Generating random data for benchmark...
Data generated.
Starting benchmark...
Intersecting 1000000 triangles with rays.
Benchmark done.
Time = 95 ms.
Ray per second = 10526315.79
[/quote]
It runs on two cores. Isn't 10m rays per second on a two core cpu too much? I think it's wrong. What do you think?
Are those results right?
Author
Hey, I was writing a simple benchmark, it intersects 1.000.000 triangles with rays, and this is what results I got:
What algorithm do you use to find the intersections? How are you timing the benchmark? Are you sure the compiler isn't optimising away a good part of your benchmark?
Off the top of my head, it sounds a little high, but not impossible.
Off the top of my head, it sounds a little high, but not impossible.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement