Archived

This topic is now archived and is closed to further replies.

D3D vs. OGL (drivers) speed tests?

This topic is 5626 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Okay the forum moderators are ALREADY hovering over the "close thread" button, but please let me finish.. I''m not posting a D3D vs. OGL topic, but I was curious to find out about SPEED/PERFORMANCE tests between (well written) OpenGL and Direct3d8 drivers. ie. Take an NVidia card and update it to the latest Detonator drivers. Now NVidia seems to be in the middle of the road of the OpenGL vs. Direct3D8, so I''m curious to know if there exists any timing information between the two sets of drivers. All I''m after is numbers here, realizing that SOME video card driver writers are more biased towards either OpenGL or Direct3D for their numbers..(is ATI sort of in the middle as well?) thanks for any links, etc..

Share this post


Link to post
Share on other sites
Just curious - why?

Also, I would think it would be difficult to decouple the effects of the driver from the effects of the software that uses the driver. How will you get good "driver only" numbers? I think you can get some info, but I don''t know how useful it will be.

Share this post


Link to post
Share on other sites
quote:

Okay the forum moderators are ALREADY hovering over the "close thread" button, but please let me finish..



If it''s a valid question that hasn''t been asked a thousand times before, it may just stay open for a bit

quote:

All I''m after is numbers here, realizing that SOME video card driver writers are more biased towards either OpenGL or Direct3D for their numbers..(is ATI sort of in the middle as well?)



If the hardware functionality is the same, then all functions calls, regardless of API, should run at the same speed. How many ways is there to draw a polygon - all that differs is the setup (API specific). Also, the slow-down from one API usage to the other is probably because emulation is used.

You could be right, however, as a manufacturer being biased on one or the other. It''s happening right now with other companies (Sony degradates non-Sony PS2 memory cards, Motorolla drains non-motorolla cell batteries faster, etc).



Jim Adams
home.att.net/~rpgbook
Author, Programming Role-Playing Games with DirectX

Share this post


Link to post
Share on other sites