Jump to content
  • Advertisement
Sign in to follow this  
dnaxx

OpenGL Texture compression vs. speed

This topic is 4838 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

Hello, Does someone know, if texture compressions results in higher framerates with modern hardware (in other words: makes it sense to use texture compression, to speed up my opengl app?)? Thanks,

Share this post


Link to post
Share on other sites
Advertisement
it depends where the bottleneck is in your program, however using compressed textures does reduce the bandwidth requirements for texture sampling, which can make a difference

Share this post


Link to post
Share on other sites
use texture compression on ALL hardware not just modern
the only place u mightnt want to use texture compression is for stuff where u want pixel perfect textures eg HUD textures but for the rest of the scene use it

Share this post


Link to post
Share on other sites
is it correct, that only the upload-process is faster due to reduced bandwidth requirements?

Share this post


Link to post
Share on other sites
Quote:
Original post by dnaxx
is it correct, that only the upload-process is faster due to reduced bandwidth requirements?


No. Lower internal memory (VRAM) bandwidth requirments increase performance.

Share this post


Link to post
Share on other sites
I implemented the texture compression and now the framerate is 1-2 fps lower. perhaps speed increases when more textures are in use.

Share this post


Link to post
Share on other sites
Quote:
Original post by dnaxx
I implemented the texture compression and now the framerate is 1-2 fps lower. perhaps speed increases when more textures are in use.


It requires the GPU to do additional work to decode them so it is not going to be faster in the way you are thinking. Yes it makes more of a difference when you have alot more textures. It allows you to pack more on the card and when it does need to be transferred on the bus there is less (compressed) data to deal with.

Share this post


Link to post
Share on other sites
it also, as I said, depends on where the bottleneck in your application is, if you are GPU load then adding more work for the GPU todo isnt going to improve things.

You said you lost '1 or 2 fps', what fps did you start at? Also, how complex is the scene you are testing? and what card do you have?

Share this post


Link to post
Share on other sites
I am currently using a Radeon 9000 with 64 MB. The fps is between 40 and 60 with ~ 10 mb of textures.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!