From Dietrich_Joerg at t-online.de Mon Jan 2 14:55:52 2006 From: Dietrich_Joerg at t-online.de (Joerg Dietrich) Date: Mon, 02 Jan 2006 20:55:52 +0100 Subject: Ogg Vorbis Decoder Message-ID: <43B98548.5090308@t-online.de> Hello, attached is the result of another holiday season with the family, sometimes you need to get out ;) It is a working Ogg Vorbis Decoder. I have written it by myself based on the codec framework of Stuart Dalton and of course idsoftware's work. It uses libvorbisfile, libvorbis and libogg which are under a BSD license. The Decoder uses the High Level API of vorbis : libvorbisfile. This adds a little bit of overhead but only to the loading times. While in game the streaming background music needs some CPU-time too. Only one minor addition to existing code was neccessary : The struct snd_stream_t in client/snd_codec.h now contains an additional field "length" with the filesize of the compressed soundfile. This length is obtained in S_CodecutilOpen() in client/snd_codec.c from the function FS_FOpenFileRead(). I have tested this patch only on Linux on x86_32. For this I have oggified the Demo-PAK and the programm itself : encoded all WAVs as OGGs and replaced all calls to *.wav to *.ogg . The resulting PAK is some 18 % smaller than the original. The program runs flawless, there is no stutter in the sounds or the background music. A framerate drop is not noticeable. In the Makefile I have only added support for the Ogg Vorbis Decoder for Linux. The maintainers of the other ports must define USE_CODEC_VORBIS in the CFLAGS and link with libvorbisfile, libvorbis and libogg when they want Ogg Vorbis. The whole decoder is optional. If you don't want Ogg Vorbis support, undefine USE_CODEC_VORBIS in the Makefile. Happy New Year! Joerg -------------- next part -------------- A non-text attachment was scrubbed... Name: ogg_codec.patch.gz Type: application/x-gzip Size: 3772 bytes Desc: not available URL: From rstotts at cox.net Mon Jan 2 21:40:02 2006 From: rstotts at cox.net (Ryan Stotts) Date: Mon, 2 Jan 2006 20:40:02 -0600 Subject: Who broke the code? (win32 / 354) Message-ID: <004201c6100e$ffc8e3e0$640fa8c0@amd> I was able to compile the previous version. Now this happens with the latest code and no changes to my system: $ make make -C code/unix release make[1]: Entering directory `/q3src/code/unix' make -C ../tools/lcc install make[2]: Entering directory `/q3src/code/tools/lcc' install -s -m 0755 build-mingw32-x86/q3lcc.exe ../ install -s -m 0755 build-mingw32-x86/q3cpp.exe ../ install -s -m 0755 build-mingw32-x86/q3rcc.exe ../ make[2]: Leaving directory `/q3src/code/tools/lcc' make -C ../tools/asm install make[2]: Entering directory `/q3src/code/tools/asm' install -s -m 0755 q3asm.exe ../ make[2]: Leaving directory `/q3src/code/tools/asm' make targets B=release-mingw32-x86 CFLAGS=" -Wall -fno-strict-aliasing -Wimpli it -Wstrict-prototypes -DUSE_OPENAL=1 -DUSE_OPENAL_DLOPEN=1 -m32 -DUSE_LOCAL_HE make[2]: Entering directory `/q3src/code/unix' gcc -DDEDICATED -Wall -fno-strict-aliasing -Wimplicit -Wstrict-prototypes -DUS ../unix/linux_signals.c: In function `InitSig': ../unix/linux_signals.c:54: error: `SIGHUP' undeclared (first use in this funct ../unix/linux_signals.c:54: error: (Each undeclared identifier is reported only ../unix/linux_signals.c:54: error: for each function it appears in.) ../unix/linux_signals.c:55: error: `SIGQUIT' undeclared (first use in this func ../unix/linux_signals.c:57: error: `SIGTRAP' undeclared (first use in this func ../unix/linux_signals.c:58: error: `SIGIOT' undeclared (first use in this funct ../unix/linux_signals.c:59: error: `SIGBUS' undeclared (first use in this funct make[2]: *** [release-mingw32-x86/ded/linux_signals.o] Error 1 make[2]: Leaving directory `/q3src/code/unix' make[1]: *** [build_release] Error 2 make[1]: Leaving directory `/q3src/code/unix' make: *** [release] Error 2 Thanks From marc.leeman at gmail.com Tue Jan 3 06:49:41 2006 From: marc.leeman at gmail.com (Marc Leeman) Date: Tue, 3 Jan 2006 12:49:41 +0100 Subject: [quake3] Who broke the code? (win32 / 354) In-Reply-To: <004201c6100e$ffc8e3e0$640fa8c0@amd> References: <004201c6100e$ffc8e3e0$640fa8c0@amd> Message-ID: <20060103114941.GC9154@scorpius.homelinux.org> > I was able to compile the previous version. Now this happens with the > latest code and no changes to my system: You should be able to trace this in the SVN repos :) > ../unix/linux_signals.c: In function `InitSig': If you compiling for win32, should you be compiling unix/linux files? -- greetz, marc If your hand is still there in one microt, I'll snap it off and use it as a good luck charm. Chiana - Bone To Be Wild scorpius.homelinux.org 2.6.14 #1 PREEMPT Sat Oct 29 09:45:46 CEST 2005 GNU/Linux -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From tim at ngus.net Tue Jan 3 10:15:48 2006 From: tim at ngus.net (Tim Angus) Date: Tue, 3 Jan 2006 15:15:48 +0000 Subject: quake3-1.33-svn20051104-5 In-Reply-To: <20051231130350.GH27287@scorpius.homelinux.org> References: <20051230210535.GF27287@scorpius.homelinux.org> <200512311310.55660.ludwig.nussel@suse.de> <20051231130350.GH27287@scorpius.homelinux.org> Message-ID: <20060103151548.542253c4.tim@ngus.net> On Sat, 31 Dec 2005 14:03:50 +0100 Marc wrote: > I just changed the 'default' setting to 0 instead of 1. When running a > server, an admin would have a nicely tweaked cfg file where he > enables a pure server. If the admin is just gonna turn it back on, why make the default 0? You're just asking for trouble imho. From tim at ngus.net Tue Jan 3 10:16:13 2006 From: tim at ngus.net (Tim Angus) Date: Tue, 3 Jan 2006 15:16:13 +0000 Subject: Who broke the code? (win32 / 354) In-Reply-To: <20060103114941.GC9154@scorpius.homelinux.org> References: <004201c6100e$ffc8e3e0$640fa8c0@amd> <20060103114941.GC9154@scorpius.homelinux.org> Message-ID: <20060103151613.0cfc246d.tim@ngus.net> On Tue, 3 Jan 2006 12:49:41 +0100 Marc wrote: > > I was able to compile the previous version. Now this happens with > > the latest code and no changes to my system: > > You should be able to trace this in the SVN repos :) > > > ../unix/linux_signals.c: In function `InitSig': > > If you compiling for win32, should you be compiling unix/linux files? The latest change to the Makefile broke the MinGW build. I'll fix it soon... From vincent at cojot.name Tue Jan 3 10:37:36 2006 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 3 Jan 2006 16:37:36 +0100 (CET) Subject: [quake3] Re: Who broke the code? (win32 / 354) In-Reply-To: <20060103151613.0cfc246d.tim@ngus.net> References: <004201c6100e$ffc8e3e0$640fa8c0@amd> <20060103114941.GC9154@scorpius.homelinux.org> <20060103151613.0cfc246d.tim@ngus.net> Message-ID: On Tue, 3 Jan 2006, Tim Angus wrote: > The latest change to the Makefile broke the MinGW build. I'll fix it > soon... Hello Tim, Before you change too much stuff back into the makefile, could you please check my small solaris patch too? It changes a few things in the Makefile too... Thanks, Vincent -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From tim at ngus.net Tue Jan 3 10:50:18 2006 From: tim at ngus.net (Tim Angus) Date: Tue, 3 Jan 2006 15:50:18 +0000 Subject: Who broke the code? (win32 / 354) In-Reply-To: References: <004201c6100e$ffc8e3e0$640fa8c0@amd> <20060103114941.GC9154@scorpius.homelinux.org> <20060103151613.0cfc246d.tim@ngus.net> Message-ID: <20060103155018.70817418.tim@ngus.net> On Tue, 3 Jan 2006 16:37:36 +0100 (CET) vincent at cojot.name wrote: > Before you change too much stuff back into the makefile, could you > please check my small solaris patch too? It changes a few things in > the Makefile too... Yes, it's on the list. From tim at ngus.net Fri Jan 6 12:43:17 2006 From: tim at ngus.net (Tim Angus) Date: Fri, 6 Jan 2006 17:43:17 +0000 Subject: Ogg Vorbis Decoder In-Reply-To: <43B98548.5090308@t-online.de> References: <43B98548.5090308@t-online.de> Message-ID: <20060106174317.24f571cd.tim@ngus.net> On Mon, 02 Jan 2006 20:55:52 +0100 Joerg wrote: > attached is the result of another holiday season with the family, > sometimes you need to get out ;) You may or may not be aware but BlackAura (Stuart Dalton) also wrote an ogg decoder. I put off integrating that as I wanted to ensure the abstract codec system was bug free before adding more potentially unstable code. I've given both implementations a cursory glance and yours seems the more robust, so I've committed it. It is disabled by default as it causes a support nightmare where people don't have ogg vorbis installed (and besides for playing music, it isn't of much use to standard Q3 anyway). Thanks! From icculus at icculus.org Fri Jan 13 16:24:49 2006 From: icculus at icculus.org (Ryan C. Gordon) Date: Fri, 13 Jan 2006 16:24:49 -0500 Subject: Intel Mac support patch... Message-ID: <43C81AA1.9020505@icculus.org> https://bugzilla.icculus.org/show_bug.cgi?id=2580 Game appears to be fully playable if you use SDL instead of OpenAL for audio, once you apply that patch. --ryan. From stowellt at gmail.com Fri Jan 13 17:12:48 2006 From: stowellt at gmail.com (Tim Stowell) Date: Fri, 13 Jan 2006 15:12:48 -0700 Subject: Having trouble compiling under MSVC 2005 express Message-ID: <994f7fe90601131412n78f13da2lcbc7951e982b4b42@mail.gmail.com> Hi all, I am getting linker errors about missing external symbols regarding snd_dma.c amongst other files. Am I missing a linker dependency? Thanks, I can get the code to compile with minGW, but would like to use VC if possible. -Tim -------------- next part -------------- An HTML attachment was scrubbed... URL: From icculus at icculus.org Sat Jan 14 04:48:43 2006 From: icculus at icculus.org (Ryan C. Gordon) Date: Sat, 14 Jan 2006 04:48:43 -0500 Subject: vm vs. compiled vs. native... Message-ID: <43C8C8FB.40207@icculus.org> Can someone please verify that the game .dll (or .so, etc) is actually working? If I run the game with: ./ioquake3.whatever +set vm_ui 0 Then on the main menu, running the "vminfo" console command will report that the UI module is "native" but if I then load the single player game, "vminfo" says it's interpreted. I just saw this on the Intel Mac, but I recall this happening on Linux, too. Am I doing something wrong, or did we break (re)loading of native libraries? --ryan. From marc.leeman at gmail.com Sat Jan 14 05:20:15 2006 From: marc.leeman at gmail.com (Marc Leeman) Date: Sat, 14 Jan 2006 11:20:15 +0100 Subject: [quake3] vm vs. compiled vs. native... In-Reply-To: <43C8C8FB.40207@icculus.org> References: <43C8C8FB.40207@icculus.org> Message-ID: <20060114102015.GW9300@scorpius.homelinux.org> > Can someone please verify that the game .dll (or .so, etc) is actually > working? If I run the game with: > > ./ioquake3.whatever +set vm_ui 0 > > Then on the main menu, running the "vminfo" console command will report > that the UI module is "native" but if I then load the single player > game, "vminfo" says it's interpreted. I just saw this on the Intel Mac, > but I recall this happening on Linux, too. Am I doing something wrong, > or did we break (re)loading of native libraries? Started tty console (use +set ttycon 0 to disable) ]\vminfo Registered virtual machines: ui : native ... ]\vminfo Registered virtual machines: qagame : compiled on load code length : 1137840 table length: 586688 data length : 2097152 ui : compiled on load code length : 594408 table length: 319976 data length : 1048576 cgame : compiled on load code length : 786818 table length: 399612 data length : 4194304 Crash^7 entered the game -- greetz, marc Repent? We have less than an arn. I was a Dominar. It'd take me longer than that to repent. Rygel - Suns and Lovers scorpius.homelinux.org 2.6.15 #1 PREEMPT Thu Jan 5 22:01:16 CET 2006 GNU/Linux -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From auerswal at unix-ag.uni-kl.de Sat Jan 14 05:41:05 2006 From: auerswal at unix-ag.uni-kl.de (Erik Auerswald) Date: Sat, 14 Jan 2006 11:41:05 +0100 Subject: [quake3] vm vs. compiled vs. native... In-Reply-To: <43C8C8FB.40207@icculus.org> References: <43C8C8FB.40207@icculus.org> Message-ID: <20060114104105.GA773@sushi.unix-ag.uni-kl.de> Hi, On Sat, Jan 14, 2006 at 04:48:43AM -0500, Ryan C. Gordon wrote: > Can someone please verify that the game .dll (or .so, etc) is actually > working? If I run the game with: > > ./ioquake3.whatever +set vm_ui 0 > > Then on the main menu, running the "vminfo" console command will report > that the UI module is "native" but if I then load the single player > game, "vminfo" says