "2" seta vm_game "2" seta vm_ui "2" unbindall -------------- next part -------------- bind 0 "weapon 10" bind 1 "weapon 1" bind 2 "weapon 2" bind 3 "weapon 3" bind 4 "weapon 4" bind 5 "weapon 5" bind 6 "weapon 6" bind 7 "weapon 7" bind 8 "weapon 8" bind 9 "weapon 9" bind ALT "+strafe" bind a "+moveleft" bind c "+movedown" bind CTRL "+attack" bind DEL "+lookdown" bind d "+moveright" bind DOWNARROW "+back" bind END "centerview" bind ENTER "+button2" bind ESCAPE "togglemenu" bind F11 "screenshot" bind F1 "vote yes" bind F2 "vote no" bind F3 "ui_teamorders" bind LEFTARROW "+left" bind \ "+mlook" bind MOUSE1 "+attack" bind MOUSE2 "+moveup" bind MOUSE3 "+zoom" bind MWHEELDOWN "weapnext" bind MWHEELUP "weapprev" bind PAUSE "pause" bind PGDN "+lookup" bind RIGHTARROW "+right" bind s "+back" bind SHIFT "+speed" bind _ "sizedown" bind - "sizedown" bind = "sizeup" bind + "sizeup" bind SPACE "+moveup" bind TAB "+scores" bind t "messagemode" bind ` "toggleconsole" bind ~ "toggleconsole" bind UPARROW "+forward" bind / "weapnext" bind ] "weapnext" bind [ "weapprev" bind w "+forward" // generated by quake, do not modify seta capturelimit "5" seta cg_autoswitch "1" seta cg_bobpitch "0.002" seta cg_bobroll "0.002" seta cg_brassTime "2500" seta cg_cameraOrbitDelay "50" seta cg_crosshairHealth "1" seta cg_crosshairSize "24" seta cg_crosshairX "0" seta cg_crosshairY "0" seta cg_deferPlayers "1" seta cg_draw2D "1" seta cg_draw3dIcons "1" seta cg_drawAmmoWarning "1" seta cg_drawAttacker "1" seta cg_drawCrosshair "4" seta cg_drawCrosshairNames "1" seta cg_drawFPS "0" seta cg_drawFriend "1" seta cg_drawGun "1" seta cg_drawIcons "1" seta cg_drawRewards "1" seta cg_drawSnapshot "0" seta cg_drawStatus "1" seta cg_drawTeamOverlay "0" seta cg_drawTimer "1" seta cg_forceModel "0" seta cg_fov "90" seta cg_gibs "1" seta cg_hudFiles "ui/hud.txt" seta cg_lagometer "1" seta cg_marks "1" seta cg_noProjectileTrail "0" seta cg_noTaunt "0" seta cg_noVoiceChats "0" seta cg_noVoiceText "0" seta cg_oldPlasma "1" seta cg_oldRail "1" seta cg_oldRocket "1" seta cg_predictItems "1" seta cg_railTrailTime "400" seta cg_runpitch "0.002" seta cg_runroll "0.005" seta cg_scorePlums "1" seta cg_selectedPlayer "0" seta cg_selectedPlayerName "" seta cg_shadows "1" seta cg_simpleItems "0" seta cg_smoothClients "0" seta cg_stereoSeparation "0.4" seta cg_teamChatHeight "0" seta cg_teamChatsOnly "0" seta cg_teamChatTime "3000" seta cg_trueLightning "0.0" seta cg_viewsize "100" seta cg_zoomfov "22.5" seta cl_allowDownload "0" seta cl_anonymous "0" seta cl_freelook "1" seta cl_maxpackets "30" seta cl_maxPing "800" seta cl_mouseAccel "0" seta cl_packetdup "1" seta cl_pitchspeed "140" seta cl_run "1" seta cl_yawspeed "140" seta cm_playerCurveClip "1" seta color1 "4" seta color2 "5" seta com_blood "1" seta com_hunkMegs "56" seta com_introplayed "1" seta com_maxfps "85" seta com_soundMegs "8" seta com_zoneMegs "16" seta dmflags "0" seta fraglimit "10" seta g_allowVote "1" seta g_banIPs "" seta g_blueTeam "Stroggs" seta g_filterBan "1" seta g_friendlyFire "0" seta g_log "games.log" seta g_logSync "0" seta g_maxGameClients "0" seta g_redTeam "Crusaders" seta g_spAwards "\a5\2\a4\15" seta g_spScores1 "" seta g_spScores2 "\l0\1\l24\1" seta g_spScores3 "" seta g_spScores4 "" seta g_spScores5 "" seta g_spSkill "2" seta g_spVideos "\tier1\1" seta g_teamAutoJoin "0" seta g_teamForceBalance "0" seta g_warmup "15" seta handicap "100" seta headmodel "sarge/default" seta in_dgamouse "1" seta in_joystick "0" seta in_mouse "1" seta in_subframe "1" seta joy_threshold "0.150000" seta m_filter "0" seta m_forward "0.25" seta model "sarge/default" seta m_pitch "-0.022000" seta m_side "0.25" seta m_yaw "0.022" seta name "Player" seta r_allowExtensions "1" seta rate "25000" seta r_colorbits "0" seta r_customaspect "1" seta r_customheight "1024" seta r_customwidth "1600" seta r_depthbits "0" seta r_detailtextures "1" seta r_dlightBacks "1" seta r_drawSun "0" seta r_dynamiclight "1" seta r_ext_compiled_vertex_array "1" seta r_ext_compressed_textures "0" seta r_ext_gamma_control "1" seta r_ext_multitexture "1" seta r_ext_texture_env_add "0" seta r_facePlaneCull "1" seta r_fastsky "0" seta r_finish "0" seta r_flares "0" seta r_fullscreen "1" seta r_gamma "1" seta r_glDriver "libGL.so.1" seta r_ignoreFastPath "1" seta r_ignoreGLErrors "1" seta r_ignorehwgamma "0" seta r_inGameVideo "1" seta r_lastValidRenderer "GeForce 6600/PCI/SSE2" seta r_lodbias "0" seta r_lodCurveError "250" seta r_mode "6" seta r_overBrightBits "1" seta r_picmip "0" seta r_primitives "0" seta r_railCoreWidth "6" seta r_railSegmentLength "32" seta r_railWidth "16" seta r_roundImagesDown "1" seta r_simpleMipMaps "1" seta r_smp "0" seta r_stencilbits "0" seta r_stereo "0" seta r_subdivisions "4" seta r_swapInterval "0" seta r_texturebits "0" seta r_textureMode "GL_LINEAR_MIPMAP_LINEAR" seta r_vertexLight "0" seta sdldevsamps "0" seta sdlmixsamps "0" seta s_doppler "1" seta sensitivity "5" seta server1 "" seta server10 "" seta server11 "" seta server12 "" seta server13 "" seta server14 "" seta server15 "" seta server16 "" seta server2 "" seta server3 "" seta server4 "" seta server5 "" seta server6 "" seta server7 "" seta server8 "" seta server9 "" seta sex "male" seta s_khz "22" seta s_mixahead "0.2" seta s_mixPreStep "0.05" seta s_musicvolume "0.25" seta snaps "20" seta sndbits "16" seta sndchannels "2" seta snddevice "/dev/dsp" seta sndspeed "0" seta s_separation "0.5" seta sv_floodProtect "1" seta sv_hostname "noname" seta sv_lanForceRate "1" seta sv_master2 "" seta sv_master3 "" seta sv_master4 "" seta sv_master5 "" seta sv_maxclients "6" seta sv_maxPing "0" seta sv_maxRate "0" seta sv_minPing "0" seta s_volume "0.8" seta sv_pure "0" seta sv_strictAuth "1" seta team_headmodel "*James" seta team_model "James" seta timelimit "0" seta ui_actualNetGametype "3" seta ui_bigFont "0.4" seta ui_blueteam1 "0" seta ui_blueteam2 "0" seta ui_blueteam3 "0" seta ui_blueteam4 "0" seta ui_blueteam5 "0" seta ui_blueteam "Stroggs" seta ui_browserGameType "0" seta ui_browserMaster "0" seta ui_browserShowEmpty "1" seta ui_browserShowFull "1" seta ui_browserSortKey "4" seta ui_ctf_capturelimit "8" seta ui_ctf_friendly "0" seta ui_ctf_timelimit "30" seta ui_currentMap "2" seta ui_currentNetMap "0" seta ui_currentOpponent "0" seta ui_currentTier "0" seta ui_dedicated "0" seta ui_ffa_fraglimit "20" seta ui_ffa_timelimit "0" seta ui_findPlayer "Sarge" seta ui_gametype "3" seta ui_joinGametype "0" seta ui_lastServerRefresh_0 "" seta ui_lastServerRefresh_1 "" seta ui_lastServerRefresh_2 "" seta ui_lastServerRefresh_3 "" seta ui_mapIndex "2" seta ui_menuFiles "ui/menus.txt" seta ui_netGametype "3" seta ui_netSource "0" seta ui_opponentName "Stroggs" seta ui_q3model "0" seta ui_recordSPDemo "0" seta ui_redteam1 "0" seta ui_redteam2 "0" seta ui_redteam3 "0" seta ui_redteam4 "0" seta ui_redteam5 "0" seta ui_redteam "Pagans" seta ui_scoreAccuracy "0%" seta ui_scoreAssists "0" seta ui_scoreBase "0" seta ui_scoreCaptures "0" seta ui_scoreDefends "0" seta ui_scoreExcellents "0" seta ui_scoreGauntlets "0" seta ui_scoreImpressives "0" seta ui_scorePerfect "0" seta ui_scoreScore "0" seta ui_scoreShutoutBonus "0" seta ui_scoreSkillBonus "0" seta ui_scoreTeam "0 to 0" seta ui_scoreTime "00:00" seta ui_scoreTimeBonus "0" seta ui_serverStatusTimeOut "7000" seta ui_smallFont "0.25" seta ui_teamArenaFirstRun "1" seta ui_team_fraglimit "0" seta ui_team_friendly "1" seta ui_teamName "Crusaders" seta ui_team_timelimit "20" seta ui_tourney_fraglimit "0" seta ui_tourney_timelimit "15" seta vm_cgame "0" seta vm_game "0" seta vm_ui "0" unbindall -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From tim at ngus.net Fri Sep 16 10:09:16 2005 From: tim at ngus.net (Tim Angus) Date: Fri, 16 Sep 2005 15:09:16 +0100 Subject: Client timer resolution bug patch Message-ID: <20050916150916.3e80f027.tim@ngus.net> http://forums.wireheadstudios.org/index.php?act=ST&f=11&t=2749 http://test.g-rave.nl/quake3-phoenix.diff This is one of those bugs that id should have fixed looooooooong ago. From zakk at timedoctor.org Fri Sep 16 15:45:27 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 16 Sep 2005 12:45:27 -0700 Subject: [quake3] Client timer resolution bug patch In-Reply-To: <20050916150916.3e80f027.tim@ngus.net> References: <20050916150916.3e80f027.tim@ngus.net> Message-ID: <20050916194527.GA2196@timedoctor.org> On Fri, Sep 16, 2005 at 03:09:16PM +0100, Tim Angus wrote: > http://forums.wireheadstudios.org/index.php?act=ST&f=11&t=2749 > http://test.g-rave.nl/quake3-phoenix.diff > > This is one of those bugs that id should have fixed looooooooong ago. Can anybody get to this test.g-rave.nl server? I can't. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From xipher at xipher.us Fri Sep 16 15:56:11 2005 From: xipher at xipher.us (Joshua Hoppes) Date: Fri, 16 Sep 2005 14:56:11 -0500 Subject: [quake3] Client timer resolution bug patch In-Reply-To: <20050916194527.GA2196@timedoctor.org> References: <20050916150916.3e80f027.tim@ngus.net> <20050916194527.GA2196@timedoctor.org> Message-ID: <432B235B.8090005@xipher.us> yea, I got it, and I attached the patch Zachary J. Slater wrote: > On Fri, Sep 16, 2005 at 03:09:16PM +0100, Tim Angus wrote: >> http://forums.wireheadstudios.org/index.php?act=ST&f=11&t=2749 >> http://test.g-rave.nl/quake3-phoenix.diff >> >> This is one of those bugs that id should have fixed looooooooong ago. > > Can anybody get to this test.g-rave.nl server? > I can't. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: quake3-phoenix.diff URL: From arny at ats.s.bawue.de Fri Sep 16 17:26:48 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 16 Sep 2005 23:26:48 +0200 Subject: MD4 Model Support Message-ID: <200509162326.51025.arny@ats.s.bawue.de> Hello, Is there any mod, etc. that actually uses the md4 model routines that seem to be in Quake3? They seem like half finished and it doesn't look like the main quake3 game makes use of these. Raven software seems to have "finished" md4 capabilities (bone animation and stuff), and called the format ".mdr". So I was asking, what your plan with these routines is. I have managed to make the GPL'ed md3view source code released by Raven work on the project, and can provide patches soon that would somewhat complete the md4 stubs currently still in the codebase. Is there any interest in this happening? If not I could save myself some work of doing this in a clean way :) -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From zakk at timedoctor.org Fri Sep 16 21:32:07 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 16 Sep 2005 18:32:07 -0700 Subject: [quake3] MD4 Model Support In-Reply-To: <200509162326.51025.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> Message-ID: <20050917013207.GC2196@timedoctor.org> I'll accept MD4 model support as long as it doesn't effect the game besides adding this support, is reasonably clean, and I would prefer that it be ifdef'd (on by default, but able to be off) just in case. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Fri Sep 16 21:37:40 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 16 Sep 2005 18:37:40 -0700 Subject: [quake3] Client timer resolution bug patch In-Reply-To: <20050916150916.3e80f027.tim@ngus.net> References: <20050916150916.3e80f027.tim@ngus.net> Message-ID: <20050917013740.GD2196@timedoctor.org> Thanks, applied. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From tim at ngus.net Fri Sep 16 21:39:56 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 17 Sep 2005 02:39:56 +0100 Subject: Client timer resolution bug patch In-Reply-To: <20050917013740.GD2196@timedoctor.org> References: <20050916150916.3e80f027.tim@ngus.net> <20050917013740.GD2196@timedoctor.org> Message-ID: <20050917023956.5ec7769a.tim@ngus.net> On Fri, 16 Sep 2005 18:37:40 -0700 Zachary wrote: > Thanks, applied. I better point out I didn't actually write this patch, just linked to it. From zakk at timedoctor.org Fri Sep 16 21:44:19 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 16 Sep 2005 18:44:19 -0700 Subject: [quake3] Re: Client timer resolution bug patch In-Reply-To: <20050917023956.5ec7769a.tim@ngus.net> References: <20050916150916.3e80f027.tim@ngus.net> <20050917013740.GD2196@timedoctor.org> <20050917023956.5ec7769a.tim@ngus.net> Message-ID: <20050917014419.GE2196@timedoctor.org> AHDAHDSHDAS. Please do not submit patches that aren't yours. I try to make a good faith attempt here on everything. The GPL be damned, I don't ever want to hurt somebodies ego because we accepted a patch they didn't actually intend for us. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From tim at ngus.net Fri Sep 16 21:49:07 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 17 Sep 2005 02:49:07 +0100 Subject: Client timer resolution bug patch In-Reply-To: <20050917014419.GE2196@timedoctor.org> References: <20050916150916.3e80f027.tim@ngus.net> <20050917013740.GD2196@timedoctor.org> <20050917023956.5ec7769a.tim@ngus.net> <20050917014419.GE2196@timedoctor.org> Message-ID: <20050917024907.2e2cf7d0.tim@ngus.net> On Fri, 16 Sep 2005 18:44:19 -0700 Zachary wrote: > AHDAHDSHDAS. > Please do not submit patches that aren't yours. > I try to make a good faith attempt here on everything. > The GPL be damned, I don't ever want to hurt somebodies ego > because we accepted a patch they didn't actually intend for > us. Sorry, I thought it was kind of obvious. From arny at ats.s.bawue.de Fri Sep 16 23:15:59 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Sat, 17 Sep 2005 05:15:59 +0200 Subject: [quake3] MD4 Model Support In-Reply-To: <20050917013207.GC2196@timedoctor.org> References: <200509162326.51025.arny@ats.s.bawue.de> <20050917013207.GC2196@timedoctor.org> Message-ID: <200509170516.02223.arny@ats.s.bawue.de> On Saturday 17 September 2005 03:32, Zachary J. Slater wrote: > I'll accept MD4 model support as long as it doesn't effect the game > besides adding this support, is reasonably clean, and I would prefer > that it be ifdef'd (on by default, but able to be off) just in case. Yes, I don't think this will be a problem. The thing is, that I am not sure about whether the current md4 routines are being used. You will find md4 structures defined in qcommon/qfiles.h as well as them being used in tr_model.c and tr_animation.c. Before I break compatibility with any other mod, game etc. that is out there, I would like to know whether I shall leave the current md4 implementation "as-is" and use an entirely different set of structures, or whether I should - as you say - just ifdef all the existing md4 stuff and add my own to to the support trick. It doesn't look like the current md4 implementation is much more than something half-finished. But I could be wrong. Maybe Ryan could give a more clear answer on this one? -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From zakk at timedoctor.org Fri Sep 16 23:23:20 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 16 Sep 2005 20:23:20 -0700 Subject: [quake3] MD4 Model Support In-Reply-To: <200509170516.02223.