[freespace2] Questions.

Taylor Richards mtrs at bellsouth.net
Sat Jun 18 18:08:57 EDT 2005


On Sat, 2005-06-18 at 23:33 +0200, Michal Wozniak wrote: 
>   First of all, hi to all on the list as this is my first port.
>  I ported fs2 to MorphOS (it's a PPC OS compatible with AmigaOS).

Cool!

>  It's using mine implementation (subset) of OpenGL. The game runs
>  fine, but i have some problems with sound. During briefing the
>  speech sound sometimes stops, sometimes skips parts of sentences.
>  Sometimes when rewinding it doesn't restart (seems to be something
>  with OpenAL, as alSourceUnqueueBuffers() fails on assert). This
>  skipping also occurs during missions. I would like to know if it's
>  a known issue or i should look for a bug in openal/sdl/whatever
>  here. Also some clues where to look or what to check would be
>  apreciated.

I've got updated code which should fix all of this but haven't had a
chance to merge it in yet.  I'm going to split audiostr.cpp into it
original version and an OpenAL version so that it will be easier to work
on in the future.

>   Secondly, i have some small fixes. in movie/decoder16.cpp
>  in dispatchDecoder16() case 0xb should be:
> 
>   for (j = 0; j < 8 ; j++ )
>     ( *pFrame ) [ j ] = GETPIXEL( pData, j * 2 );   
> 
>   instead of:
>   
>   memcpy( *pFrame, *pData, 16 );

Sweet.  That's been on the todo list for over a year so it's good that
someone finally fixed it.  I'll try and get this updated before the
weekend is over.

>   This fixes some broken blocks on bigendian systems. Another
>  one is in movie/mveplayer.cpp. Error is not resetted before
>  alGenSources(1, &mas->source_id); in **mve_audio_createbuf()*,*
> * *so if some error occured before it will prohibit movie
>  playig. Maybe not a bug and maybe it only happens here, but
>  it won't hurt to add alGetError() before it.

Part of the newer audio code updates are a better set of error checking
macros.  I've already updated the movie code to use them and that would
take care of this too.  Every OpenAL call is now run inside of these two
macros (one just prints errors, the other fails on them) so it should be
easier to debug in the future.


Taylor

-- 
Taylor Richards <taylor at icculus.org>




More information about the freespace2 mailing list