Is exist any realtion between resolution of color/pixel buffer and speed of ARB(NV)_occluding_query?
For me is not really ineresting, how much pixels is visible(important only visibility, like in HP_ocl_test), but very useful quering of results.
Thanks.
I think the speed is dependent on the number of pixels rendered in the occlusion query and not directly dependent on the resolution of the buffer. Obviously if you have a high res pixel buffer then you will be rendering/testing more pixels in the occlusion query. I would have thought the HP extension would be effected in the same way.
[This message has been edited by Adrian (edited 11-13-2003).]
2Adrian
thanks…
General idea was: if VPU support some high-accuracy internal calculation, it can use fixed(low) resolution(or hierarchical depth on ATI) for all tests, result can be received through some coefficient.
may be I think to complicated about this extension - if its only drawing to depth buffer(withot any additional algorithms) with follow reading of result - relation is exist(linear).