←back to thread

AI capex is so big that it's affecting economic statistics

(paulkedrosky.com)
337 points throw0101c | 1 comments | 18 Jul 25 19:59 UTC | HN request time: 0.207s | source
Show context
mikewarot ◴[18 Jul 25 21:11 UTC] No.44609878[source]▶
I'm waiting for the shoe to drop when someone comes out with an FPGA optimized for reconfigurable computing and lowers the cost of llm compute by 90% or better.
replies(6): >>44609898 #>>44609932 #>>44610004 #>>44610118 #>>44610319 #>>44610367 #
pdhborges ◴[18 Jul 25 21:14 UTC] No.44609898[source]▶
Where is that improvement coming from? Hardware is already here to compute gemm as fast as possible.
replies(1): >>44609936 #
1. leakyfilter ◴[18 Jul 25 21:17 UTC] No.44609936[source]▶
Raw gemm computation was never the real bottleneck, especially on the newer GPUs. Feeding the matmuls i.e memory bandwidth is where it’s at, especially in the newer GPUs.