←back to thread

Measuring Acceleration Structures

(zeux.io)
78 points ibobev | 1 comments | 02 Apr 25 12:28 UTC | HN request time: 0.241s | source
1. vardump ◴[02 Apr 25 18:17 UTC] No.43559646[source]▶
Smaller data is where it’s at when optimizing nowadays. Less bandwidth required and higher cache hit rate.

You can compute a ton per bit transferred from DRAM. On both CPUs and GPUs.