Jump to content
  • Advertisement
Sign in to follow this  

Tiled Resources on DirectX 11.2

This topic is 696 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



I am doing tiled resources tests and I have discovered that on my AMD Radeon 280X when I 


map my texture 16384x16484 ( 7 mips )  DXGI_FORMAT_R8G8B8A8 to a null tile takes

around 15-20 seconds in the startup of the application.


After that if I update around 20 tile mappings it gets to around 30-40 ms in the frame.


Also the times are significantly down if i do not use mapping to a single tile and just map the tiles to null, but still around 3-4ms.


Seems that there is some kind of linear or quadratic dependency in the tile mappings on the number of tiles already mapped.


I have noticed that the tile mappings times are down also when one uses BC1, since it uses less tiles around 2731 instead of 20000+ and is managable to 0.3-0.4ms per frame, which is also slow, since they are just 20 mappings for 1 texture.


On NVidia hardware is much better. I have not been able to find documentation about these timings apart from the fact that they say, batch your tile mappings, which i do already.


So if somebody provides an explanation why there is such a dependency will be great. 


I can post source code, if needed on git hub.


On the internet there is one example of tiled resources from 2013 from Microsoft, which also suffers from this slow down.



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!