Re: GLX in bad state

You are the CTO of Transgaming and this is your forum.  I am hoping that 
this forum is free enough for me to continue pursuing this topic with 
other interested Transgamers.

As far as your dismissal of this topic is concerned, you are wrong.  I 
have managed to produce just under 5 fps just by introducing a more efficient
video memory allocation scheme.  It is not perfect but it does prove my point.

It is possible that there are other factors besides the absence of 
ARB_render_texture that are slowing the game down.  However, if you took 
away from Windows the texture rendering capability there is no doubt that 
C4 would run slower.  And at around 30 fps for my system the graphics card 
would need to push about 0.5 extra GB every second to maintain this 
frame rate.

C4 is a benchmark and the shadow and reflection effects it produces are 
not unique.  And they are becoming more prevalent.

I am looking forward to reading performance measurements from a variety of 
different systems posted by various Transgamers.  I think it would be very 
helpful.

I have added comments below. 

On Fri, 2 Apr 2004, Gavriel State wrote:

> >>Nadav Frum wrote:
> >
> > The problem with the absence of a render_texture extension is two fold:
> >
> > 1.For each texture that becomes a render target WineX creates a pbuffer.  
> >   That means double the amount of video memory is used for each such
> >   texture.  If you have 20 MB of renderable textures that means that you 
> >   need to allocate 20 MB more of video memory.  Do not expect to run the 
> >   C4 benchmark with a 32 MB video card.
> 
> That's only if it renders to every texture, which is very unlikely.Most games
> that do this kind of thing render to 2-3 textures, no more.

Let us see.  There are several shadow casting objects:  choppers, numerous 
jeeps, plenty of vegetation and buildings, etc.  For reflections there is 
the ship, the chopper and the general scenery.  It is more than 2-3.  
There are also the textures used to bypass the 2 active texture limit on my 
card.   All in all, around 70.

When you start drawing shadows, every visible object will require one.  
Most games have more than 2-3 visible objects.  Some shadows can be 
aggregated.

> >>It should be easy enough to estimate the speedup such an extension would get
> >>by removing the glCopyTexSubImage2D call from X11DRV_GLX_UpdateTexture.
> >
> > You are welcome to try and to post your results.
> 
> I have. There was no noticable difference in performance in the two cases.

This is interesting.  Please post your results and your system specs.

> The problem is somewhere else.

Where?

> Regardless, even if it was an issue, there's little that anyone outside of
> NVidia, ATI, and the ARB can do about it.And they're already working on
> something called 'Superbuffers' which will hopefully address this issue.

>From a quick look at the net, Super Buffers is the name given to the ARB 
working group put together to study and produce a common industry proposal 
for renderable textures and other buffer schemes.

The discussion I am attempting to start is simply to identify problems.  
When the problems have been identified perhaps somebody will agree to 
address them.  I am sure that there are plenty people at NVidia, ATI and 
the ARB that do not spend all their time in the Super Buffers working 
group's deliberations.

Nadav Frum
----------