it's interpreted. I just saw this on the Intel Mac, > but I recall this happening on Linux, too. Am I doing something wrong, > or did we break (re)loading of native libraries? Just checked this on linux x86: In the main menu the native ui is used (can be easily checked by verifying that there is a "Very High Quality" video settings profile), but after starting single player the vm ui is used. The "vminfo" command reports it that way as well. Erik From tim at ngus.net Sat Jan 14 10:06:12 2006 From: tim at ngus.net (Tim Angus) Date: Sat, 14 Jan 2006 15:06:12 +0000 Subject: vm vs. compiled vs. native... In-Reply-To: <43C8C8FB.40207@icculus.org> References: <43C8C8FB.40207@icculus.org> Message-ID: <20060114150612.5119b479.tim@ngus.net> On Sat, 14 Jan 2006 04:48:43 -0500 Ryan wrote: > Can someone please verify that the game .dll (or .so, etc) is > actually working? If I run the game with: > > ./ioquake3.whatever +set vm_ui 0 > > Then on the main menu, running the "vminfo" console command will > report that the UI module is "native" but if I then load the single > player game, "vminfo" says it's interpreted. I just saw this on the > Intel Mac, but I recall this happening on Linux, too. Am I doing > something wrong, or did we break (re)loading of native libraries? You'll need "+set sv_pure 0" on the command line too, or the file system layer will ignore discrete files (such as native libraries). From marc.leeman at gmail.com Sat Jan 14 10:22:52 2006 From: marc.leeman at gmail.com (Marc Leeman) Date: Sat, 14 Jan 2006 16:22:52 +0100 Subject: [quake3] Re: vm vs. compiled vs. native... In-Reply-To: <20060114150612.5119b479.tim@ngus.net> References: <43C8C8FB.40207@icculus.org> <20060114150612.5119b479.tim@ngus.net> Message-ID: <20060114152252.GY9300@scorpius.homelinux.org> > You'll need "+set sv_pure 0" on the command line too, or the file > system layer will ignore discrete files (such as native libraries). ]\vminfo Registered virtual machines: ui : native ... ]\vminfo Registered virtual machines: qagame : compiled on load code length : 1137840 table length: 586688 data length : 2097152 ui : native cgame : compiled on load code length : 786818 table length: 399612 data length : 4194304 ]\quit -- greetz, marc You made me drink piss? Jool - Different Destinations scorpius.homelinux.org 2.6.15 #1 PREEMPT Thu Jan 5 22:01:16 CET 2006 GNU/Linux -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From stowellt at gmail.com Sat Jan 14 12:45:19 2006 From: stowellt at gmail.com (Tim Stowell) Date: Sat, 14 Jan 2006 10:45:19 -0700 Subject: Can't get code to compile in VC Express Message-ID: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> There seems to be something wrong with the sound linkage, do I need to have something set in my linker dependencies for this to work? 1>cl_main.obj : error LNK2019: unresolved external symbol _CL_CloseAVI referenced in function _CL_Disconnect 1>cl_main.obj : error LNK2019: unresolved external symbol _CL_VideoRecording referenced in function _CL_Disconnect 1>snd_dma.obj : error LNK2001: unresolved external symbol _CL_VideoRecording 1>snd_mix.obj : error LNK2001: unresolved external symbol _CL_VideoRecording 1>cl_main.obj : error LNK2019: unresolved external symbol _CL_TakeVideoFrame referenced in function _CL_Frame 1>cl_main.obj : error LNK2019: unresolved external symbol _CL_WriteAVIVideoFrame referenced in function _CL_InitRef 1>cl_main.obj : error LNK2019: unresolved external symbol _CL_OpenAVIForWriting referenced in function _CL_Video_f 1>snd_mix.obj : error LNK2019: unresolved external symbol _CL_WriteAVIAudioFrame referenced in function _S_TransferStereo16 -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Sat Jan 14 13:13:45 2006 From: tim at ngus.net (Tim Angus) Date: Sat, 14 Jan 2006 18:13:45 +0000 Subject: Can't get code to compile in VC Express In-Reply-To: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> References: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> Message-ID: <20060114181345.43ba1ab6.tim@ngus.net> On Sat, 14 Jan 2006 10:45:19 -0700 Tim wrote: > There seems to be something wrong with the sound linkage, do I need > to have something set in my linker dependencies for this to work? Try adding cl_avi.c to the project. Nobody seems interested in maintaining the MSVC build I'm afraid. It's best to use MinGW really, as it gets much more testing and the tool chain is more familiar to the developers. From stowellt at gmail.com Sat Jan 14 13:44:40 2006 From: stowellt at gmail.com (Tim Stowell) Date: Sat, 14 Jan 2006 11:44:40 -0700 Subject: [quake3] Re: Can't get code to compile in VC Express In-Reply-To: <20060114181345.43ba1ab6.tim@ngus.net> References: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> <20060114181345.43ba1ab6.tim@ngus.net> Message-ID: <994f7fe90601141044p18c25143j29f12210135ab46c@mail.gmail.com> Thanks for the tip, that did it (have been stuck on this one) Curious, I have done research on this but can't seem to get a straight answer. On win32, does VC yield more optimized code, or is that just a rumor? I would be more than happy to use MinGW if there is no real disadvantage, thanks. On 1/14/06, Tim Angus wrote: > > On Sat, 14 Jan 2006 10:45:19 -0700 Tim wrote: > > There seems to be something wrong with the sound linkage, do I need > > to have something set in my linker dependencies for this to work? > > Try adding cl_avi.c to the project. Nobody seems interested in > maintaining the MSVC build I'm afraid. It's best to use MinGW really, > as it gets much more testing and the tool chain is more familiar to the > developers. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Sat Jan 14 14:18:08 2006 From: tim at ngus.net (Tim Angus) Date: Sat, 14 Jan 2006 19:18:08 +0000 Subject: Can't get code to compile in VC Express In-Reply-To: <994f7fe90601141044p18c25143j29f12210135ab46c@mail.gmail.com> References: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> <20060114181345.43ba1ab6.tim@ngus.net> <994f7fe90601141044p18c25143j29f12210135ab46c@mail.gmail.com> Message-ID: <20060114191808.093c6fc0.tim@ngus.net> On Sat, 14 Jan 2006 11:44:40 -0700 Tim wrote: > On win32, does VC yield more optimized code, or is > that just a rumor? If may ever so slightly, because VC is targetted only at x86 and thus can afford to take shortcuts that wouldn't be acceptable in gcc. The code that is compiled in the MinGW and MSVC builds is for all intents and purposes identical, so there is no performance advantage there. > I would be more than happy to use MinGW if there > is no real disadvantage, thanks. Based purely on my own anecdotal testing of 1.32b versus ioq3 compiled with MinGW, there is no measurable difference in terms of framerate, so I don't perceive there to be a disadvantage. From stowellt at gmail.com Sat Jan 14 14:26:06 2006 From: stowellt at gmail.com (Tim Stowell) Date: Sat, 14 Jan 2006 12:26:06 -0700 Subject: [quake3] Re: Can't get code to compile in VC Express In-Reply-To: <20060114191808.093c6fc0.tim@ngus.net> References: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> <20060114181345.43ba1ab6.tim@ngus.net> <994f7fe90601141044p18c25143j29f12210135ab46c@mail.gmail.com> <20060114191808.093c6fc0.tim@ngus.net> Message-ID: <994f7fe90601141126o6cf7b8ffj9ae06a6c2d1cbd60@mail.gmail.com> I really appreciate the info, that clears up a lot of confusion. It would seem from what I have gleaned by searching all over the web that the Icculus build seems to be the most current and active Quake 3 project (unless I'm wrong?), so I think it would benefit me to switch to MinGW to use it. On 1/14/06, Tim Angus wrote: > > On Sat, 14 Jan 2006 11:44:40 -0700 Tim wrote: > > On win32, does VC yield more optimized code, or is > > that just a rumor? > > If may ever so slightly, because VC is targetted only at x86 and thus > can afford to take shortcuts that wouldn't be acceptable in gcc. The > code that is compiled in the MinGW and MSVC builds is for all intents > and purposes identical, so there is no performance advantage there. > > > I would be more than happy to use MinGW if there > > is no real disadvantage, thanks. > > Based purely on my own anecdotal testing of 1.32b versus ioq3 compiled > with MinGW, there is no measurable difference in terms of framerate, so > I don't perceive there to be a disadvantage. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From scott at scottbrooks.ca Sat Jan 14 14:58:53 2006 From: scott at scottbrooks.ca (Scott Brooks) Date: Sat, 14 Jan 2006 12:58:53 -0700 Subject: GCC 4's autovectorization Message-ID: I was wondering if anyone has any experience with gcc4's auto vectorization, and how much of an improvement it has over 3.4, or VC. Thanks From icculus at icculus.org Sat Jan 14 17:18:48 2006 From: icculus at icculus.org (Ryan C. Gordon) Date: Sat, 14 Jan 2006 17:18:48 -0500 Subject: [quake3] Re: Can't get code to compile in VC Express In-Reply-To: <20060114191808.093c6fc0.tim@ngus.net> References: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> <20060114181345.43ba1ab6.tim@ngus.net> <994f7fe90601141044p18c25143j29f12210135ab46c@mail.gmail.com> <20060114191808.093c6fc0.tim@ngus.net> Message-ID: <43C978C8.7030804@icculus.org> > If may ever so slightly, because VC is targetted only at x86 and thus > can afford to take shortcuts that wouldn't be acceptable in gcc. The > code that is compiled in the MinGW and MSVC builds is for all intents > and purposes identical, so there is no performance advantage there. It's generally believed that MSVC generates significantly better code than gcc, but for something like quake3, the difference is probably unnoticable (or at least, will net you say, 150 fps instead of 140 or something). --ryan. From tim at ngus.net Sat Jan 14 17:28:13 2006 From: tim at ngus.net (Tim Angus) Date: Sat, 14 Jan 2006 22:28:13 +0000 Subject: Can't get code to compile in VC Express In-Reply-To: <43C978C8.7030804@icculus.org> References: <994f7fe90601140945n671b2fbev56d7c9d6f8ee2ae7@mail.gmail.com> <20060114181345.43ba1ab6.tim@ngus.net> <994f7fe90601141044p18c25143j29f12210135ab46c@mail.gmail.com> <20060114191808.093c6fc0.tim@ngus.net> <43C978C8.7030804@icculus.org> Message-ID: <20060114222813.50c7ec9e.tim@ngus.net> On Sat, 14 Jan 2006 17:18:48 -0500 Ryan wrote: > It's generally believed that MSVC generates significantly better code > than gcc, I would believe that with gcc 2.7, which is about when I first heard that claim, but I'm kind of dubious this is still true. > but for something like quake3, the difference is probably > unnoticable (or at least, will net you say, 150 fps instead of 140 or > something). Well, like I say, my own experiments indicate nothing like as big a margin as that. No detectable margin at all in fact. Q3 is a pretty crap comparison though, as it's hardware accelerated and will spend a lot of time in the GL library. Using Q2 with the software renderer would make a lot more sense. From stowellt at gmail.com Sat Jan 14 17:59:26 2006 From: stowellt at gmail.com (Tim Stowell) Date: Sat, 14 Jan 2006 15:59:26 -0700 Subject: Programming reference resources Message-ID: <994f7fe90601141459j1aec1392h51ac5c434eb2eb17@mail.gmail.com> I am new to the Quake 3 programming world and as such have been scouring the web for many hours trying to get some references to the code. I realize it has not all been documented, and that most people probably just go in and reverse-engineer the best they can with the comments in the code, but I wondered what resources others here have found to help them? I think it would be great to have a master documentation reference to this end. I have found the following links: http://www.dxquake3.dsl.pipex.com/dxquake3_links.htm (which includes many useful links) http://www.planetquake.com/code3arena/ (Has some good articles, although it may be a bit dated) Would anyone else care to add in a link or two? I am new at this, so please forgive my ignorance if this is common knowledge -------------- next part -------------- An HTML attachment was scrubbed... URL: From lmsmercykiller at gmail.com Sat Jan 14 19:55:20 2006 From: lmsmercykiller at gmail.com (Stephen Youts) Date: Sat, 14 Jan 2006 18:55:20 -0600 Subject: [quake3] Programming reference resources In-Reply-To: <994f7fe90601141459j1aec1392h51ac5c434eb2eb17@mail.gmail.com> References: <994f7fe90601141459j1aec1392h51ac5c434eb2eb17@mail.gmail.com> Message-ID: I remember seeing some site based in germany that is essentially a codemap, showing relationship diagrams and such. However, I don't have a link