- The first suggestion is don't: if you only require a minimal subset of GDI, build your GDI to OpenGL translator, it will get hw accelerated (GDI is on software on Vista) and the world will be a...

Huh? Why would anybody want to do that?

It's also unfortunate that you keep being in denial when it's a highly skilled professional OpenGL driver developer with many years of experience answering.

Like I said Windows is not an "embedded...

Well, frankly speaking, if you are using REALTIME_PRIORITY, you get what you deserve:

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dllproc/base/setthreadpriority.asp
...

All this is good, but - and please take this as constructive criticism - why doesn't the OpenGL wiki have anonymous or free login editing like Wikipedia?

I don't think that having an "elitist"...

Originally posted by atreyu:
Apparently we've been compiling the graphics layer dlls using Multithreaded DLL's.

I would dare to say that's the right thing to do when you have multiple threads.
...

I assume you are loading your graphics layer DLL dynamically.

What driver & card are you having problems with?
Are you doing things in DllMain of your graphics layer DLL you shouldn't be doing?...

The problem here is that OpenGL doesn't specify how a polygon is broken into triangles, for a quad with vertices 0123, in some cases you may get triangles 012, 230 and in others you may get 301, 123...

The driver can already do that without the need of new extensions, there's no reason why it cannot hash the shaders and persist them on disk.

Check if your graphics card supports overlay planes in OpenGL.
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/opengl/ntopnglr_1awk.asp
...

Huh? WinXP 64 is as 'locked' as WinXP 32 is. In both systems you have to install ICDs, they don't come inbox.

What happens is that on WinXP 32 there's a thin D3D wrapper for quake3-style games....

I know NVIDIA uses it (UBB is NVIDIA's naming) I believe it used to be a selling point differenciator between Quadro's and geForces (the geForce driver would turn it off).
I don't think so....

There's a common optimization that ICDs use to minimize the memory footprint allocated to backbuffers. It's called "unified backbuffer" and what they do is to allocate a single desktop-size...

Check your alphabits, the colorbits tell you nothing about alpha:

typedef struct tagPIXELFORMATDESCRIPTOR { // pfd
WORD nSize;
WORD nVersion;
DWORD dwFlags;
BYTE iPixelType; ...

He said he broke a "very large bitmap" into tiles to "get around the OpenGL bitmap size limitation". That sounds exactly like having one big image and tiling it in such a way that each tile is a...

CLAMP_TO_EDGE is the wrong mode for tiling if he is using bilinear filtering. He wants to use GL_CLAMP and textures with border, where the border is the neighbour row/column from the neighbour tile.

That is a FAQ .

Try a 4096x1 texture. The maximum texture size is just that, a static maximum (normally limited by the size of the internal calculations of the texture addressing unit of the graphics chip).
In...

Check a previous post on that. Using FBOs or pbuffers instead of glReadPixels is just a variation.

Also, if your floating point operations are going off, try enabling improve floating point consistency with the /Op flag.

In "unextended" WGL the content of the backbuffer is undefined after calling SwapBuffers.

If glAddSwapRectWIN is supported, you can use the PIXELFORMATDESCRIPTOR flags PFD_SWAP_COPY and...

Interesting. That card probably doesn't support stippling in hardware (or if it supports it, it's disabled by the driver and only enabled on the Quadro counterpart).

That means that the stippled...

My guess is that actually both behave the same, both allocate the full mipmap chain disregarding BASE_LEVEL, but because 3Dlabs supports virtual addressing, it only pages in the mipmap levels of the...

Hmmm I don't agree with that, the whole point of LOD clamping is deferred texture uploads, not reduced memory usage.

Otherwise when the LOD clamp changes (and it will change, otherwise you would...

Are you showing the texture magnified or minified?
Are you using mipmaps in both cases?
Did you try GL_NEAREST_MIPMAP_NEAREST?