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <20050917013207.GC2196@timedoctor.org> <200509170516.02223.arny@ats.s.bawue.de> Message-ID: <20050917032319.GA6182@timedoctor.org> To be clear, I'd want yours ifdef'd, the other stuff should just be nuked if it isn't being used and you provide equivilent/ better support. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From simon at 16hz.net Sat Sep 17 05:04:11 2005 From: simon at 16hz.net (Simon Morgan) Date: Sat, 17 Sep 2005 10:04:11 +0100 Subject: [quake3] Client timer resolution bug patch In-Reply-To: <20050917013740.GD2196@timedoctor.org> References: <20050916150916.3e80f027.tim@ngus.net> <20050917013740.GD2196@timedoctor.org> Message-ID: <20050917090411.GA22780@bollo.16hz.net> On Fri, Sep 16, 2005 at 06:37:40PM -0700, Zachary J. Slater wrote: > Thanks, applied. I know it's anal but you should fix the fact that the guy who wrote the patch thinks that the code base should accomodate his coding style. If you don't want to but are willing to accept a patch that does this then I'll do it. -- UNIX was half a billion (500000000) seconds old on Tue Nov 5 00:53:20 1985 GMT (measuring since the time(2) epoch). -- Andy Tannenbaum From floam at sh.nu Sat Sep 17 13:26:01 2005 From: floam at sh.nu (Aaron Gyes) Date: Sat, 17 Sep 2005 10:26:01 -0700 Subject: [quake3] Client timer resolution bug patch In-Reply-To: <20050917090411.GA22780@bollo.16hz.net> References: <20050916150916.3e80f027.tim@ngus.net> <20050917013740.GD2196@timedoctor.org> <20050917090411.GA22780@bollo.16hz.net> Message-ID: <1126977961.12823.0.camel@localhost> On Sat, 2005-09-17 at 10:04 +0100, Simon Morgan wrote: > On Fri, Sep 16, 2005 at 06:37:40PM -0700, Zachary J. Slater wrote: > > Thanks, applied. > > I know it's anal but you should fix the fact that the guy who wrote the > patch thinks that the code base should accomodate his coding style. If > you don't want to but are willing to accept a patch that does this then > I'll do it. Yeah. I was kind of suprised zakk applied it verbatim. From icculus at clutteredmind.org Sat Sep 17 18:09:52 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Sat, 17 Sep 2005 15:09:52 -0700 Subject: [quake3] MD4 Model Support In-Reply-To: <200509170516.02223.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <20050917013207.GC2196@timedoctor.org> <200509170516.02223.arny@ats.s.bawue.de> Message-ID: <432C9430.8040006@clutteredmind.org> > Maybe Ryan could give a more clear answer on this one? I don't know, sorry. --ryan. From tim at ngus.net Sun Sep 18 21:28:37 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 19 Sep 2005 02:28:37 +0100 Subject: A bunch of patches Message-ID: <20050919022837.5c3cb028.tim@ngus.net> Derived from http://www.quakesrc.org/forums/viewtopic.php?t=5374&postdays=0&postorder=asc&start=0 The CL_Rcon_f fix is just a series of stupid buffer overflow bugs. It looks as if the q3infoboom bug has already been fixed in ioQ3 in a different way, though this patch addresses the cause. The existing fix should stay since it's a sensible sanity check anyway. The rest are fairly self explanatory. Disclaimer: I created the patches, but I didn't discover the bugs. For future reference: Tim Angus = Timbo = http://tremulous.net/ -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Q3-CL_Rcon_f-overflow.diff URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Q3-COM_ParseExt-overwrite.diff URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Q3-missing-trap_FS_FCloseFile.diff URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Q3-q3infoboom-fix.diff URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Q3-q3msgboom-fix.diff URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Q3-qboolean-confusion.diff URL: From tim at ngus.net Sun Sep 18 21:32:31 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 19 Sep 2005 02:32:31 +0100 Subject: MD4 Model Support In-Reply-To: <200509170516.02223.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <20050917013207.GC2196@timedoctor.org> <200509170516.02223.arny@ats.s.bawue.de> Message-ID: <20050919023231.0d999902.tim@ngus.net> On Sat, 17 Sep 2005 05:15:59 +0200 Thilo wrote: > Yes, I don't think this will be a problem. The thing is, that I am not > sure about whether the current md4 routines are being used. You will > find md4 structures defined in qcommon/qfiles.h as well as them being > used in tr_model.c and tr_animation.c. Before I break compatibility > with any other mod, game etc. that is out there, To the best of my knowledge none of the MD4 stuff is used. There certainly isn't any interface available to cgame at any rate, so it isn't possible to use it from a mod. MD4 was something that was promised to the mod community in "the next patch" for a couple of iterations, but it never quite materialised. The implementation that is there is most likely throughly unfinished and/or cruft. From arny at ats.s.bawue.de Tue Sep 20 11:34:54 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Tue, 20 Sep 2005 17:34:54 +0200 Subject: [quake3] MD4 Model Support In-Reply-To: <20050917032319.GA6182@timedoctor.org> References: <200509162326.51025.arny@ats.s.bawue.de> <200509170516.02223.arny@ats.s.bawue.de> <20050917032319.GA6182@timedoctor.org> Message-ID: <200509201735.02404.arny@ats.s.bawue.de> On Saturday 17 September 2005 05:23, Zachary J. Slater wrote: > To be clear, I'd want yours ifdef'd, the other stuff should > just be nuked if it isn't being used and you provide equivilent/ > better support. Done as you wish. Attached is the patch, it should apply cleanly against the most current revision. The header qcommon/qfiles.h contains the new md4 (mdr) structures that support the mdr file format. It is also there, where I have placed "#define RAVENMD4", so by undefining this all mdr support will be turned off. Much of this stuff is my own work, and if I reused code it was only from icculus' quake3 itself. Of course, I tested everything before making this patch, and everything worked fine. The patch also contains a little update to the 8 bit jpeg support - it will obliviate the need for a second malloc for RGBA conversion, so loading will be more efficient. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: ioquake3-mdrsupport.diff Type: text/x-diff Size: 43001 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From arny at ats.s.bawue.de Tue Sep 20 11:42:40 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Tue, 20 Sep 2005 17:42:40 +0200 Subject: [quake3] MD4 Model Support In-Reply-To: <200509201735.02404.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <20050917032319.GA6182@timedoctor.org> <200509201735.02404.arny@ats.s.bawue.de> Message-ID: <200509201742.42479.arny@ats.s.bawue.de> Ouch. I sent the 40k patch to the mailing list, should have done directly zo Zakk, sorry. On Tuesday 20 September 2005 17:34, Thilo Schulz wrote: > Much of this stuff is my own work, and if I reused code it was only from > icculus' quake3 itself. Also a correction: I am using the MC uncompression routines from Raven's GPLed md3view. This would be the only thing though. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mandin.patrice at wanadoo.fr Tue Sep 20 14:24:02 2005 From: mandin.patrice at wanadoo.fr (Patrice Mandin) Date: Tue, 20 Sep 2005 20:24:02 +0200 Subject: Using SDL functions to load libraries, and cross-platform stuff Message-ID: <20050920202402.3ed176e1.mandin.patrice@wanadoo.fr> Hello, As SDL support has been added for audio and video, it would be nice to also use it to load libraries. This will remove some non-cross platform stuff. Also, in code/qcommon/common.c, is included if system is neither linux nor macosx, which is wrong. It should be included only on win32 systems. There are many more things like that in the source, like code/game/q_shared.h file that holds many system-dependent stuff, or code/unix/unix_main.c which is full of #ifdef stuff. This is where having autotools support would be nice, but I know some people here don't want it atm. -- Patrice Mandin WWW: http://pmandin.atari.org/ Programmeur Linux, Atari Sp?cialit?: D?veloppement, jeux From zakk at timedoctor.org Tue Sep 20 14:47:03 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 20 Sep 2005 11:47:03 -0700 Subject: [quake3] Using SDL functions to load libraries, and cross-platform stuff In-Reply-To: <20050920202402.3ed176e1.mandin.patrice@wanadoo.fr> References: <20050920202402.3ed176e1.mandin.patrice@wanadoo.fr> Message-ID: <20050920184702.GA8632@timedoctor.org> On Tue, Sep 20, 2005 at 08:24:02PM +0200, Patrice Mandin wrote: > There are many more things like that in the source, like > code/game/q_shared.h file that holds many system-dependent stuff, or > code/unix/unix_main.c which is full of #ifdef stuff. This is where > having autotools support would be nice, but I know some people > here don't want it atm. > Autotools support will never be added, real package maintainers (gentoo people), have sent me patches for things and do not require it. It will never be added, and I honestly hope this is the last time anyone asks. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From icculus at clutteredmind.org Thu Sep 22 00:25:33 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Thu, 22 Sep 2005 00:25:33 -0400 Subject: MingW32... Message-ID: <4332323D.8080608@clutteredmind.org> A couple of comments, brought about by the big-ass Mingw32 patch that just showed up in Subversion unannounced... There is this: + // The only difference is _ added to the C symbols. It seems mingw + // mangles all symbols this way, like linux gcc does when producing + // a.out instead of elf In the GCC case, you should either use the __attribute__((alias("_sym"))) format to make sure we find the right thing: http://www.ohse.de/uwe/articles/gcc-attributes.html#func-alias Or just do this: __asm__("_doAsmCall: \n\t" "doAsmCall: \n\t" " movl (%%edi),%eax \n\t" "_doAsmCall" and "doAsmCall" will both be valid symbols in the binary with the same address, so the linker can choose whichever it wants for the platform and get the right thing. This tends to be cleaner and more portable (as far as x86 goes). It also prevents a massive cut-and-paste of error-prone asm code. There are a couple of these around the codebase: #if !( (defined __linux__ || __FreeBSD__ || __MINGW32__) && (defined __i386__) && (!defined C_ONLY)) // Can we replace all this nonsense with either a check for __GNUC__? This is only going to get uglier with each new platform, and be error prone for the ones that are missed. "3. Find the following in DXSDK/Include/dinput.h..." Ugh, we have to modify a system header?! --ryan. From tim at ngus.net Thu Sep 22 09:51:53 2005 From: tim at ngus.net (Tim Angus) Date: Thu, 22 Sep 2005 14:51:53 +0100 Subject: MingW32... In-Reply-To: <4332323D.8080608@clutteredmind.org> References: <4332323D.8080608@clutteredmind.org> Message-ID: <20050922145153.10387703.tim@ngus.net> On Thu, 22 Sep 2005 00:25:33 -0400 Ryan wrote: > A couple of comments, brought about by the big-ass Mingw32 patch that > just showed up in Subversion unannounced... I announce it on IRC :P. (#icculus.org/quake3). > There is this: > > + // The only difference is _ added to the C symbols. It seems > mingw + // mangles all symbols this way, like linux gcc does when > producing + // a.out instead of elf > > In the GCC case, you should either use the > __attribute__((alias("_sym"))) format to make sure we find the right > thing: Hmm, good idea. I dicked about with __attribute__s a little when I was trying to figure out what was preventing it from compiling/running, but it didn't occur to me to use them this way. > __asm__("_doAsmCall: \n\t" > "doAsmCall: \n\t" > " movl (%%edi),%eax \n\t" This only works in this instance, i.e. C referencing an asm symbol. The asm also references C symbols, and obviously you can't try calling multiple symbols until it works. Therefore, the alias solution seems preferable. > There are a couple of these around the codebase: > > #if !( (defined __linux__ || __FreeBSD__ || __MINGW32__) && (defined > __i386__) && (!defined C_ONLY)) // > > Can we replace all this nonsense with either a check for __GNUC__? > This is only going to get uglier with each new platform, and be error > prone for the ones that are missed. That seems sensible. > "3. Find the following in DXSDK/Include/dinput.h..." > > Ugh, we have to modify a system header?! Yeah, stupid isn't it? Some cretinous Microsoft engineer has typedefed the same thing twice in the header. It's not like it's even that obscure, they're right next to each other -- perhaps this is generated code. In any case, while MSVC merrily compiles the multiple typedefs without complaining, gcc shits a brick. Two things occur to me: 1) Perhaps there is a way of convincing gcc to allow the multiple typedefs with some GNU C preproccesor voodoo surrounding the #include line. 2) Presumably wine(lib) has equivalents of the required headers that are correct enough to work. If such headers exist, they could probably be included with ioQ3, which saves a lot of hassle. From scott at scottbrooks.ca Thu Sep 22 21:00:30 2005 From: scott at scottbrooks.ca (Scott Brooks) Date: Thu, 22 Sep 2005 19:00:30 -0600 Subject: [quake3] MD4 Model Support In-Reply-To: <200509201742.42479.