handy. Maybe someone knows what I'm talking about and has a link. On 1/14/06, Tim Stowell wrote: > > I am new to the Quake 3 programming world and as such have been scouring > the web for many hours trying to get some references to the code. I realize > it has not all been documented, and that most people probably just go in and > reverse-engineer the best they can with the comments in the code, but I > wondered what resources others here have found to help them? I think it > would be great to have a master documentation reference to this end. I have > found the following links: > > http://www.dxquake3.dsl.pipex.com/dxquake3_links.htm (which includes many > useful links) > http://www.planetquake.com/code3arena/ (Has some good articles, although > it may be a bit dated) > > Would anyone else care to add in a link or two? I am new at this, so > please forgive my ignorance if this is common knowledge > > > -- -MercyKiller -------------- next part -------------- An HTML attachment was scrubbed... URL: From kell at spoonix.com Sat Jan 14 21:26:43 2006 From: kell at spoonix.com (K. Spoon) Date: Sat, 14 Jan 2006 20:26:43 -0600 Subject: [quake3] Programming reference resources In-Reply-To: References: <994f7fe90601141459j1aec1392h51ac5c434eb2eb17@mail.gmail.com> Message-ID: <20060115022642.GA12981@inverness.spoonix.com> On Sat, Jan 14, 2006 at 06:55:20PM -0600, Stephen Youts wrote: > I remember seeing some site based in germany that is essentially a codemap, > showing relationship diagrams and such. However, I don't have a link handy. > Maybe someone knows what I'm talking about and has a link. Hrm. Haven't seen it and I can't find the right incantation to make google cough up a link. :( But, in case someone else might find it useful, I've got the output of a run through doxygen up at: http://www.spoonix.org/downloads/ioq3-doxygen/ > On 1/14/06, Tim Stowell wrote: > > > > I am new to the Quake 3 programming world and as such have been scouring > > the web for many hours trying to get some references to the code. I realize > > it has not all been documented, and that most people probably just go in and > > reverse-engineer the best they can with the comments in the code, but I > > wondered what resources others here have found to help them? I think it > > would be great to have a master documentation reference to this end. I have > > found the following links: > > > > http://www.dxquake3.dsl.pipex.com/dxquake3_links.htm (which includes many > > useful links) > > http://www.planetquake.com/code3arena/ (Has some good articles, although > > it may be a bit dated) > > > > Would anyone else care to add in a link or two? I am new at this, so > > please forgive my ignorance if this is common knowledge > > > > > > > > > -- > -MercyKiller -- K. Spoon From rstotts at cox.net Sat Jan 14 22:34:53 2006 From: rstotts at cox.net (Ryan Stotts) Date: Sat, 14 Jan 2006 21:34:53 -0600 Subject: [quake3] Programming reference resources References: <994f7fe90601141459j1aec1392h51ac5c434eb2eb17@mail.gmail.com> Message-ID: <000e01c61984$a5ef2ad0$640fa8c0@amd> >I remember seeing some site based in germany that is essentially a codemap, >showing relationship diagrams and such. However, I don't have a link handy. >Maybe someone knows what I'm talking about and has a link. Was it this? http://www.nanobit.net/quake3_doxy/ Here is one for Quake 2... http://submarine.org.uk/ref_dx/ref_gl_big.png This is interesting too.. http://www.quakesrc.org/forums/viewtopic.php?t=4551 -------------- next part -------------- An HTML attachment was scrubbed... URL: From lmsmercykiller at gmail.com Sat Jan 14 22:39:40 2006 From: lmsmercykiller at gmail.com (Stephen Youts) Date: Sat, 14 Jan 2006 21:39:40 -0600 Subject: [quake3] Programming reference resources In-Reply-To: <000e01c61984$a5ef2ad0$640fa8c0@amd> References: <994f7fe90601141459j1aec1392h51ac5c434eb2eb17@mail.gmail.com> <000e01c61984$a5ef2ad0$640fa8c0@amd> Message-ID: Isn't that, but it is something very similar. I managed to dig it up on google with "EV_Fire_Weapon", go figure. http://soclose.de/q3doc/index.htm On 1/14/06, Ryan Stotts wrote: > Was it this? > > http://www.nanobit.net/quake3_doxy/ > -- -MercyKiller -------------- next part -------------- An HTML attachment was scrubbed... URL: From rbottoms at gmail.com Sun Jan 15 14:06:35 2006 From: rbottoms at gmail.com (Richard Bottoms) Date: Sun, 15 Jan 2006 11:06:35 -0800 Subject: Fade to Black Message-ID: <958e99670601151106g9d472b1u64f37a15dd621447@mail.gmail.com> Anyone have a function that lets the server tell the client to fade to black at some specified time out? I am working on an RPG like game and I want to have more cinematic-like transitions. In this case I'd like to modify 'TeleportPlayer' to fade to black and up from black .It's really part of a need for players to be able to Teleport from one map to the next. After fade out, I want the server to tell the client to execute a function that disconnects it from "server:port a" then reconnects at "server:port b" The blackout hides the flicker that occurs sometimes when changing maps and it looks cooler too. r.b. -------------- next part -------------- An HTML attachment was scrubbed... URL: From rstotts at cox.net Sun Jan 15 14:22:10 2006 From: rstotts at cox.net (Ryan Stotts) Date: Sun, 15 Jan 2006 13:22:10 -0600 Subject: Now compiles really fast! Message-ID: <000901c61a08$fb8cb140$640fa8c0@amd> It used to take upwards of ~30 seconds to compile on my machine with a lot of message spew. Now it compiles in like ~5 seconds and in one large paragraph. Thanks! -------- Revision: 483 Author: icculus Date: 3:16:16 AM, Saturday, January 14, 2006 Message: Fixed compiler warning about uninitialized variables. ---- Modified : /trunk/code/tools/lcc/cpp/eval.c From auerswal at unix-ag.uni-kl.de Sun Jan 15 23:53:08 2006 From: auerswal at unix-ag.uni-kl.de (Erik Auerswald) Date: Mon, 16 Jan 2006 05:53:08 +0100 Subject: [PATCH] additional vidModes Message-ID: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> Hi, attached patch adds some video modes to quake3. Most of these modes are copied from icculus.org quake2, the 1280x960 mode is my preferred 4:3 video mode on 19" CRTs. Erik -------------- next part -------------- Index: code/q3_ui/ui_video.c =================================================================== --- code/q3_ui/ui_video.c (revision 495) +++ code/q3_ui/ui_video.c (working copy) @@ -775,10 +775,20 @@ "960x720", "1024x768", "1152x864", - "1280x1024", + "1280x1024", /* 5:4 aspect ratio */ "1600x1200", "2048x1536", "856x480 wide screen", + "1280x960", /* 4:3 aspect ratio */ + "1024x480", /* sony vaio pocketbook */ + "1152x768", /* Apple TiBook */ + "1280x854", /* Apple TiBook */ + "640x400", /* generic 16:10 widescreen resolutions */ + "800x500", /* as found on many modern notebooks */ + "1024x640", + "1280x800", + "1680x1050", + "1920x1200", NULL }; static const char *filter_names[] = Index: code/renderer/tr_init.c =================================================================== --- code/renderer/tr_init.c (revision 495) +++ code/renderer/tr_init.c (working copy) @@ -295,7 +295,17 @@ { "Mode 8: 1280x1024", 1280, 1024, 1 }, { "Mode 9: 1600x1200", 1600, 1200, 1 }, { "Mode 10: 2048x1536", 2048, 1536, 1 }, - { "Mode 11: 856x480 (wide)",856, 480, 1 } + { "Mode 11: 856x480 (wide)", 856, 480, 1 }, + { "Mode 12: 1280x960", 1280, 960, 1 }, + { "Mode 13: 1024x480", 1024, 480, 1 }, + { "Mode 14: 1152x768", 1152, 768, 1 }, + { "Mode 15: 1280x854", 1280, 854, 1 }, + { "Mode 16: 640x400", 640, 400, 1 }, + { "Mode 17: 800x500", 800, 500, 1 }, + { "Mode 18: 1024x640", 1024, 640, 1 }, + { "Mode 19: 1280x800", 1280, 800, 1 }, + { "Mode 20: 1680x1050", 1680, 1050, 1 }, + { "Mode 21: 1920x1200", 1920, 1200, 1 } }; static int s_numVidModes = ( sizeof( r_vidModes ) / sizeof( r_vidModes[0] ) ); From Dietrich_Joerg at t-online.de Mon Jan 16 12:51:03 2006 From: Dietrich_Joerg at t-online.de (Joerg Dietrich) Date: Mon, 16 Jan 2006 18:51:03 +0100 Subject: [quake3] [PATCH] additional vidModes In-Reply-To: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> Message-ID: <43CBDD07.4050304@t-online.de> While we are at it, may I propose a 640x512 mode. It would allow 1280x1024 LCD-owners with slow PCs to run Quake 3 in quarter resolution without ugly scaling effects. Joerg Erik Auerswald wrote: > Hi, > > attached patch adds some video modes to quake3. Most of these modes are > copied from icculus.org quake2, the 1280x960 mode is my preferred 4:3 > video mode on 19" CRTs. > > Erik From gofish at evilbill.org Mon Jan 16 12:39:27 2006 From: gofish at evilbill.org (Bill Adams) Date: Mon, 16 Jan 2006 09:39:27 -0800 (PST) Subject: [quake3] [PATCH] additional vidModes In-Reply-To: <43CBDD07.4050304@t-online.de> References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> Message-ID: <3088.12.168.226.4.1137433167.squirrel@ghod.com> And 1280x720 for my 16:9 projector. Perhaps it would be worth it to implement general custom video modes? Not that I have the time or knowhow for such a feat. b. > While we are at it, may I propose a 640x512 mode. > It would allow 1280x1024 LCD-owners with slow PCs > to run Quake 3 in quarter resolution without ugly > scaling effects. > > Joerg > > > Erik Auerswald wrote: >> Hi, >> >> attached patch adds some video modes to quake3. Most of these modes are >> copied from icculus.org quake2, the 1280x960 mode is my preferred 4:3 >> video mode on 19" CRTs. >> >> Erik > From davebytes at comcast.net Mon Jan 16 14:26:06 2006 From: davebytes at comcast.net (David Chait) Date: Mon, 16 Jan 2006 14:26:06 -0500 Subject: [quake3] [PATCH] additional vidModes References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> <3088.12.168.226.4.1137433167.squirrel@ghod.com> Message-ID: <004001c61ad2$b3281280$6e01a8c0@sixfour> As I was involved in developing widescreen mode support for a racing game, thought I'd throw in what formats I know of, in case someone just wants to encode them. Note that for many of them, the FOV needs tweaking and can't/shouldn't be direct-calculated from the pixel aspect ratio... And of course, assumes the monitor & card have proper driver and/or timings support for the wacky modes. ;) new wide lcd formats 1680x1050 1920x1200 2560x1600 wide lcd + laptop (there's a TON of these, this is just a sampling...) 1200x800 1280x854 1400x900 1440x900 1440x960 standard LCD TV/etc formats: 852x480 1280x720 1280x768 1366x768 1920x1080 wide plasmas 1024x768 1024x1024 triple-head support: 1920x480 (640x3) 2400x600 (800x3) 3072x768 (1024x3) -d ----- Original Message ----- From: "Bill Adams" To: Sent: Monday, January 16, 2006 12:39 PM Subject: Re: [quake3] [PATCH] additional vidModes | And 1280x720 for my 16:9 projector. | | Perhaps it would be worth it to implement general custom video modes? Not | that I have the time or knowhow for such a feat. | From ben at zygoat.ca Mon Jan 16 14:29:10 2006 From: ben at zygoat.ca (Ben Kennedy) Date: Mon, 16 Jan 2006 14:29:10 -0500 Subject: [quake3] [PATCH] additional vidModes In-Reply-To: <004001c61ad2$b3281280$6e01a8c0@sixfour> References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> <3088.12.168.226.4.1137433167.squirrel@ghod.com> <004001c61ad2$b3281280$6e01a8c0@sixfour> Message-ID: <20060116192910.5288@mail.zygoat.ca> If there are more than about 5 modes in a circulating list, it will be infuriating to use. The UI will have to be improved, to at minimum provide next/previous buttons, or even better, a list or pop-up. -b David Chait wrote at 2:26 PM (-0500) on 1/16/06: >As I was involved in developing widescreen mode support for a racing game, >thought I'd throw in what formats I know of, in case someone just wants to >encode them. Note that for many of them, the FOV needs tweaking and >can't/shouldn't be direct-calculated from the pixel aspect ratio... > >And of course, assumes the monitor & card have proper driver and/or timings >support for the wacky modes. ;) > >new wide lcd formats >1680x1050 >1920x1200 >2560x1600 > >wide lcd + laptop (there's a TON of these, this is just a sampling...) >1200x800 >1280x854 >1400x900 >1440x900 >1440x960 > >standard LCD TV/etc formats: >852x480 >1280x720 >1280x768 >1366x768 >1920x1080 > >wide plasmas >1024x768 >1024x1024 > >triple-head support: >1920x480 (640x3) >2400x600 (800x3) >3072x768 (1024x3) > >-d > >----- Original Message ----- >From: "Bill Adams" >To: >Sent: Monday, January 16, 2006 12:39 PM >Subject: Re: [quake3] [PATCH] additional vidModes > > >| And 1280x720 for my 16:9 projector. >| >| Perhaps it would be worth it to implement general custom video modes? Not >| that I have the time or knowhow for such a feat. >| > -- Ben Kennedy, chief magician zygoat creative technical services 613-228-3392 | 1-866-466-4628 http://www.zygoat.ca From zakk at timedoctor.org Mon Jan 16 14:32:10 2006 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 16 Jan 2006 11:32:10 -0800 Subject: [quake3] [PATCH] additional vidModes In-Reply-To: <3088.12.168.226.4.1137433167.squirrel@ghod.com> References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> <3088.12.168.226.4.1137433167.squirrel@ghod.com> Message-ID: <8175E90B-CFFC-45CD-8ACB-6014773607F2@timedoctor.org> In general if a mode isn't supported in-game, you can edit your config file and set r_customheight/width and r_mode -1 Though I agree that we need at least some of these extra modes added. I am a bit concerned though about two bugs I'm seeing with random unusual modes, for instance at 1680x1050, what my widescreen LCD runs, full-screen cinematics aren't playing. I am also unsure if the widescreen video modes actually give you more vision to the left and right. I don't think they do. - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From tfogal at cisunix.unh.edu Mon Jan 16 16:40:49 2006 From: tfogal at cisunix.unh.edu (tom fogal) Date: Mon, 16 Jan 2006 16:40:49 -0500 Subject: [quake3] [PATCH] additional vidModes In-Reply-To: Your message of "Mon, 16 Jan 2006 14:29:10 EST." <20060116192910.5288@mail.zygoat.ca> Message-ID: <200601162140.k0GLent0032328@euler.unh.edu> <20060116192910.5288 at mail.zygoat.ca>"Ben Kennedy" writes: > If there are more than about 5 modes in a circulating list, it will be > infuriating to use. The UI will have to be improved, to at minimum > provide next/previous buttons, or even better, a list or pop-up. For the long term, couldn't the list be just the modes supported by the graphics system setup? Apparently one can query SDL for what modes a given device supports [1]. Presumably any one setup will not support *too* many modes, so the existing UI might suffice. Maybe. -tom [1] http://www.devolution.com/pipermail/sdl/2004-October/065446.html > >As I was involved in developing widescreen mode support for a racing game, > >thought I'd throw in what formats I know of, in case someone just wants to > >encode them. Note that for many of them, the FOV needs tweaking and > >can't/shouldn't be direct-calculated from the pixel aspect ratio... " * A: Because it reverses the logical flow of conversation. * Q: Why is top posting frowned upon? " From tim at ngus.net Mon Jan 16 16:54:48 2006 From: tim at ngus.net (Tim Angus) Date: Mon, 16 Jan 2006 21:54:48 +0000 Subject: additional vidModes In-Reply-To: <200601162140.k0GLent0032328@euler.unh.edu> References: <20060116192910.5288@mail.zygoat.ca> <200601162140.k0GLent0032328@euler.unh.edu> Message-ID: <20060116215448.62c62770.tim@ngus.net> On Mon, 16 Jan 2006 16:40:49 -0500 tom wrote: > For the long term, couldn't the list be just the modes supported by > the graphics system setup? Apparently one can query SDL for what > modes a given device supports [1]. This would certainly be my prefered solution. You only see the resolutions appropriate to your system, which can set using the perfectly adequate existing custom resolution code. Trying to statically support every conceivable resolution available (or even a subset of them) is pretty dumb IMO, for the reasons already covered. > " > * A: Because it reverses the logical flow of conversation. > * Q: Why is top posting frowned upon? > " My own thoughts: http://a.ngus.net/topposting.html ;) From davebytes at comcast.net Mon Jan 16 19:02:04 2006 From: davebytes at comcast.net (David Chait) Date: Mon, 16 Jan 2006 19:02:04 -0500 Subject: [quake3] [PATCH] additional vidModes References: <200601162140.k0GLent0032328@euler.unh.edu> Message-ID: <007101c61af9$403df920$6e01a8c0@sixfour> My windows display control panel is currently showing support for more than a dozen video modes, FYI (a 1680x1050 native display...). I'm sure if you're under linux and hand crafted your modelines, you likely only have a few -- but using EDID information, my Radeon X8xx + Dell 20FPW can direct detect and support 'more than about 5'... -d ----- Original Message ----- From: "tom fogal" Sent: Monday, January 16, 2006 4:40 PM | <20060116192910.5288 at mail.zygoat.ca>"Ben Kennedy" writes: | > If there are more than about 5 modes in a circulating list, it will be | > infuriating to use. The UI will have to be improved, to at minimum | > provide next/previous buttons, or even better, a list or pop-up. | | For the long term, couldn't the list be just the modes supported by the | graphics system setup? Apparently one can query SDL for what modes a | given device supports [1]. | | Presumably any one setup will not support *too* many modes, so the | existing UI might suffice. Maybe. | | -tom From davebytes at comcast.net Mon Jan 16 19:07:57 2006 From: davebytes at comcast.net (David Chait) Date: Mon, 16 Jan 2006 19:07:57 -0500 Subject: [quake3] [PATCH] additional vidModes References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> <3088.12.168.226.4.1137433167.squirrel@ghod.com> <8175E90B-CFFC-45CD-8ACB-6014773607F2@timedoctor.org> Message-ID: <007501c61afa$12a57fa0$6e01a8c0@sixfour> It would be good if it could at least autodetect the desktop display setting, and make sure that is in the list (add it if not!). That'd be a decent 'starting' solution. If the system isn't autodetecting the proper aspect ratio based on the aspect resolution being non-4x3, it'll need to be enhanced... ;) Again, the tricky thing is that some displays aren't square pixels, but I wouldn't worry about them to start with (could add something later to compensate). I'm sure there's a FOV setting in the config somewhere, but given that we've got 16:9, 16:10, 15:9, and wacky intermediates, best to compute a relative FOV it if the mode is obviously non-4:3 ratio. -d ----- Original Message ----- From: "Zachary J. Slater" Sent: Monday, January 16, 2006 2:32 PM | In general if a mode isn't supported in-game, you can edit your | config file and set | r_customheight/width and r_mode -1 | | Though I agree that we need at least some of these extra modes added. | | I am a bit concerned though about two bugs I'm seeing with random | unusual modes, | for instance at 1680x1050, what my widescreen LCD runs, full-screen | cinematics aren't | playing. | | I am also unsure if the widescreen video modes actually give you more | vision to the left | and right. I don't think they do. From auerswal at unix-ag.uni-kl.de Tue Jan 17 03:43:54 2006 From: auerswal at unix-ag.uni-kl.de (Erik Auerswald) Date: Tue, 17 Jan 2006 09:43:54 +0100 Subject: [quake3] [PATCH] additional vidModes In-Reply-To: <8175E90B-CFFC-45CD-8ACB-6014773607F2@timedoctor.org> References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> <3088.12.168.226.4.1137433167.squirrel@ghod.com> <8175E90B-CFFC-45CD-8ACB-6014773607F2@timedoctor.org> Message-ID: <20060117084354.GA32703@sushi.unix-ag.uni-kl.de> Hi, On Mon, Jan 16, 2006 at 11:32:10AM -0800, Zachary J. Slater wrote: > I am also unsure if the widescreen video modes actually give you more > vision to the left > and right. I don't think they do. Widescreen modes reduce the up and down vision (just tested). For any given cg_fov value the horizontal field of view stays the same for every resolution. Erik From cournia at gmail.com Tue Jan 17 17:17:42 2006 From: cournia at gmail.com (Nathan Cournia) Date: Tue, 17 Jan 2006 17:17:42 -0500 Subject: BUG? Used Zone Memory Message-ID: <6e036f0e0601171417h20663438s2053ddee390c425b@mail.gmail.com> Hi, I just started looking though the Q3A source and think I may have found a possible bug. In Z_ClearZone, zone->used is set to 0 while zone->size is set to size (the number of bytes malloc'd from the OS). It seems to me that zone->used should initially be set to sizeof(memzone_t), as the first part of the memory zone is dedicated to the memzone_t struct. BTW I don't think that setting zone->used to 0 has any serious ill effects, other than Z_AvailableZoneMemory will be sizeof(memzone_t) bytes off. Any thoughts? nathan From tim at ngus.net Tue Jan 17 20:25:40 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 01:25:40 +0000 Subject: Legal stuff Message-ID: <20060118012540.5e94cf5a.tim@ngus.net> I just checked in a GPL MD4 implementation derived from the Linux kernel. I'm not 100% sure it's 64bit and big-endian safe, so some testing on x86_64 and ppc would be good. The advertising clause in the Berkeley license is not GPL compatible. Fortunately there is a precedent for simple removing it: ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change So I suggest we do just that.. unless anybody disagrees? From jaq at spacepants.org Tue Jan 17 22:41:01 2006 From: jaq at spacepants.org (Jamie Wilkinson) Date: Wed, 18 Jan 2006 14:41:01 +1100 Subject: [quake3] Legal stuff In-Reply-To: <20060118012540.5e94cf5a.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> Message-ID: <20060118034101.GB4684@spacepants.org> This one time, at band camp, Tim Angus wrote: >I just checked in a GPL MD4 implementation derived from the Linux >kernel. I'm not 100% sure it's 64bit and big-endian safe, so some >testing on x86_64 and ppc would be good. I dunno about x86_64, but the mdfour.c in quakeforge and quake2forge is big-endian-safe, GPLed derived from Samba. From zakk at timedoctor.org Tue Jan 17 23:49:20 2006 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 17 Jan 2006 20:49:20 -0800 Subject: [quake3] Legal stuff In-Reply-To: <20060118012540.5e94cf5a.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> Message-ID: <4E861C69-8597-47C9-87B7-6B77662FF46B@timedoctor.org> As long as your best judgement says this is OK, and we can revert back using SVN, I'm for it. So do it unless Ryan or Angst voice any serious objections, which we can hopefully mc hammer out. - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From fuddl at gmx.de Wed Jan 18 02:17:23 2006 From: fuddl at gmx.de (Bruno Kleinert) Date: Wed, 18 Jan 2006 08:17:23 +0100 Subject: [quake3] Legal stuff In-Reply-To: <20060118012540.5e94cf5a.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> Message-ID: <20060118081723.450e74f5@localhost.localdomain> On Wed, 18 Jan 2006 01:25:40 +0000 Tim Angus wrote: > I just checked in a GPL MD4 implementation derived from the Linux > kernel. I'm not 100% sure it's 64bit and big-endian safe, so some > testing on x86_64 and ppc would be good. > > The advertising clause in the Berkeley license is not GPL compatible. > Fortunately there is a precedent for simple removing it: > > ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change > > So I suggest we do just that.. unless anybody disagrees? > I would appreciate it very much if one would take time to clean up the license issues. ATM that's the only reason why I couldn't get Ubuntu dapper packages through the Ubuntu reviewing process. From auerswal at unix-ag.uni-kl.de Wed Jan 18 03:10:01 2006 From: auerswal at unix-ag.uni-kl.de (Erik Auerswald) Date: Wed, 18 Jan 2006 09:10:01 +0100 Subject: [quake3] [PATCH] additional vidModes In-Reply-To: <20060116192910.5288@mail.zygoat.ca> References: <20060116045308.GA26028@sushi.unix-ag.uni-kl.de> <43CBDD07.4050304@t-online.de> <3088.12.168.226.4.1137433167.squirrel@ghod.com> <004001c61ad2$b3281280$6e01a8c0@sixfour> <20060116192910.5288@mail.zygoat.ca> Message-ID: <20060118081001.GA15886@sushi.unix-ag.uni-kl.de> Hi, On Mon, Jan 16, 2006 at 02:29:10PM -0500, Ben Kennedy wrote: > If there are more than about 5 modes in a circulating list, it will be > infuriating to use. The UI will have to be improved, to at minimum > provide next/previous buttons, or even better, a list or pop-up. The next/previous functionality is already there, you have to use the cursor keys. Erik From auerswal at unix-ag.uni-kl.de Wed Jan 18 03:14:06 2006 From: auerswal at unix-ag.uni-kl.de (Erik Auerswald) Date: Wed, 18 Jan 2006 09:14:06 +0100 Subject: [quake3] Re: additional vidModes In-Reply-To: <20060116215448.62c62770.tim@ngus.net> References: <20060116192910.5288@mail.zygoat.ca> <200601162140.k0GLent0032328@euler.unh.edu> <20060116215448.62c62770.tim@ngus.net> Message-ID: <20060118081406.GB15886@sushi.unix-ag.uni-kl.de> Hi, On Mon, Jan 16, 2006 at 09:54:48PM +0000, Tim Angus wrote: > On Mon, 16 Jan 2006 16:40:49 -0500 tom wrote: > > For the long term, couldn't the list be just the modes