[r321] Speed vs Quality rendering problems

[Agent ME]

More than likely this is just my laptop's video card acting up, but thought this was odd.
With the OpenGL rendering setting set to Speed, I often get a few white pixels.
With it set to Quality, it generally looks a lot better, until I stumbled on this on Doom2's MAP16:

It doesn't happen in the Speed rendering setting.

Any idea what could cause this besides my video card?

(This is in r321, because it's a revision I know compiles and works well in Linux.)

[Rachael]

This is a definite video card/driver issue, most likely outdated drivers. You might go to your card's website and see if there's a more recent Linux driver.

Any idea other than that? Sorry, no.

[Enjay]

This looks similar to the pics posted in this thread:

http://forum.drdteam.org/viewtopic.php?f=24&t=3941

[Gez]

Do you have "smooth sprite edge" on? If not, try with it.

As weird as it seems, I had the same issue and that's what fixed it.

[Agent ME]

Changing "Force Additive Lighting" to on fixes this (but seems to make lights not as bright).
Changing "Enable Brightness Maps" to off fixes this (but I prefer to have brightness maps).
Changing "Shaders for Fog" to off fixes this.
Changing "Fog Mode" to off fixes this (but I prefer to have some fog).
Changing "Smooth Sprite Edges" to on fixes this (guess I'll stick with this setting as it seems beneficial).

About my drivers, they're already the latest automatically installed from the Ubuntu repository, but I might try to upgrade them to Nvidia's latest manually.
My video card is an Nvidia 8600M GS.

[Graf Zahl]

Changing "Smooth Sprite Edges" to on fixes this (guess I'll stick with this setting as it seems beneficial).

Thanks for this piece of information. It looks like having sprites rendered along with non-translucent walls somehow corrupts the shader state and leaves an active brightmap where none is supposed to be.

[GuntherDW]

I'm also "suffering" of this problem, but most of the time i just ignore it.

I'm always using the latest (beta) drivers though, (185.18.14 atm) and this still occurs (in r324).

I'll update my source revision and check if it's still occuring, it's rather random though so i'm not sure about wether it'll be gone or not.
(edit: i'm using a 9800GTX+ from MSI, N9800GTX+ OC)