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <20050917032319.GA6182@timedoctor.org> <200509201735.02404.arny@ats.s.bawue.de> <200509201742.42479.arny@ats.s.bawue.de> Message-ID: Do you, or does anyone else have any docs on this file format? I've tried to hunt around google/raven's site and I can't really find anything usefull. Thanks. On 20-Sep-05, at 9:42 AM, Thilo Schulz wrote: > Ouch. > > I sent the 40k patch to the mailing list, should have done directly > zo Zakk, > sorry. > > On Tuesday 20 September 2005 17:34, Thilo Schulz wrote: > >> Much of this stuff is my own work, and if I reused code it was >> only from >> icculus' quake3 itself. >> > > Also a correction: I am using the MC uncompression routines from > Raven's GPLed > md3view. This would be the only thing though. > > -- > Thilo Schulz > From jaq at spacepants.org Fri Sep 23 01:53:51 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Fri, 23 Sep 2005 15:53:51 +1000 Subject: [quake3-commits] r99 - in trunk/code: client game qcommon renderer unix win32 In-Reply-To: <20050923025916.22200.qmail@icculus.org> References: <20050923025916.22200.qmail@icculus.org> Message-ID: <20050923055351.GA18662@spacepants.org> This one time, at band camp, DONOTREPLY at icculus.org wrote: >Author: tma >Date: 2005-09-22 22:59:15 -0400 (Thu, 22 Sep 2005) >New Revision: 99 > >Log: >* Replaced lots of __linux__ || __FreeBSD__ with __GNUC__ >* MinGW port now uses asm versions of SnapVector and ftol >* Improved the handling of mangled symbols in vm_x86.c I hate to be so anal, especially as I'm not contributing any patches myself (so bring on the flames if I'm out of line), but can you break each of these into separate commits in the future? (i.e. so one commit == one feature/fix) It's easier to read the patches if they carry a single meaning, and if there's a problem it's easier to rollback. Thanks. From arny at ats.s.bawue.de Fri Sep 23 08:22:23 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 23 Sep 2005 14:22:23 +0200 Subject: [quake3] MD4 Model Support In-Reply-To: References: <200509162326.51025.arny@ats.s.bawue.de> <200509201742.42479.arny@ats.s.bawue.de> Message-ID: <200509231422.26543.arny@ats.s.bawue.de> On Friday 23 September 2005 03:00, Scott Brooks wrote: > Do you, or does anyone else have any docs on this file format? I've > tried to hunt around google/raven's site and I can't really find > anything usefull. Yes, there are some tools that can be used to handle these files: http://www.3ddownloads.com/eliteforce/official md3view to look at the models Here is a milkshape plugin to export mdr files: http://eliteforce2.filefront.com/file/Milkshape_MDR_Exporter;21437 More releases, including carcass that outputs to mdr format from several source files (XSI animation files and the models): http://eliteforce2.filefront.com/files/Elite_Force/Official_Releases/Others;385 http://dynamic.gamespy.com/~thesynapse/skinning.php Here's a skinning tutorial for .mdr files http://forums.ravensoft.com/ib/ikonboard.pl?act=ST;f=4;t=2081 Soldier of Fortune seems to have some goul stuff in there, but the animations should be pretty much the same. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From tim at ngus.net Fri Sep 23 13:13:42 2005 From: tim at ngus.net (Tim Angus) Date: Fri, 23 Sep 2005 18:13:42 +0100 Subject: MD4 Model Support In-Reply-To: <200509201735.02404.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <200509170516.02223.arny@ats.s.bawue.de> <20050917032319.GA6182@timedoctor.org> <200509201735.02404.arny@ats.s.bawue.de> Message-ID: <20050923181342.0dcafb3f.tim@ngus.net> On Tue, 20 Sep 2005 17:34:54 +0200 Thilo wrote: > Done as you wish. Attached is the patch, it should apply cleanly > against the most current revision. > The header qcommon/qfiles.h contains the new md4 (mdr) structures that > support the mdr file format. It is also there, where I have placed > "#define RAVENMD4", so by undefining this all mdr support will be > turned off. I applied the patch, albeit with RAVENMD4 undefined by default. I'm not sure I see the value of having MDR support without an interface into cgame. Trouble is, creating such an interface breaks compatibility with baseq3, and I believe that is something that is to be avoided at the moment. Such an interface could be similarly #ifdefed to this patch though; I don't think anybody would object to that. From tim at ngus.net Fri Sep 23 13:17:16 2005 From: tim at ngus.net (Tim Angus) Date: Fri, 23 Sep 2005 18:17:16 +0100 Subject: MingW32... In-Reply-To: <20050922145153.10387703.tim@ngus.net> References: <4332323D.8080608@clutteredmind.org> <20050922145153.10387703.tim@ngus.net> Message-ID: <20050923181716.2ee53717.tim@ngus.net> On Thu, 22 Sep 2005 14:51:53 +0100 Tim wrote: > Hmm, good idea. I dicked about with __attribute__s a little when I was > trying to figure out what was preventing it from compiling/running, > but it didn't occur to me to use them this way. So I hacked about with this for a while, but couldn't get it to work. If you feel like having a go, feel free :). Instead I have this slightly cleaner than before approach: #ifdef __MINGW32__ // _ is prepended to compiled symbols #define CMANG(sym) "_"#sym #else #define CMANG(sym) #sym #endif ... void AsmCall( void ) { asm( CMANG(doAsmCall) ": \n\t" \ " movl (%%edi),%%eax \n\t" \ " subl $4,%%edi \n\t" \ > > There are a couple of these around the codebase: > > > > #if !( (defined __linux__ || __FreeBSD__ || __MINGW32__) && (defined > > __i386__) && (!defined C_ONLY)) // > > > > Can we replace all this nonsense with either a check for __GNUC__? > > This is only going to get uglier with each new platform, and be > > error prone for the ones that are missed. > > That seems sensible. Done, mostly. From tim at ngus.net Fri Sep 23 13:23:09 2005 From: tim at ngus.net (Tim Angus) Date: Fri, 23 Sep 2005 18:23:09 +0100 Subject: r99 - in trunk/code: client game qcommon renderer unix win32 In-Reply-To: <20050923055351.GA18662@spacepants.org> References: <20050923055351.GA18662@spacepants.org> Message-ID: <20050923182309.542dcf92.tim@ngus.net> On Fri, 23 Sep 2005 15:53:51 +1000 Jamie wrote: > >* Replaced lots of __linux__ || __FreeBSD__ with __GNUC__ > >* MinGW port now uses asm versions of SnapVector and ftol > >* Improved the handling of mangled symbols in vm_x86.c > > I hate to be so anal, especially as I'm not contributing any patches > myself (so bring on the flames if I'm out of line), but can you break > each of these into separate commits in the future? (i.e. so one > commit == one feature/fix) > > It's easier to read the patches if they carry a single meaning, and if > there's a problem it's easier to rollback. I appreciate your concern, but you must realise that sometimes it's difficult to do things sequentially. The above changes all really come under the banner of "Portability improvements". Would you have objected if I had used that as the change log instead? I do try to work atomically, but sometimes it's not convenient. From arny at ats.s.bawue.de Fri Sep 23 15:01:34 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 23 Sep 2005 21:01:34 +0200 Subject: [quake3] Re: MD4 Model Support In-Reply-To: <20050923181342.0dcafb3f.tim@ngus.net> References: <200509162326.51025.arny@ats.s.bawue.de> <200509201735.02404.arny@ats.s.bawue.de> <20050923181342.0dcafb3f.tim@ngus.net> Message-ID: <200509232101.36269.arny@ats.s.bawue.de> On Friday 23 September 2005 19:13, Tim Angus wrote: > I applied the patch, albeit with RAVENMD4 undefined by default. I'm not > sure I see the value of having MDR support without an interface into > cgame. Trouble is, creating such an interface breaks compatibility with > baseq3, and I believe that is something that is to be avoided at the > moment. Such an interface could be similarly #ifdefed to this patch > though; I don't think anybody would object to that. Actually, no interface change is needed at all. When calling RE_RegisterModel that function will check whether the suffix is ".mdr" and will then try to load the model as an mdr model. The only thing that needs to be adapted slightly is the cgame to support new kinds of models. Doing animations is also entity based (frame xxx is now to be "played") and does not need any modification either. So enabling it by default will not hurt compatibility at all - but it may still be desirable to have it disabled by default, so you don't get any argument from me. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From tim at ngus.net Fri Sep 23 16:52:59 2005 From: tim at ngus.net (Tim Angus) Date: Fri, 23 Sep 2005 21:52:59 +0100 Subject: MD4 Model Support In-Reply-To: <200509232101.36269.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <200509201735.02404.arny@ats.s.bawue.de> <20050923181342.0dcafb3f.tim@ngus.net> <200509232101.36269.arny@ats.s.bawue.de> Message-ID: <20050923215259.76a9e02a.tim@ngus.net> On Fri, 23 Sep 2005 21:01:34 +0200 Thilo wrote: > Actually, no interface change is needed at all. Oh OK. Nevertheless, If I wanted to get a raven player model working in ioQ3, I would need to change cg_players.c somewhat. Perhaps you could create some example code that demonstrates its use. From arny at ats.s.bawue.de Fri Sep 23 17:20:42 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 23 Sep 2005 23:20:42 +0200 Subject: [quake3] Re: MD4 Model Support In-Reply-To: <20050923215259.76a9e02a.tim@ngus.net> References: <200509162326.51025.arny@ats.s.bawue.de> <200509232101.36269.arny@ats.s.bawue.de> <20050923215259.76a9e02a.tim@ngus.net> Message-ID: <200509232320.45551.arny@ats.s.bawue.de> On Friday 23 September 2005 22:52, Tim Angus wrote: > Oh OK. Nevertheless, If I wanted to get a raven player model working in > ioQ3, I would need to change cg_players.c somewhat. Perhaps you could > create some example code that demonstrates its use. With the links I sent to Scott a few mails ago on this thread, there's also a place where you can download the original EliteForce multiplayer vms source code. The cg_players.c is contained in there and shows how their models are being used. I would refer to their source code to see how the models are properly initialized and used. I ran the original vms from EliteForce on the icculus.org quake3 engine, so I did not have to modify quake3's original vms to get models displayed. You can see the results here: http://thilo.kickchat.com/efport-progress/neelix.jpg http://thilo.kickchat.com/efport-progress/ctfworks1.jpg If you are interested, I can write up a little .txt as howto for inclusion into the repository, for using the new model system, and how it generally works, so others who still mod for quake3 can use it if they wish. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From zakk at timedoctor.org Fri Sep 23 19:32:56 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 23 Sep 2005 16:32:56 -0700 Subject: [quake3] Re: MD4 Model Support In-Reply-To: <200509232320.45551.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <200509232101.36269.arny@ats.s.bawue.de> <20050923215259.76a9e02a.tim@ngus.net> <200509232320.45551.arny@ats.s.bawue.de> Message-ID: <20050923233256.GA9074@timedoctor.org> This elite force stuff is awesome! Finally I can murder Neelix! -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From tim at ngus.net Fri Sep 23 21:52:56 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 24 Sep 2005 02:52:56 +0100 Subject: MD4 Model Support In-Reply-To: <200509232320.45551.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <200509232101.36269.arny@ats.s.bawue.de> <20050923215259.76a9e02a.tim@ngus.net> <200509232320.45551.arny@ats.s.bawue.de> Message-ID: <20050924025256.0094d7a6.tim@ngus.net> On Fri, 23 Sep 2005 23:20:42 +0200 Thilo wrote: > If you are interested, I can write up a little .txt as howto for > inclusion into the repository, for using the new model system, and > how it generally works, so others who still mod for quake3 can use it > if they wish. I think that's a good idea, otherwise it becomes like the existing MD4 code -- "Is this used for anything? Why is this code here? What does it do?" From vincent at cojot.name Mon Sep 26 12:30:58 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Mon, 26 Sep 2005 18:30:58 +0200 (CEST) Subject: Makefile problem In-Reply-To: <200509232101.36269.arny@ats.s.bawue.de> References: <200509162326.51025.arny@ats.s.bawue.de> <200509201735.02404.arny@ats.s.bawue.de> <20050923181342.0dcafb3f.tim@ngus.net> <200509232101.36269.arny@ats.s.bawue.de> Message-ID: Hi everyone, Since the last few updates, I've been getting this: [raistlin at thorbardin quake3]$ gmake gmake -C code/unix release gmake[1]: Entering directory `/tmp/quake3/code/unix' Makefile:736: *** missing separator. Stop. gmake[1]: Leaving directory `/tmp/quake3/code/unix' gmake: *** [release] Error 2 Just so you know.. Vincent From andreas at syndrom23.de Mon Sep 26 12:45:09 2005 From: andreas at syndrom23.de (Andreas Kohn) Date: Mon, 26 Sep 2005 18:45:09 +0200 Subject: [quake3] Makefile problem In-Reply-To: References: <200509162326.51025.arny@ats.s.bawue.de> <200509201735.02404.arny@ats.s.bawue.de> <20050923181342.0dcafb3f.tim@ngus.net> <200509232101.36269.arny@ats.s.bawue.de> Message-ID: <1127753110.1007.27.camel@klamath.syndrom23.de> On Mon, 2005-09-26 at 18:30 +0200, vincent at cojot.name wrote: > Hi everyone, > > Since the last few updates, I've been getting this: > > [raistlin at thorbardin quake3]$ gmake > gmake -C code/unix release > gmake[1]: Entering directory `/tmp/quake3/code/unix' > Makefile:736: *** missing separator. Stop. > gmake[1]: Leaving directory `/tmp/quake3/code/unix' > gmake: *** [release] Error 2 > > Just so you know.. Hi, does this happen with a fresh checkout? Could be that you have conflicts in code/unix/Makefile. HTH, -- Andreas -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 187 bytes Desc: This is a digitally signed message part URL: From vincent at cojot.name Mon Sep 26 12:53:02 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Mon, 26 Sep 2005 18:53:02 +0200 (CEST) Subject: [quake3] Makefile problem In-Reply-To: <1127753110.1007.27.camel@klamath.syndrom23.de> References: <200509162326.51025.arny@ats.s.bawue.de> <200509201735.02404.arny@ats.s.bawue.de> <20050923181342.0dcafb3f.tim@ngus.net> <200509232101.36269.arny@ats.s.bawue.de> <1127753110.1007.27.camel@klamath.syndrom23.de> Message-ID: On Mon, 26 Sep 2005, Andreas Kohn wrote: Hi Andreas, Nope, it doesn't. I guess I'm too new to svn and 'svn update' isn't enough to keep my source svn tree clean (I usually copy it elsewhere and work from there...). Off to read the docs... Thanks, Vincent > Hi, > > does this happen with a fresh checkout? Could be that you have conflicts > in code/unix/Makefile. > > HTH, > -- > Andreas From mail at cynapses.org Mon Sep 26 13:47:04 2005 From: mail at cynapses.org (Andreas Schneider) Date: Mon, 26 Sep 2005 19:47:04 +0200 Subject: [quake3] Makefile problem In-Reply-To: References: <200509162326.51025.arny@ats.s.bawue.de> <200509201735.02404.arny@ats.s.bawue.de> <20050923181342.0dcafb3f.tim@ngus.net> <200509232101.36269.arny@ats.s.bawue.de> <1127753110.1007.27.camel@klamath.syndrom23.de> Message-ID: <43383418.6090902@cynapses.org> I think you have modified the Makefile and now the update was not able to merge it the right way. svn revert code/unix/Makefile should do the trick :) -- andreas vincent at cojot.name wrote: > On Mon, 26 Sep 2005, Andreas Kohn wrote: > > Hi Andreas, > > Nope, it doesn't. I guess I'm too new to svn and 'svn update' isn't > enough to keep my source svn tree clean (I usually copy it elsewhere and > work from