supported by > > the graphics system setup? Apparently one can query SDL for what > > modes a given device supports [1]. > > This would certainly be my prefered solution. You only see the > resolutions appropriate to your system, which can set using the > perfectly adequate existing custom resolution code. I'd like this solution as well. AFAIK this has to wait until SDL is used on every plattform. Until then I'm working on a patch to set a custom resolution via the UI. Erik From zakk at timedoctor.org Wed Jan 18 03:17:23 2006 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 18 Jan 2006 00:17:23 -0800 Subject: [quake3] Legal stuff In-Reply-To: <20060118081723.450e74f5@localhost.localdomain> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118081723.450e74f5@localhost.localdomain> Message-ID: <2F781851-EB23-492F-B41F-08E45C4E6CD8@timedoctor.org> On Jan 17, 2006, at 11:17 PM, Bruno Kleinert wrote: > > I would appreciate it very much if one would take time to clean up the > license issues. ATM that's the only reason why I couldn't get Ubuntu > dapper packages through the Ubuntu reviewing process. > While I realize that debian might have problems with licenses, we do not have a lawyer, nor can we afford one. Anyone who expects open-source projects to be magically vetted by the lawyer-faeries is bound to be sorely disappointed. Not that I don't want to see icculus.org/quake3 running on every distribution, just that we can't ever be 100% sure of any legal status without lawyers. I'd like to see some kind of process appear for doing this (making sure licenses are complied with), but that would be a costly task. - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From ludwig.nussel at suse.de Wed Jan 18 03:51:53 2006 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Wed, 18 Jan 2006 09:51:53 +0100 Subject: [quake3] Legal stuff In-Reply-To: <2F781851-EB23-492F-B41F-08E45C4E6CD8@timedoctor.org> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118081723.450e74f5@localhost.localdomain> <2F781851-EB23-492F-B41F-08E45C4E6CD8@timedoctor.org> Message-ID: <200601180951.53724.ludwig.nussel@suse.de> On Wednesday 18 January 2006 09:17, Zachary J. Slater wrote: > On Jan 17, 2006, at 11:17 PM, Bruno Kleinert wrote: > > > > I would appreciate it very much if one would take time to clean up the > > license issues. ATM that's the only reason why I couldn't get Ubuntu > > dapper packages through the Ubuntu reviewing process. > > > > While I realize that debian might have problems with licenses, > we do not have a lawyer, nor can we afford one. Anyone > who expects open-source projects to be magically vetted by the > lawyer-faeries is bound to be sorely disappointed. > > Not that I don't want to see icculus.org/quake3 running on every > distribution, > just that we can't ever be 100% sure of any legal status without > lawyers. > > I'd like to see some kind of process appear for doing this (making sure > licenses are complied with), but that would > be a costly task. Our build system does some automatic grepping to find offending licenses. It found that the included jpeg code has an advertising clause. lcc has some free of charge constraints. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Wed Jan 18 08:59:18 2006 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 18 Jan 2006 14:59:18 +0100 (CET) Subject: [quake3] Legal stuff In-Reply-To: <20060118012540.5e94cf5a.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> Message-ID: On Wed, 18 Jan 2006, Tim Angus wrote: > I just checked in a GPL MD4 implementation derived from the Linux > kernel. I'm not 100% sure it's 64bit and big-endian safe, so some > testing on x86_64 and ppc would be good. I don't know if that's related but on Solaris/Sparc I now get this: ************************************************** WARNING: pak0.pk3 is present but its checksum (1600873053) is not correct. Please re-copy pak0.pk3 from your legitimate Q3 CDROM. ************************************************** It started happening with this morning's build (rev 503)... Cmd line was: build/release-SunOS-sparc/ioq3ded.sparc +set fs_cdpath \ /usr/local/share/games/quake3 +set vm_game 1 +set vm_cgame 1 \ +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 The prior build I tested was just before the md4 changes.. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, 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 tim at ngus.net Wed Jan 18 09:45:31 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 14:45:31 +0000 Subject: Legal stuff In-Reply-To: References: <20060118012540.5e94cf5a.tim@ngus.net> Message-ID: <20060118144531.3603910e.tim@ngus.net> On Wed, 18 Jan 2006 14:59:18 +0100 (CET) vincent at cojot.name wrote: > ************************************************** > WARNING: pak0.pk3 is present but its checksum (1600873053) > is not correct. Please re-copy pak0.pk3 from your > legitimate Q3 CDROM. > ************************************************** > > It started happening with this morning's build (rev 503)... > > Cmd line was: > > build/release-SunOS-sparc/ioq3ded.sparc +set fs_cdpath \ > /usr/local/share/games/quake3 +set vm_game 1 +set vm_cgame 1 \ > +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 > > The prior build I tested was just before the md4 changes.. Well, bollocks. The most likely is related. From tim at ngus.net Wed Jan 18 09:47:50 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 14:47:50 +0000 Subject: Legal stuff In-Reply-To: <200601180951.53724.ludwig.nussel@suse.de> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118081723.450e74f5@localhost.localdomain> <2F781851-EB23-492F-B41F-08E45C4E6CD8@timedoctor.org> <200601180951.53724.ludwig.nussel@suse.de> Message-ID: <20060118144750.69166107.tim@ngus.net> On Wed, 18 Jan 2006 09:51:53 +0100 Ludwig wrote: > Our build system does some automatic grepping to find offending > licenses. It found that the included jpeg code has an advertising > clause. lcc has some free of charge constraints. I think lcc is OK because the lcc binary is not being distributed. In the worse case the lcc source can be stripped from any source packages. I didn't notice the jpeg advertising clause... From tim at ngus.net Wed Jan 18 09:57:21 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 14:57:21 +0000 Subject: Legal stuff In-Reply-To: <20060118034101.GB4684@spacepants.org> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118034101.GB4684@spacepants.org> Message-ID: <20060118145721.379be9be.tim@ngus.net> On Wed, 18 Jan 2006 14:41:01 +1100 Jamie wrote: > This one time, at band camp, Tim Angus wrote: > >I just checked in a GPL MD4 implementation derived from the Linux > >kernel. I'm not 100% sure it's 64bit and big-endian safe, so some > >testing on x86_64 and ppc would be good. > > I dunno about x86_64, but the mdfour.c in quakeforge and quake2forge > is big-endian-safe, GPLed derived from Samba. ...But missing functionality needed for Q3. The mdfour_update function doesn't work like the RSA implementation. From tim at ngus.net Wed Jan 18 10:54:34 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 15:54:34 +0000 Subject: Legal stuff In-Reply-To: <20060118145721.379be9be.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118034101.GB4684@spacepants.org> <20060118145721.379be9be.tim@ngus.net> Message-ID: <20060118155434.74a42d1f.tim@ngus.net> On Wed, 18 Jan 2006 14:57:21 +0000 Tim wrote: > > I dunno about x86_64, but the mdfour.c in quakeforge and quake2forge > > is big-endian-safe, GPLed derived from Samba. > > ...But missing functionality needed for Q3. The mdfour_update function > doesn't work like the RSA implementation. I've worked around this restriction though, hopefully this works better on big-endian machines. From tim at ngus.net Wed Jan 18 10:54:52 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 15:54:52 +0000 Subject: Legal stuff In-Reply-To: <20060118144531.3603910e.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118144531.3603910e.tim@ngus.net> Message-ID: <20060118155452.24ba0aa1.tim@ngus.net> On Wed, 18 Jan 2006 14:45:31 +0000 Tim wrote: > On Wed, 18 Jan 2006 14:59:18 +0100 (CET) vincent at cojot.name wrote: > > ************************************************** > > WARNING: pak0.pk3 is present but its checksum (1600873053) > > is not correct. Please re-copy pak0.pk3 from your > > legitimate Q3 CDROM. > > ************************************************** > > > > It started happening with this morning's build (rev 503)... > > > > Cmd line was: > > > > build/release-SunOS-sparc/ioq3ded.sparc +set fs_cdpath \ > > /usr/local/share/games/quake3 +set vm_game 1 +set vm_cgame 1 \ > > +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 > > > > The prior build I tested was just before the md4 changes.. > > Well, bollocks. The most likely is related. Give 504 a shot. From vincent at cojot.name Wed Jan 18 11:19:35 2006 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 18 Jan 2006 17:19:35 +0100 (CET) Subject: [quake3] Re: Legal stuff In-Reply-To: <20060118155452.24ba0aa1.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118144531.3603910e.tim@ngus.net> <20060118155452.24ba0aa1.tim@ngus.net> Message-ID: On Wed, 18 Jan 2006, Tim Angus wrote: > On Wed, 18 Jan 2006 14:45:31 +0000 Tim wrote: >> On Wed, 18 Jan 2006 14:59:18 +0100 (CET) vincent at cojot.name wrote: >>> ************************************************** >>> WARNING: pak0.pk3 is present but its checksum (1600873053) >>> is not correct. Please re-copy pak0.pk3 from your >>> legitimate Q3 CDROM. >>> ************************************************** >>> >>> It started happening with this morning's build (rev 503)... >> Well, bollocks. The most likely is related. > > Give 504 a shot. Yup, 504 works fine here. Thanks Tim! I tested with q3ded and with the client as well.. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, 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 tim at ngus.net Wed Jan 18 13:48:49 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 18 Jan 2006 18:48:49 +0000 Subject: Legal stuff In-Reply-To: <200601180951.53724.ludwig.nussel@suse.de> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118081723.450e74f5@localhost.localdomain> <2F781851-EB23-492F-B41F-08E45C4E6CD8@timedoctor.org> <200601180951.53724.ludwig.nussel@suse.de> Message-ID: <20060118184849.4992caf2.tim@ngus.net> On Wed, 18 Jan 2006 09:51:53 +0100 Ludwig wrote: > It found that the included jpeg code has an advertising > clause. > (2) If only executable code is distributed, then the accompanying > documentation must state that "this software is based in part on the > work of the Independent JPEG Group". As far as I can see this isn't a problem as it only applies when not distributing source... and we are. From fuddl at gmx.de Wed Jan 18 17:37:26 2006 From: fuddl at gmx.de (Bruno Kleinert) Date: Wed, 18 Jan 2006 23:37:26 +0100 Subject: [quake3] Re: additional vidModes In-Reply-To: <20060118081406.GB15886@sushi.unix-ag.uni-kl.de> References: <20060116192910.5288@mail.zygoat.ca> <200601162140.k0GLent0032328@euler.unh.edu> <20060116215448.62c62770.tim@ngus.net> <20060118081406.GB15886@sushi.unix-ag.uni-kl.de> Message-ID: <20060118233726.7f715b9c@localhost.localdomain> On Wed, 18 Jan 2006 09:14:06 +0100 Erik Auerswald wrote: > Hi, > > On Mon, Jan 16, 2006 at 09:54:48PM +0000, Tim Angus wrote: > > On Mon, 16 Jan 2006 16:40:49 -0500 tom wrote: > > > For the long term, couldn't the list be just the modes supported by > > > the graphics system setup? Apparently one can query SDL for what > > > modes a given device supports [1]. > > > > This would certainly be my prefered solution. You only see the > > resolutions appropriate to your system, which can set using the > > perfectly adequate existing custom resolution code. > > I'd like this solution as well. AFAIK this has to wait until SDL is used > on every plattform. Until then I'm working on a patch to set a custom > resolution via the UI. > > Erik > would it even be possible, in windowed mode, to change the resolution by resizing the window? i can imagine this might be impossible because of that vid_restart q3 command... From jaq at spacepants.org Wed Jan 18 22:19:19 2006 From: jaq at spacepants.org (Jamie Wilkinson) Date: Thu, 19 Jan 2006 14:19:19 +1100 Subject: [quake3] Re: Legal stuff In-Reply-To: <20060118145721.379be9be.tim@ngus.net> References: <20060118012540.5e94cf5a.tim@ngus.net> <20060118034101.GB4684@spacepants.org> <20060118145721.379be9be.tim@ngus.net> Message-ID: <20060119031919.GC19590@spacepants.org> This one time, at band camp, Tim Angus wrote: >On Wed, 18 Jan 2006 14:41:01 +1100 Jamie wrote: >> This one time, at band camp, Tim Angus wrote: >> >I just checked in a GPL MD4 implementation derived from the Linux >> >kernel. I'm not 100% sure it's 64bit and big-endian safe, so some >> >testing on x86_64 and ppc would be good. >> >> I dunno about x86_64, but the mdfour.c in quakeforge and quake2forge >> is big-endian-safe, GPLed derived from Samba. > >...But missing functionality needed for Q3. The mdfour_update function >doesn't work like the RSA implementation. Ah, d'oh. From icculus at icculus.org Thu Jan 19 04:48:42 2006 From: icculus at icculus.org (Ryan C. Gordon) Date: Thu, 19 Jan 2006 04:48:42 -0500 Subject: native DLLs and R_GetShaderByHandle Message-ID: <43CF607A.2090704@icculus.org> Ok, ignore my issues with qvm vs. native loading...you DO need to disable sv_pure, even in single player. Thanks, Tim. However, if I use the native .DLLs on the Intel Mac, I get this when running a timedemo: ^3R_GetShaderByHandle: out of range hShader '276' ^3R_GetShaderByHandle: out of range hShader '283' ^3R_GetShaderByHandle: out of range hShader '286' ^3R_GetShaderByHandle: out of range hShader '290' ^3R_GetShaderByHandle: out of range hShader '295' ...these seem to match up with the avatar icons on the loading screen, but in-game rendering seems fine. Ran game like this: ./ioquake3 +set vm_ui 0 +set vm_game 0 +set vm_cgame 0 +set sv_pure 0 Can someone test this on another platform? I want to file this in Bugzilla, but want to see if it's exclusively an Intel Mac (gcc4?) issue first. --ryan. From tim at ngus.net Thu Jan 19 11:45:03 2006 From: tim at ngus.net (Tim Angus) Date: Thu, 19 Jan 2006 16:45:03 +0000 Subject: native DLLs and R_GetShaderByHandle In-Reply-To: <43CF607A.2090704@icculus.org> References: <43CF607A.2090704@icculus.org> Message-ID: <20060119164503.0f8e3d2f.tim@ngus.net> On Thu, 19 Jan 2006 04:48:42 -0500 Ryan wrote: > Can someone test this on another platform? I want to file this in > Bugzilla, but want to see if it's exclusively an Intel Mac (gcc4?) > issue first. I'm not seeing anything like that here. From jamlacey at gmail.com Tue Jan 24 19:00:11 2006 From: jamlacey at gmail.com (James Lacey) Date: Tue, 24 Jan 2006 17:00:11 -0700 Subject: working on Win64 port Message-ID: <000b01c62142$5062d3d0$6d00000a@interactiontech.local> I just wanted to send a heads-up to the mailing list in case someone else is working on this so we don't duplicate any effort. I noticed on the status page that an x86_64 build is only supported on Linux. I just built a new computer with an Athlon 64 X2 processor so I thought it would be a fun project in my free time to try to get quake3 to build on MSVC 8 for x86_64. Since this is a spare-time project, I don't have an estimate for completion but I did already run into a nasty ICE when building for x86_64 so it looks like the task will be a little harder than I initially thought. Later, James From zakk at timedoctor.org Tue Jan 24 20:04:21 2006 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 24 Jan 2006 17:04:21 -0800 Subject: Binaries!!!11111 Message-ID: <43D6CE95.6040500@timedoctor.org> This is getting ridiculous, we really need to put out binaries. Is anyone seriously knowledgeable about the nullsoft installer (besides Timbo), AND can go through the mingw stuff to get a playable windows binary, then the installer. Ryan's promised me an OS X installer and binaries. I'd like a sample version of that, but if he is too busy can anyone else do that? Linux binaries are already taken care of by various people doing rpms and installers. Thanks to Angst especially for knocking the installer thing out. Mainly though I would like to see a working installer for Windows and OS X sooner rather than later, OR to know about (on this list) what patches or whatever would hold up Windows and OS X. - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From jamlacey at gmail.com Tue Jan 24 20:26:21 2006 From: jamlacey at gmail.com (James Lacey) Date: Tue, 24 Jan 2006 18:26:21 -0700 Subject: [quake3] Binaries!!!11111 In-Reply-To: <43D6CE95.6040500@timedoctor.org> Message-ID: <000c01c6214e$5a593800$6d00000a@interactiontech.local> I packaged an installer using NSIS for a UT2004 mod project I was on. I am not an expert with NSIS by any stretch of the imagination (we use InstallShield at my work), but I could probably write the script if no one with more expertise wants to tackle it. As far as an OS X installer, I don't really think an "installer" per-se is necessary. Quake 3 installed to a folder in OS X (as opposed to an application bundle) so using MojoPatch would be overkill. It easiest thing to provide would probably be a .dmg with the binary that the user would then copy into his Quake 3 folder. James -----Original Message----- From: Zachary J. Slater [mailto:zakk at timedoctor.org] Sent: Tuesday, January 24, 2006 6:04 PM To: quake3 at icculus.org Subject: [quake3] Binaries!!!11111 This is getting ridiculous, we really need to put out binaries. Is anyone seriously knowledgeable about the nullsoft installer (besides Timbo), AND can go through the mingw stuff to get a playable windows binary, then the installer. Ryan's promised me an OS X installer and binaries. I'd like a sample version of that, but if he is too busy can anyone else do that? Linux binaries are already taken care of by various people doing rpms and installers. Thanks to Angst especially for knocking the installer thing out. Mainly though I would like to see a working installer for Windows and OS X sooner rather than later, OR to know about (on this list) what patches or whatever would hold up Windows and OS X. - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From marc.leeman at gmail.com Wed Jan 25 16:13:02 2006 From: marc.leeman at gmail.com (Marc Leeman) Date: Wed, 25 Jan 2006 22:13:02 +0100 Subject: [quake3] Binaries!!!11111 In-Reply-To: <000c01c6214e$5a593800$6d00000a@interactiontech.local> References: <43D6CE95.6040500@timedoctor.org> <000c01c6214e$5a593800$6d00000a@interactiontech.local> Message-ID: <20060125211302.GJ18000@scorpius.homelinux.org> > This is getting ridiculous, we really need to put out binaries. Just provide a tag and we can sync the binaries. I'll take care of the deb packages. -- greetz, marc This will work, trust me. We just have to find a place to get our clothes off. Crichton - Liars, Guns and Money - A Not So Simple Plan scorpius.homelinux.org 2.6.15 #1 PREEMPT Thu Jan 5 22:01:16 CET 2006 GNU/Linux -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From rstotts at cox.net Wed Jan 25 17:40:14 2006 From: rstotts at cox.net (Ryan Stotts) Date: Wed, 25 Jan 2006 16:40:14 -0600 Subject: Misc console messages Message-ID: <000501c62200$4f3d4e40$640fa8c0@amd> Just browsing through it and noticed some things... ...detecting CPU, found generic (really is a Athlon64 3200+ (2ghz) http://valid.x86-secret.com/show_oc?id=66788 ) Joystick is not active. (is there anyway I can disable the joystick? I thought there used to be a command line option? +set in_joystick 0 ?) ...WGL_3DFX_gamma_control not found (can I disable that in the .cfg?) Failed to load library: "OpenAL32.dll". (is it supposed to be like that? I have working sound...) From tim at ngus.net Wed Jan 25 18:40:18 2006 From: tim at ngus.net (Tim Angus) Date: Wed, 25 Jan 2006 23:40:18 +0000 Subject: Misc console messages In-Reply-To: <000501c62200$4f3d4e40$640fa8c0@amd> References: <000501c62200$4f3d4e40$640fa8c0@amd> Message-ID: <20060125234018.17b51d5b.tim@ngus.net> On Wed, 25 Jan 2006 16:40:14 -0600 Ryan wrote: > Just browsing through it and noticed some things... I assume this is Windows, compiled with MinGW? > ...detecting CPU, found generic (really is a Athlon64 3200+ (2ghz) > http://valid.x86-secret.com/show_oc?id=66788 ) The CPU detection code uses a load of intel syntax inline asm. It's a waste of time porting it since Q3 does no (runtime) CPU specific optimisation. > Joystick is not active. (is there anyway I can disable the joystick? I > thought there used to be a command line option? +set in_joystick 0 ?) Uh.. maybe. It's just a message :). > ...WGL_3DFX_gamma_control not found (can I disable that in the .cfg?) Probably not, but again, it's just informational.. it's doing no harm. > Failed to load library: "OpenAL32.dll". (is it supposed to be like > that? I have working sound...) That means OpenAL failed to load, which implies you don't have it installed, or (for whatever reason) your OpenAL lib is inaccesible to Q3. If OpenAL fails to load ioq3 falls back on the base sound system. You can see which sound system you're using via the value of s_backend and/or the output of \s_info. From spamhans at yahoo.de Wed Jan 25 21:15:22 2006 From: spamhans at yahoo.de (Per Hansen) Date: Thu, 26 Jan 2006 03:15:22 +0100 Subject: [lists] [quake3] Binaries!!!11111 In-Reply-To: <43D6CE95.6040500@timedoctor.org> References: <43D6CE95.6040500@timedoctor.org> Message-ID: <43D830BA.2040207@yahoo.de> On windows the best installer would be a *.msi file! http://www.microsoft.com/downloads/details.aspx?FamilyId=A55B6B43-E24F-4EA3-A93E-40C0EC4F68E5&displaylang=en http://wix.sourceforge.net/ http://izfree.sourceforge.net/ http://msi-repackaging.sourceforge.net/ http://www.aksdb.org/ http://www.qwerty-msi.com/ http://www.advancedinstaller.com/ http://users.cyberone.com.au/dbareis/intro.htm On GNU/Linux I would like to see a "debian" dir included in the sourcepackage. http://people.debian.org/~mpalmer/debian-mentors_FAQ.html#firstpackage http://www.infodrom.org/Debian/doc/maint/Maintenance-pkgbuild.html -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 481 bytes Desc: OpenPGP digital signature URL: From rstotts at cox.net Wed Jan 25 21:29:43 2006 From: rstotts at cox.net (Ryan Stotts) Date: Wed, 25 Jan 2006 20:29:43 -0600 Subject: [quake3] Re: Misc console messages References: <000501c62200$4f3d4e40$640fa8c0@amd> <20060125234018.17b51d5b.tim@ngus.net> Message-ID: <000e01c62220$5e5a1e60$640fa8c0@amd> > I assume this is Windows, compiled with MinGW? Yes it is, I should have mentioned that. > That means OpenAL failed to load, which implies you don't have it > installed, It was indeed because I didn't install anything "OpenAL" related. I just downloaded this and installed it: http://www.openal.org/openal_webstf/downloads/OpenALwEAX.exe http://www.openal.org/ Check out the before and after results: Before: ============================== ------ Initializing Sound ------ Loading "OpenAL32.dll"... Failed to load library: "OpenAL32.dll". Initializing DirectSound locked hardware. ok ----- Sound Info ----- 1 stereo 32768 samples 16 samplebits 1 submission_chunk 22050 speed 0xcc98060 dma buffer No background file. ---------------------- Sound intialization successful. -------------------------------- Sound memory manager started ]\s_backend "s_backend" is:"base" default:"" ]\s_info ----- Sound Info ----- 1 stereo 32768 samples 16 samplebits 1 submission_chunk 22050 speed 0xcc98060 dma buffer No background file. ============================== After: ============================== ------ Initializing Sound ------ Loading "OpenAL32.dll"... Allocated 30 sources. OpenAL info: Vendor: Creative Labs Inc. Version: 1.1 Renderer: Software Extensions: EAX EAX2.0 EAX3.0 EAX4.0 EAX5.0 EAX3.0EMULATED EAX4.0EMULATED AL_EXT_OFFSET AL_EXT_LINEAR_DISTANCE AL_EXT_EXPONENT_DISTANCE Sound intialization successful. -------------------------------- ]\s_info OpenAL info: Vendor: Creative Labs Inc. Version: 1.1 Renderer: Software Extensions: EAX EAX2.0 EAX3.0 EAX4.0 EAX5.0 EAX3.0EMULATED EAX4.0EMULATED AL_EXT_OFFSET AL_EXT_LINEAR_DISTANCE AL_EXT_EXPONENT_DISTANCE ]\s_backend "s_backend" is:"OpenAL" default:"" ============================== I gained about one or two fps in the timedemo using OpenAL vs's the default... From zakk at timedoctor.org Wed Jan 25 21:43:34 2006 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 25 Jan 2006 18:43:34 -0800 Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: <43D830BA.2040207@yahoo.de> References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> Message-ID: <43D83756.80603@timedoctor.org> Per Hansen wrote: > On GNU/Linux I would like to see a "debian" dir included in the > sourcepackage. > > http://people.debian.org/~mpalmer/debian-mentors_FAQ.html#firstpackage > http://www.infodrom.org/Debian/doc/maint/Maintenance-pkgbuild.html No. Distributions can do whatever they want with the