Vector Class Discussion

simple ray casting test slower when vectorized
Author:  Date: 2013-03-14 13:02
Hi Epsilon. I looked at your code and it looks to me like your not using SIMD optimally.

For ray casting you will get a lot more throughput by using packet based ray casting. In other words using an array of struct of arrays (AoSoA). For SSE this means you store four rays like xxxxyyyyzzzz. You should get a four times speed up with SSE and about 6x with AVX. See for example
graphics.stanford.edu/~boulos/papers/cook_gi07.pdf

 
thread simple ray casting test slower when vectorized new - epsilon - 2013-03-02
last replythread simple ray casting test slower when vectorized new - Agner - 2013-03-03
last replythread simple ray casting test slower when vectorized new - epsilon - 2013-03-03
last replythread simple ray casting test slower when vectorized new - Agner - 2013-03-03
last reply simple ray casting test slower when vectorized - chad - 2013-03-14