there...). Off to read the docs... > > Thanks, > > Vincent > >> Hi, >> >> does this happen with a fresh checkout? Could be that you have conflicts >> in code/unix/Makefile. >> >> HTH, >> -- >> Andreas -- http://www.cynapses.org/ - cybernetic synapses From debeuk at gmail.com Mon Sep 26 15:53:39 2005 From: debeuk at gmail.com (Sven Beukenex) Date: Mon, 26 Sep 2005 21:53:39 +0200 Subject: revision 112 doesn't compile on x86_64 Message-ID: <9a224f8b05092612535322a534@mail.gmail.com> Got this error compiling after a fresh checkout (revision 112) on Fedora Core 4 x86_64: gcc -pipe -Wall -Wimplicit -Wstrict-prototypes -Werror -DUSE_SDL=1 -I/usr/include/SDL -D_REENTRANT -DHAVE_VM_NATIVE -DNDEBUG -O3 -fomit-frame-pointer -ffast-math -falign-loops=2 -falign-jumps=2 -falign-functions=2 -fstrength-reduce -fno-strict-aliasing -o releasex86_64-glibc/client/tr_world.o -c ../renderer/tr_world.c -I/usr/X11R6/include gcc -pipe -Wall -Wimplicit -Wstrict-prototypes -Werror -DUSE_SDL=1 -I/usr/include/SDL -D_REENTRANT -DHAVE_VM_NATIVE -DNDEBUG -O3 -fomit-frame-pointer -ffast-math -falign-loops=2 -falign-jumps=2 -falign-functions=2 -fstrength-reduce -fno-strict-aliasing -o releasex86_64-glibc/client/vm_none.o -c ../qcommon/vm_none.c ../qcommon/vm_none.c:3: error: conflicting types for 'VM_CallCompiled' ../qcommon/vm_local.h:171: error: previous declaration of 'VM_CallCompiled' was here make[2]: *** [releasex86_64-glibc/client/vm_none.o] Error 1 make[2]: Leaving directory `/home/sven/quake3/code/unix' make[1]: *** [build_release] Error 2 make[1]: Leaving directory `/home/sven/quake3/code/unix' make: *** [release] Error 2 Noticed the comment "hopefully this doesn't disrupt the MSVC and/or 64bit builds". Sorry, it does. HTH /Sven -- Why are the pretty ones always insane? -- J.G. Thirlwell -------------- next part -------------- An HTML attachment was scrubbed... URL: From fatty at gsklan.de Mon Sep 26 16:09:02 2005 From: fatty at gsklan.de (fatty) Date: Mon, 26 Sep 2005 22:09:02 +0200 Subject: [quake3] OpenAL In-Reply-To: <200509071732.23646.badcdev@gmail.com> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> <200509071732.23646.badcdev@gmail.com> Message-ID: <4338555E.5010308@gsklan.de> Hello Mailinglist! Is there nobody else who wants to use that OpenAL implementation? From my point of view OpenAL rocks and regarding to the thread in QuakeSRC.org's forums (http://www.quakesrc.org/forums/viewforum.php?f=20) BlackAura has support for Win32 ready, too. btw.: if you run into problems with latest revision and the provided patch, you should alter the #endif tags from #endif !USE_OPENAL to #endif // !USE_OPENAL My GCC (3.4) did complain about some function declarations, too. So I had to change them from void function() to void function(void) but that should be anyway clear. Stephan From tim at ngus.net Mon Sep 26 16:10:17 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 26 Sep 2005 21:10:17 +0100 Subject: revision 112 doesn't compile on x86_64 In-Reply-To: <9a224f8b05092612535322a534@mail.gmail.com> References: <9a224f8b05092612535322a534@mail.gmail.com> Message-ID: <20050926211017.659e9359.tim@ngus.net> On Mon, 26 Sep 2005 21:53:39 +0200 Sven wrote: > Noticed the comment "hopefully this doesn't disrupt the MSVC and/or > 64bit builds". Sorry, it does. Try again with 114. From zakk at timedoctor.org Mon Sep 26 16:19:19 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 26 Sep 2005 13:19:19 -0700 Subject: [quake3] OpenAL In-Reply-To: <4338555E.5010308@gsklan.de> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> <200509071732.23646.badcdev@gmail.com> <4338555E.5010308@gsklan.de> Message-ID: <20050926201919.GA14644@timedoctor.org> As I told you before, I need approval from BlackAura directly, AND word that his code is complete. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From fatty at gsklan.de Mon Sep 26 17:00:56 2005 From: fatty at gsklan.de (fatty) Date: Mon, 26 Sep 2005 23:00:56 +0200 Subject: [quake3] OpenAL In-Reply-To: <20050926201919.GA14644@timedoctor.org> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> <200509071732.23646.badcdev@gmail.com> <4338555E.5010308@gsklan.de> <20050926201919.GA14644@timedoctor.org> Message-ID: <43386188.7070806@gsklan.de> Zachary J. Slater wrote: > As I told you before, > I need approval from BlackAura directly, > AND word that his code is complete. > Sorry, I didn't want to force you to do anything. I only wondered why there was so little response to that thread and recognised some errors with latest revision, so I decided to let everyone now about what I did to get it running again. Maybe my wording wasn't correct so you misunderstood my intention. Sorry again for my impetuous behavior, Stephan From zakk at timedoctor.org Mon Sep 26 18:27:17 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 26 Sep 2005 15:27:17 -0700 Subject: [quake3] OpenAL In-Reply-To: <43386188.7070806@gsklan.de> References: <200509060337.j863bn0O004209@euler.unh.edu> <003401c5b29e$767bf820$6e01a8c0@sixfour> <431E83B0.2080004@clutteredmind.org> <200509071732.23646.badcdev@gmail.com> <4338555E.5010308@gsklan.de> <20050926201919.GA14644@timedoctor.org> <43386188.7070806@gsklan.de> Message-ID: <20050926222717.GB14644@timedoctor.org> I was under the impression that BlackAura had been asked, and agreed to submit his code to us when it was ready, but that it wasn't ready yet. I feel that this project would be best served if he, or someone else, wrote an openal backend specifically against our code base, so that we don't get other patches in as well. Especially any patches derived from the questionable dusk3d stuff, etc. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From debeuk at gmail.com Mon Sep 26 19:11:43 2005 From: debeuk at gmail.com (Sven Beukenex) Date: Tue, 27 Sep 2005 01:11:43 +0200 Subject: [quake3] Re: revision 112 doesn't compile on x86_64 In-Reply-To: <20050926211017.659e9359.tim@ngus.net> References: <9a224f8b05092612535322a534@mail.gmail.com> <20050926211017.659e9359.tim@ngus.net> Message-ID: <9a224f8b05092616113379f240@mail.gmail.com> On 9/26/05, Tim Angus wrote: > > On Mon, 26 Sep 2005 21:53:39 +0200 Sven wrote: > > Noticed the comment "hopefully this doesn't disrupt the MSVC and/or > > 64bit builds". Sorry, it does. > > Try again with 114. > Yup, fixed. (it was up to 119 when I updated the tree) Thanks, Sven -- Why are the pretty ones always insane? -- J.G. Thirlwell -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Mon Sep 26 19:17:15 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 27 Sep 2005 00:17:15 +0100 Subject: revision 112 doesn't compile on x86_64 In-Reply-To: <9a224f8b05092616113379f240@mail.gmail.com> References: <9a224f8b05092612535322a534@mail.gmail.com> <20050926211017.659e9359.tim@ngus.net> <9a224f8b05092616113379f240@mail.gmail.com> Message-ID: <20050927001715.1ffb87d1.tim@ngus.net> On Tue, 27 Sep 2005 01:11:43 +0200 Sven wrote: > Yup, fixed. > (it was up to 119 when I updated the tree) Thank Ludwig. It was my inane keyboard stabbing that broke it in the first place. :) From anthaus at tutopia.com Mon Sep 26 22:40:28 2005 From: anthaus at tutopia.com (Carlos Alberto Lopez Romero) Date: Mon, 26 Sep 2005 21:40:28 -0500 Subject: Error compiling jdhuff.c Message-ID: <4338B11C.9090603@tutopia.com> Greetings. I get the following message beginning