package once they get it, otherwise they get nothing special inside besides patches they want included for interesting or serious reasons. Debian directories in source packages are neither serious nor interesting. (This is final unless somebody buys Timbo and/or Angst a mac in the cause of Debian, at which point I will reconsider) -- - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From vincent at cojot.name Thu Jan 26 04:23:56 2006 From: vincent at cojot.name (vincent at cojot.name) Date: Thu, 26 Jan 2006 10:23:56 +0100 (CET) Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: <43D83756.80603@timedoctor.org> References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> <43D83756.80603@timedoctor.org> Message-ID: On Wed, 25 Jan 2006, Zachary J. Slater wrote: > Per Hansen wrote: >> On GNU/Linux I would like to see a "debian" dir included in the >> sourcepackage. > No. > > Distributions can do whatever they want with the package once they get it, > otherwise they get nothing special inside besides patches they want > included for interesting or serious reasons. Debian directories in > source packages are neither serious nor interesting. > (This is final unless somebody buys Timbo and/or Angst a mac in > the cause of Debian, at which point I will reconsider) Does that include all the different dists and platforms? In ioq2, I have the Solaris pkg templates integrated into the source so that people can easily rebuild the Solaris packages if they want to. Also, I suppose that whatever is debianized for ioq3 will find its way into their mirrors but for Solaris, the 'ioq3' source is most likely the only single place where I can 'publish' the pkg templates which I will be developping. Or would you just want me to publish this stuff along with the installers? One more thing about packaging stuff: Maybe we could ask iD to re-distribute the demo data files? About two or three years ago, I asked iD for permission to package and re-distribute the q2 demo data files (with due credit and unmodified) so that people who download ioq2 for Solaris would actually have something to play with even if they didn't have the full game. I got their permission and therefore the full ioq2 for Solaris is made of two pkgs, one of them being the demo data files.. Ideas, comments? -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, 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 Jan 26 12:13:34 2006 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 26 Jan 2006 09:13:34 -0800 Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> <43D83756.80603@timedoctor.org> Message-ID: <43D9033E.3050208@timedoctor.org> vincent at cojot.name wrote: > On Wed, 25 Jan 2006, Zachary J. Slater wrote: > >> Per Hansen wrote: >>> On GNU/Linux I would like to see a "debian" dir included in the >>> sourcepackage. > >> No. >> >> Distributions can do whatever they want with the package once they get >> it, otherwise they get nothing special inside besides patches they want >> included for interesting or serious reasons. Debian directories in >> source packages are neither serious nor interesting. >> (This is final unless somebody buys Timbo and/or Angst a mac in >> the cause of Debian, at which point I will reconsider) > > Does that include all the different dists and platforms? > > In ioq2, I have the Solaris pkg templates integrated into the source so > that people can easily rebuild the Solaris packages if they want to. > Also, I suppose that whatever is debianized for ioq3 will find its way > into their mirrors but for Solaris, the 'ioq3' source is most likely the > only single place where I can 'publish' the pkg templates which I will > be developping. Or would you just want me to publish this stuff along > with the installers? > I can host whatever packages or extra files any distribution needs if you e-mail them to me, or put them up in some temporary place. (big files don't go into svn, they go on filesingularity) -- - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From sketch at openverse.com Thu Jan 26 12:13:59 2006 From: sketch at openverse.com (Sketch) Date: Thu, 26 Jan 2006 11:13:59 -0600 (CST) Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> <43D83756.80603@timedoctor.org> Message-ID: On Thu, 26 Jan 2006, vincent at cojot.name wrote: > In ioq2, I have the Solaris pkg templates integrated into the source so that > people can easily rebuild the Solaris packages if they want to. Also, I > suppose that whatever is debianized for ioq3 will find its way into their > mirrors but for Solaris, the 'ioq3' source is most likely the only single > place where I can 'publish' the pkg templates which I will be developping. Or > would you just want me to publish this stuff along with the installers? I think this is a good thing for users when it's actively maintained. It's very handy to be able to download the source to something that is not in debian proper (pick your distro), but has a debian dir so that you can build your own packages. When it's something that someone contribued 12 versions ago and it hasn't been updated to go with changes in the app, it's not so great. Same goes for RPM .spec files. > One more thing about packaging stuff: Maybe we could ask iD to re-distribute > the demo data files? About two or three years ago, I asked iD for permission > to package and re-distribute the q2 demo data files (with due credit and > unmodified) so that people who download ioq2 for Solaris would actually have > something to play with even if they didn't have the full game. I got their > permission and therefore the full ioq2 for Solaris is made of two pkgs, one > of them being the demo data files.. Wouldn't be a bad idea, though from my (IANAL) reading of the Q3A EULA included with the demo and point releases, there's nothing to stop people from doing that now. I already am. ;) -- Sketch From stowellt at gmail.com Thu Jan 26 14:05:23 2006 From: stowellt at gmail.com (Tim Stowell) Date: Thu, 26 Jan 2006 12:05:23 -0700 Subject: md4 skeletal model format Message-ID: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> Does anyone know if the icculus codebase has the md4 skeletal model code removed? I am trying to implement Gongo's md4 animation, and it won't compile with the Icculus codebase, but will with Jason's clean Quake 3 source. -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Thu Jan 26 15:51:19 2006 From: tim at ngus.net (Tim Angus) Date: Thu, 26 Jan 2006 20:51:19 +0000 Subject: md4 skeletal model format In-Reply-To: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> Message-ID: <20060126205119.78d32437.tim@ngus.net> On Thu, 26 Jan 2006 12:05:23 -0700 Tim wrote: > Does anyone know if the icculus codebase has the md4 skeletal model > code removed? I am trying to implement Gongo's md4 animation, and it > won't compile with the Icculus codebase, but will with Jason's clean > Quake 3 source. Probably related to this: http://svn.icculus.org/quake3?rev=100&view=rev From stowellt at gmail.com Thu Jan 26 15:54:57 2006 From: stowellt at gmail.com (Tim Stowell) Date: Thu, 26 Jan 2006 13:54:57 -0700 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <20060126205119.78d32437.tim@ngus.net> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <20060126205119.78d32437.tim@ngus.net> Message-ID: <994f7fe90601261254k3e112512ie8d63e3a2142975b@mail.gmail.com> Thanks, that makes sense. Anyone know of any resources about actually implementing an md4 model with the code? On 1/26/06, Tim Angus wrote: > > On Thu, 26 Jan 2006 12:05:23 -0700 Tim wrote: > > Does anyone know if the icculus codebase has the md4 skeletal model > > code removed? I am trying to implement Gongo's md4 animation, and it > > won't compile with the Icculus codebase, but will with Jason's clean > > Quake 3 source. > > Probably related to this: > > http://svn.icculus.org/quake3?rev=100&view=rev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Thu Jan 26 16:05:12 2006 From: tim at ngus.net (Tim Angus) Date: Thu, 26 Jan 2006 21:05:12 +0000 Subject: md4 skeletal model format In-Reply-To: <994f7fe90601261254k3e112512ie8d63e3a2142975b@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <20060126205119.78d32437.tim@ngus.net> <994f7fe90601261254k3e112512ie8d63e3a2142975b@mail.gmail.com> Message-ID: <20060126210512.1c727f55.tim@ngus.net> On Thu, 26 Jan 2006 13:54:57 -0700 Tim wrote: > Thanks, that makes sense. Anyone know of any resources about actually > implementing an md4 model with the code? Have a look at this thread. I don't think Thilo ever got around to writing the txt file, but it might help anyway... http://icculus.org/cgi-bin/ezmlm/ezmlm-cgi?50:sss:164:200509:lkinknodjebaofdgnnik#b From arny at ats.s.bawue.de Thu Jan 26 19:36:50 2006 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 27 Jan 2006 01:36:50 +0100 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <994f7fe90601261254k3e112512ie8d63e3a2142975b@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <20060126205119.78d32437.tim@ngus.net> <994f7fe90601261254k3e112512ie8d63e3a2142975b@mail.gmail.com> Message-ID: <200601270136.53267.arny@ats.s.bawue.de> On Thursday 26 January 2006 21:54, Tim Stowell wrote: > Thanks, that makes sense. Anyone know of any resources about actually > implementing an md4 model with the code? My code patch adds MD4 support for the Raven Software's MD4 models used in EliteForce and some other games. They're consistently labeled with "MDR" instead of "MD4" in icculus.org's codebase and can be enabled if one changes the define RAVENMD4 (in qfiles.h) The remnants of id's original model support have _not_ been touched. Raven has finished the MD4 format a long time ago. There are also some tools out there to actually edit md4 files, also community made tools like a Milkshape Plugin from theSnerd. I only reimplemented the model format for the icculus quake3 engine based on already released and GPL'ed code, because I needed it in one of my projects and found it may be useful to other people. You will find much info on this page: http://dynamic.gamespy.com/~thesynapse/skinning.php Apparently, creating new animations is more complex than I originally thought, but some animation source files and a program to compile them have been provided. -- 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 marc.leeman at gmail.com Fri Jan 27 01:08:45 2006 From: marc.leeman at gmail.com (Marc Leeman) Date: Fri, 27 Jan 2006 07:08:45 +0100 Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> <43D83756.80603@timedoctor.org> Message-ID: <20060127060845.GA2222@scorpius.homelinux.org> > I suppose that whatever is debianized for ioq3 will find its way into > their mirrors but for Solaris, the 'ioq3' source is most likely the only I'll change some minor things today, the 20051231 snapshot should hit debian early next week. -- greetz, marc If he should ask for it, what body part are you willing to offer for it, your Eminence? Pilot - DNA Mad Scientist scorpius.homelinux.org 2.6.15 #1 PREEMPT Thu Jan 5 22:01:16 CET 2006 GNU/Linux -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From fuddl at gmx.de Fri Jan 27 07:40:36 2006 From: fuddl at gmx.de (Bruno Kleinert) Date: Fri, 27 Jan 2006 13:40:36 +0100 Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: <20060127060845.GA2222@scorpius.homelinux.org> References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> <43D83756.80603@timedoctor.org> <20060127060845.GA2222@scorpius.homelinux.org> Message-ID: <20060127134036.0e797af6@localhost.localdomain> On Fri, 27 Jan 2006 07:08:45 +0100 Marc Leeman wrote: > > I suppose that whatever is debianized for ioq3 will find its way into > > their mirrors but for Solaris, the 'ioq3' source is most likely the only > > I'll change some minor things today, the 20051231 snapshot should hit > debian early next week. > > -- > greetz, marc > If he should ask for it, what body part are you willing to offer for > it, your Eminence? > Pilot - DNA Mad Scientist > scorpius.homelinux.org 2.6.15 #1 PREEMPT Thu Jan 5 22:01:16 CET 2006 GNU/Linux the time your packages become official debian ones, please tell me, so that i can make a sync to ubuntu ;) and floam schould then remove the link to my packages in the "get it" section on the website. @marc: do you already have debian packages of quake3? if yes, where can i get them atm? cheers - fuddl -- Among elephants it's not considered cool nor in any good taste to drain other elephants From stowellt at gmail.com Fri Jan 27 11:10:40 2006 From: stowellt at gmail.com (Tim Stowell) Date: Fri, 27 Jan 2006 09:10:40 -0700 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <200601270136.53267.arny@ats.s.bawue.de> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <20060126205119.78d32437.tim@ngus.net> <994f7fe90601261254k3e112512ie8d63e3a2142975b@mail.gmail.com> <200601270136.53267.arny@ats.s.bawue.de> Message-ID: <994f7fe90601270810n444b5ceai31c04ca23bd467d7@mail.gmail.com> Great work, I think this is a really useful feature. You mentioned there are some animation source files and a program to compile--Are these something you made? Where are they at? Thanks! On 1/26/06, Thilo Schulz wrote: > > On Thursday 26 January 2006 21:54, Tim Stowell wrote: > > Thanks, that makes sense. Anyone know of any resources about actually > > implementing an md4 model with the code? > > My code patch adds MD4 support for the Raven Software's MD4 models used in > EliteForce and some other games. They're consistently labeled with "MDR" > instead of "MD4" in icculus.org's codebase and can be enabled if one > changes > the define RAVENMD4 (in qfiles.h) > > The remnants of id's original model support have _not_ been touched. > > Raven has finished the MD4 format a long time ago. There are also some > tools > out there to actually edit md4 files, also community made tools like a > Milkshape Plugin from theSnerd. I only reimplemented the model format for > the > icculus quake3 engine based on already released and GPL'ed code, because I > needed it in one of my projects and found it may be useful to other > people. > You will find much info on this page: > > http://dynamic.gamespy.com/~thesynapse/skinning.php > > Apparently, creating new animations is more complex than I originally > thought, > but some animation source files and a program to compile them have been > provided. > > -- > Thilo Schulz > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From arny at ats.s.bawue.de Fri Jan 27 12:07:10 2006 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 27 Jan 2006 18:07:10 +0100 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <994f7fe90601270810n444b5ceai31c04ca23bd467d7@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601270136.53267.arny@ats.s.bawue.de> <994f7fe90601270810n444b5ceai31c04ca23bd467d7@mail.gmail.com> Message-ID: <200601271807.12782.arny@ats.s.bawue.de> On Friday 27 January 2006 17:10, Tim Stowell wrote: > Great work, I think this is a really useful feature. You mentioned there > are some animation source files and a program to compile--Are these > something you made? Where are they at? Thanks! No! I didn't make them, I didn't even have a look at them because these are windows-only tools and I didn't need to build models myself but only the support for these kinds of models. To the location of tools and informations, I can only point you to this page again: > > http://dynamic.gamespy.com/~thesynapse/skinning.php -- 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 stowellt at gmail.com Fri Jan 27 12:14:03 2006 From: stowellt at gmail.com (Tim Stowell) Date: Fri, 27 Jan 2006 10:14:03 -0700 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <200601271807.12782.arny@ats.s.bawue.de> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601270136.53267.arny@ats.s.bawue.de> <994f7fe90601270810n444b5ceai31c04ca23bd467d7@mail.gmail.com> <200601271807.12782.arny@ats.s.bawue.de> Message-ID: <994f7fe90601270914k9330b35lfeb8eb24b8014b0b@mail.gmail.com> Sorry, saw the page but didn't know if you meant something else. I read through your earlier posts in the mailing list about how you used the Elite Forces vms because it had the necessary changes to cgame. Do I need to make modifications to cgame also to use the new models? Sorry for all the questions, but I am new to this, thanks. On 1/27/06, Thilo Schulz wrote: > > On Friday 27 January 2006 17:10, Tim Stowell wrote: > > Great work, I think this is a really useful feature. You mentioned there > > are some animation source files and a program to compile--Are these > > something you made? Where are they at? Thanks! > > No! I didn't make them, I didn't even have a look at them because these > are > windows-only tools and I didn't need to build models myself but only the > support for these kinds of models. > > To the location of tools and informations, I can only point you to this > page > again: > > > http://dynamic.gamespy.com/~thesynapse/skinning.php > > -- > Thilo Schulz > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From arny at ats.s.bawue.de Fri Jan 27 12:23:42 2006 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 27 Jan 2006 18:23:42 +0100 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <994f7fe90601270914k9330b35lfeb8eb24b8014b0b@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601271807.12782.arny@ats.s.bawue.de> <994f7fe90601270914k9330b35lfeb8eb24b8014b0b@mail.gmail.com> Message-ID: <200601271823.52608.arny@ats.s.bawue.de> On Friday 27 January 2006 18:14, Tim Stowell wrote: > Sorry, saw the page but didn't know if you meant something else. I read > through your earlier posts in the mailing list about how you used the Elite > Forces vms because it had the necessary changes to cgame. Do I need to make > modifications to cgame also to use the new models? Sorry for all the > questions, but I am new to this, thanks. Not for the model format itself. If you want to use the EliteForce models though, you must make the same changes, yes. -- 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 stowellt at gmail.com Fri Jan 27 12:32:59 2006 From: stowellt at gmail.com (Tim Stowell) Date: Fri, 27 Jan 2006 10:32:59 -0700 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <200601271823.52608.arny@ats.s.bawue.de> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601271807.12782.arny@ats.s.bawue.de> <994f7fe90601270914k9330b35lfeb8eb24b8014b0b@mail.gmail.com> <200601271823.52608.arny@ats.s.bawue.de> Message-ID: <994f7fe90601270932r22218afdj2f8e48a50a78cc80@mail.gmail.com> That makes sense. I guess I'm a bit confused, as I thought the model format itself was the same as that used by the EliteForce models, and that to use the skeletal model system at all required a change to cgame. On 1/27/06, Thilo Schulz wrote: > > On Friday 27 January 2006 18:14, Tim Stowell wrote: > > Sorry, saw the page but didn't know if you meant something else. I read > > through your earlier posts in the mailing list about how you used the > Elite > > Forces vms because it had the necessary changes to cgame. Do I need to > make > > modifications to cgame also to use the new models? Sorry for all the > > questions, but I am new to this, thanks. > > Not for the model format itself. If you want to use the EliteForce models > though, you must make the same changes, yes. > > -- > Thilo Schulz > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From arny at ats.s.bawue.de Fri Jan 27 12:53:10 2006 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 27 Jan 2006 18:53:10 +0100 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <994f7fe90601270932r22218afdj2f8e48a50a78cc80@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601271823.52608.arny@ats.s.bawue.de> <994f7fe90601270932r22218afdj2f8e48a50a78cc80@mail.gmail.com> Message-ID: <200601271853.12673.arny@ats.s.bawue.de> On Friday 27 January 2006 18:32, Tim Stowell wrote: > That makes sense. I guess I'm a bit confused, as I thought the model format > itself was the same as that used by the EliteForce models, and that to use > the skeletal model system at all required a change to cgame. I am quite sure that the most significant changes are limited to the loading of the models themselves. Other areas, like doing the animations don't need many changes. Like the original md3 format the md4 format uses frames. In MD4 it's the position of each bone that is saved. The current frame number is stored in the backend.currentEntity->e structure of the renderer. It is up to the cgame to select which frame is to be played at the moment. -- 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 stowellt at gmail.com Fri Jan 27 13:02:18 2006 From: stowellt at gmail.com (Tim Stowell) Date: Fri, 27 Jan 2006 11:02:18 -0700 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <200601271853.12673.arny@ats.s.bawue.de> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601271823.52608.arny@ats.s.bawue.de> <994f7fe90601270932r22218afdj2f8e48a50a78cc80@mail.gmail.com> <200601271853.12673.arny@ats.s.bawue.de> Message-ID: <994f7fe90601271002n24f2c405r78ec474291e711de@mail.gmail.com> Ok, now that is making a lot more sense. So rather than the frame controlling vertex positions as in md3, in md4, the frame stores the position of the bone which then deforms during gameplay? On 1/27/06, Thilo Schulz wrote: > > On Friday 27 January 2006 18:32, Tim Stowell wrote: > > That makes sense. I guess I'm a bit confused, as I thought the model > format > > itself was the same as that used by the EliteForce models, and that to > use > > the skeletal model system at all required a change to cgame. > > I am quite sure that the most significant changes are limited to the > loading > of the models themselves. Other areas, like doing the animations don't > need > many changes. > Like the original md3 format the md4 format uses frames. In MD4 it's the > position of each bone that is saved. > The current frame number is stored in the backend.currentEntity->e > structure > of the renderer. It is up to the cgame to select which frame is to be > played > at the moment. > > -- > Thilo Schulz > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From arny at ats.s.bawue.de Fri Jan 27 13:46:48 2006 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Fri, 27 Jan 2006 19:46:48 +0100 Subject: [quake3] Re: md4 skeletal model format In-Reply-To: <994f7fe90601271002n24f2c405r78ec474291e711de@mail.gmail.com> References: <994f7fe90601261105y42905936u8df3d3fcec99a6e2@mail.gmail.com> <200601271853.12673.arny@ats.s.bawue.de> <994f7fe90601271002n24f2c405r78ec474291e711de@mail.gmail.com> Message-ID: <200601271946.50409.arny@ats.s.bawue.de> On Friday 27 January 2006 19:02, Tim Stowell wrote: > Ok, now that is making a lot more sense. So rather than the frame > controlling vertex positions as in md3, in md4, the frame stores the > position of the bone which then deforms during gameplay? correct. -- 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 marc.leeman at gmail.com Sat Jan 28 05:12:44 2006 From: marc.leeman at gmail.com (Marc Leeman) Date: Sat, 28 Jan 2006 11:12:44 +0100 Subject: [quake3] Re: [lists] [quake3] Binaries!!!11111 In-Reply-To: <20060127134036.0e797af6@localhost.localdomain> References: <43D6CE95.6040500@timedoctor.org> <43D830BA.2040207@yahoo.de> <43D83756.80603@timedoctor.org> <20060127060845.GA2222@scorpius.homelinux.org> <20060127134036.0e797af6@localhost.localdomain> Message-ID: <20060128101244.GC2222@scorpius.homelinux.org> > @marc: do you already have debian packages of quake3? if yes, where can > i get them atm? Should be in the archives and the ITP but: http://scorpius.homelinux.org/~marc/debian/ The quake3-data package will be removed and replaced by a message in the README.Debian, possibly a message while installing). We're (some ppl on Debian-Games) are planning to craft some more general data installer that works for more games/mods. Until that is ready, the engine will be uploaded. The current quake3-data only allows you to choose between the demo and the full CD game. The latest package has some minor menu issue after the renaming to ioquake3, but I am currently compiling an update, since I'm doing a lot of number crunching this is rather slow :-/ -- greetz, marc Have we sent the 'don't shoot us we're pathetic' transmission yet? Crichton - Out of Their Minds scorpius.homelinux.org 2.6.15 #1 PREEMPT Thu Jan 5 22:01:16 CET 2006 GNU/Linux -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From stowellt at gmail.com Tue Jan 31 19:10:48 2006 From: stowellt at gmail.com (Tim Stowell) Date: Tue, 31 Jan 2006 17:10:48 -0700 Subject: JIT compiler on PPC G4 Message-ID: <994f7fe90601311610y2fca22b8re6728b0c5044a318@mail.gmail.com> Was JIT ever working for mac G4 computers (just wondering if ID software never implemented it). Icculus Quake seems to run just fine on the mac I tested it on. My questions are: - Is there just a speed difference without having a JIT? - Are there plans to get JIT working for OS X G4 architecture? Thanks! -------------- next part -------------- An HTML attachment was scrubbed... URL: