My demo updated again ... now this time it should really wor

[Hyp-X]

On 2002-02-24 04:04, Galilee wrote:
hehe, your "game" is the first game that has been messed up with aniso on. Really strange if you are the first to run into this bug.

It might be strange for users.
It happens all the time for developers.
Drivers happen to be written for existing programs. If you use something that is not used in current titles, and it works you are simply lucky.

 

[Humus]

On 2002-02-24 02:34, Dave B(TotalVR) wrote:

...but how do I get it to work on my KYRO? re-write the driver to support those extensions?

Any plans to suppourt standard multitexturing for the difffuse lighting so it works on DX7 hardware?

Dave

Yup, rewrite the driver :smile:
Currently I feel it wouldn't be much use to put in support for anything below Radeon 7500, it's too slow already and I plan to add more stuff that may cut framerate in half again, we got around 10-20fps on my roommates Radeon 7500 when we tested it last time. Performance has been rised some since then ... but I think it's still too slow on Radeon 7500 to be really useable. Cards not supporting 3d textures in hardware will run in software btw, which means about 0.1 fps or below.

 

[Humus]

hehe, your "game" is the first game that has been messed up with aniso on. Really strange if you are the first to run into this bug.

Well, things like that happends all the time. I've identified many driver bugs in the past. Anisotropic filtering really shouldn't screw up my app, it should be completely transparent to the app. I can't see any reason how it can be my app, so it has to be a driver bug, and it doesn't get screwed up on ATi cards with anisotropic enabled either.

The extreme brightness bug earlier was more of a driver bug than app bug too btw, when investigating the spec for the GL_ARB_texture_env_dot3 I found that either there is a misinterpretation of the spec or ATi has a driver bug, cause their driver replaces the constant 4 in the expression in the spec with RGB_SCALE_ARB while nVidia keeps the 4 and does an additional scaling with RGB_SCALE_ARB. As the spec is written now nVidia does it the right way, but I'm not sure this was the original intent since the extension originates from ATi and their driver have always worked this was as far as I can tell. I've notified them about the problem though. In the mean time I've switched back to the EXT version which doesn't allow postscaling. I'm still a little confused that noone else have found the problem before me ... I can't really be alone using this extension?