with release 112 and up to the latest revision (119). gcc -pipe -Wall -Wimplicit -Wstrict-prototypes -Werror -DUSE_SDL=1 -I/usr/include/SDL -D_REENTRANT -DHAVE_VM_NATIVE -DHAVE_VM_COMPILED -m32 -DNDEBUG -O3 -march=i686 -fomit-frame-pointer -ffast-math -falign-loops=2 -falign-jumps=2 -falign-functions=2 -fno-strict-aliasing -fstrength-reduce -o releasei386-glibc/client/jdhuff.o -c ../jpeg-6/jdhuff.c cc1: warnings being treated as errors ../jpeg-6/jdhuff.c: En la funci?n ?decode_mcu?: ../jpeg-6/jdhuff.c:443: aviso: puede ser que se utilice ?br_state.get_buffer? sin inicializar en esta funci?n ../jpeg-6/jdhuff.c:443: aviso: puede ser que se utilice ?br_state.bits_left? sin inicializar en esta funci?n make[1]: *** [releasei386-glibc/client/jdhuff.o] Error 1 make[1]: Leaving directory `~/quake3/code/unix' make: *** [build_release] Error 2 I'm sorry that some messages are in Spanish. I'll try to translate as best as I can if you feel that's necessary. Carlos Lopez Fedora Core 4 / GNU Linux x86 ______________________________ Visita http://www.tutopia.com y comienza a navegar m?s r?pido en Internet. Tutopia es Internet para todos. From michel at daenzer.net Mon Sep 26 21:56:41 2005 From: michel at daenzer.net (Michel =?ISO-8859-1?Q?D=E4nzer?=) Date: Mon, 26 Sep 2005 21:56:41 -0400 Subject: [quake3] Error compiling jdhuff.c In-Reply-To: <4338B11C.9090603@tutopia.com> References: <4338B11C.9090603@tutopia.com> Message-ID: <1127786203.3824.27.camel@localhost> On Mon, 2005-09-26 at 21:40 -0500, Carlos Alberto Lopez Romero wrote: > > gcc -pipe -Wall -Wimplicit -Wstrict-prototypes -Werror -DUSE_SDL=1 > -I/usr/include/SDL -D_REENTRANT -DHAVE_VM_NATIVE -DHAVE_VM_COMPILED -m32 > -DNDEBUG -O3 -march=i686 -fomit-frame-pointer -ffast-math > -falign-loops=2 -falign-jumps=2 -falign-functions=2 -fno-strict-aliasing > -fstrength-reduce -o releasei386-glibc/client/jdhuff.o -c ../jpeg-6/jdhuff.c > cc1: warnings being treated as errors > ../jpeg-6/jdhuff.c: En la funci?n ?decode_mcu?: > ../jpeg-6/jdhuff.c:443: aviso: puede ser que se utilice > ?br_state.get_buffer? sin inicializar en esta funci?n > ../jpeg-6/jdhuff.c:443: aviso: puede ser que se utilice > ?br_state.bits_left? sin inicializar en esta funci?n > make[1]: *** [releasei386-glibc/client/jdhuff.o] Error 1 > make[1]: Leaving directory `~/quake3/code/unix' > make: *** [build_release] Error 2 > > I'm sorry that some messages are in Spanish. I'll try to translate as > best as I can if you feel that's necessary. You could run LC_ALL=C make to get English output, but I think it says ?br_state.get_buffer? ?br_state.bits_left? may be used uninitialized in ?decode_mcu?. -- Earthling Michel D?nzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer From anthaus at tutopia.com Mon Sep 26 23:26:52 2005 From: anthaus at tutopia.com (Carlos Alberto Lopez Romero) Date: Mon, 26 Sep 2005 22:26:52 -0500 Subject: [quake3] Error compiling jdhuff.c In-Reply-To: <1127786203.3824.27.camel@localhost> References: <4338B11C.9090603@tutopia.com> <1127786203.3824.27.camel@localhost> Message-ID: <4338BBFC.7070608@tutopia.com> Michel D??nzer wrote: >On Mon, 2005-09-26 at 21:40 -0500, Carlos Alberto Lopez Romero wrote: > > >>gcc -pipe -Wall -Wimplicit -Wstrict-prototypes -Werror -DUSE_SDL=1 >>-I/usr/include/SDL -D_REENTRANT -DHAVE_VM_NATIVE -DHAVE_VM_COMPILED -m32 >>-DNDEBUG -O3 -march=i686 -fomit-frame-pointer -ffast-math >>-falign-loops=2 -falign-jumps=2 -falign-functions=2 -fno-strict-aliasing >>-fstrength-reduce -o releasei386-glibc/client/jdhuff.o -c ../jpeg-6/jdhuff.c >>cc1: warnings being treated as errors >>../jpeg-6/jdhuff.c: En la funci??n ???decode_mcu???: >>../jpeg-6/jdhuff.c:443: aviso: puede ser que se utilice >>???br_state.get_buffer??? sin inicializar en esta funci??n >>../jpeg-6/jdhuff.c:443: aviso: puede ser que se utilice >>???br_state.bits_left??? sin inicializar en esta funci??n >>make[1]: *** [releasei386-glibc/client/jdhuff.o] Error 1 >>make[1]: Leaving directory `~/quake3/code/unix' >>make: *** [build_release] Error 2 >> >>I'm sorry that some messages are in Spanish. I'll try to translate as >>best as I can if you feel that's necessary. >> >> > >You could run > >LC_ALL=C make > >to get English output, but I think it says ???br_state.get_buffer??? >???br_state.bits_left??? may be used uninitialized in ???decode_mcu???. > > > > That's correct. Thanks. ______________________________ Visita http://www.tutopia.com y comienza a navegar m?s r?pido en Internet. Tutopia es Internet para todos. From tim at ngus.net Tue Sep 27 12:37:45 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 27 Sep 2005 17:37:45 +0100 Subject: Error compiling jdhuff.c In-Reply-To: <4338B11C.9090603@tutopia.com> References: <4338B11C.9090603@tutopia.com> Message-ID: <20050927173745.1e2749e0.tim@ngus.net> On Mon, 26 Sep 2005 21:40:28 -0500 Carlos wrote: > releasei386-glibc/client/jdhuff.o -c ../jpeg-6/jdhuff.c cc1: warnings > being treated as errors I guess you are using GCC 4, which warns more than 3.x. The Makefile in revision 120 no longer treats warnings as errors, it should work. I'll install GCC 4 at some point and fix the warnings; for now it's not so important. From pomac at vapor.com Tue Sep 27 16:45:41 2005 From: pomac at vapor.com (Ian Kumlien) Date: Tue, 27 Sep 2005 22:45:41 +0200 Subject: [quake3] Re: Error compiling jdhuff.c In-Reply-To: <20050927173745.1e2749e0.tim@ngus.net> References: <4338B11C.9090603@tutopia.com> <20050927173745.1e2749e0.tim@ngus.net> Message-ID: <1127853941.10083.0.camel@localhost> On Tue, 2005-09-27 at 17:37 +0100, Tim Angus wrote: > On Mon, 26 Sep 2005 21:40:28 -0500 Carlos wrote: > > releasei386-glibc/client/jdhuff.o -c ../jpeg-6/jdhuff.c cc1: warnings > > being treated as errors > > I guess you are using GCC 4, which warns more than 3.x. The Makefile in > revision 120 no longer treats warnings as errors, it should work. I'll > install GCC 4 at some point and fix the warnings; for now it's not so > important. a 32 bit arch? I have no warnings at all on amd64... -- Ian Kumlien -- http://pomac.netswarm.net -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: From anthaus at tutopia.com Tue Sep 27 17:58:04 2005 From: anthaus at tutopia.com (Carlos Alberto Lopez Romero) Date: Tue, 27 Sep 2005 16:58:04 -0500 Subject: [quake3] Re: Error compiling jdhuff.c In-Reply-To: <1127853941.10083.0.camel@localhost> References: <4338B11C.9090603@tutopia.com> <20050927173745.1e2749e0.tim@ngus.net> <1127853941.10083.0.camel@localhost> Message-ID: <4339C06C.2030705@tutopia.com> Ian Kumlien wrote: >On Tue, 2005-09-27 at 17:37 +0100, Tim Angus wrote: > > >>On Mon, 26 Sep 2005 21:40:28 -0500 Carlos wrote: >> >> >>>releasei386-glibc/client/jdhuff.o -c ../jpeg-6/jdhuff.c cc1: warnings >>>being treated as errors >>> >>> >>I guess you are using GCC 4, which warns more than 3.x. The Makefile in >>revision 120 no longer treats warnings as errors, it should work. I'll >>install GCC 4 at some point and fix the warnings; for now it's not so >>important. >> >> > >a 32 bit arch? I have no warnings at all on amd64... > > > The error disappeared with release 120. So, maybe it was a quirk on Fedora's GCC 4. I haven't seen if the warnings are still there. ______________________________ Visita http://www.tutopia.com y comienza a navegar m?s r?pido en Internet. Tutopia es Internet para todos.