IMG PowerVR RTX - what is it?
- Thread starter Shifty Geezer
- Start date
Raytracing doesn't need so much processing power as memory power. The old SAARCor raytracing processor was achieving amazing performance from a 50 MHz part or whatever it was. The problem comes with testing every object for each ray calculation, and doing repeatedly for several iterations. Shading of each pixel requires the same level of processing power as usual. I expect true reflections/refractions to add extra burden on texture consumption too. A convex reflective surface will take in a wider view and hence more surfaces meaning more textures being sampled.
Dominik D
Regular
Memory burden depends on the technique. As with everything else: there are RT methods trading memory requirements for power and vice versa. Don't know what Caustic are doing though and whether their hardware does some tricks to work around the memory requirements for the memory-heavy techniques or optimizes computations in a way that makes it possible to trace efficiently with computation-heavy techniques on a modest hardware.
I understand space partitioning has improved object searches over the old brute-force method, but is there any solution for the random memory access patterns of searching for ray hits? If a ray is cast in a direction, the only way to know if its hit an object, and where on that object, is to read the object coordinates and test them all. I don't see how that can be calculated without memory accessing of the order of number of objects in the scene.
All the primitives in the lowest level bounding volumes crossed have to be tested obviously, but with breadth first raytracing you can amortize the cost across the rays ... even without the coherence of primary/shadow rays (basically you don't finish rays packets one at a time, you buffer the rays at a given level of the hierarchy and resort them into new ray packets during tracing .... creating some coherence out of chaos).
By using anyone of the many acceleration structures known (i.e. BVH, QBVH, Kd-tree, etc.), the complexity is O(log number_of_primitives), not O(number_of_primitives).
This has been, in the past, one of the argument to promote ray tracing over classic rasterization (i.e. DirectX/OpenGL-like rendering).
However, the visit of the acceleration structure (usually some kind of tree) leads to very scattered memory accesses after few nodes and this seems the major problema the moment.
No they don't. They have upset market positions in two markets they entered recently ... but they did it with polish, not revolution. Sony up till recently was the revolutionary in the console market, but given how PS3 turned out that's unlikely to continue.
As I said, breadth first raytracing solves this to an extent.
