12:11:25
sech1:
AMD lineup from Zen 2 to Zen 5 has been tested: https://gist.github.com/SChernykh/6058ecf01c929883b9d19c7eeadc8809
12:11:33
sech1:
Need more Intel CPUs, preferably new ones
12:11:48
syntheticbird:monero.social:
Could you stop cooking please
12:12:11
syntheticbird:monero.social:
it's very embarrassing for us mere mortals
12:12:35
DataHoarder:
interesting to see even Zen2 going that well
12:13:33
sech1:
Yes, I expected Zen 5 to get better improvement
12:13:43
sech1:
Maybe I don't fully understand where the actual bottleneck is
12:13:55
sech1:
But it has to do with memory access, because program size increase helps all Ryzens
12:21:36
DataHoarder:
AMD uProf works well to gather the statistics on memory or retired ops and delays
12:22:22
DataHoarder:
though if you want specifics you probably should just create a bunch of programs ahead of time so it can annotate them
12:25:20
DataHoarder:
there's a bunch more small samples in comments
12:28:10
DataHoarder:
my weird nerfed Zen2 with just P012 should arrive in a couple of days :)
21:51:30
benchmarking420:matrix.org:
### RandomX v2 Benchmark Results
21:51:31
benchmarking420:matrix.org:
**AMD RYZEN AI MAX+ 395 w/ Radeon 8060S @ 99.86W**
21:51:33
benchmarking420:matrix.org:
| Algorithm | Hashrate | Relative Speed | Hash/Joule | VM+AES/s | VM+AES/Joule | Relative Work/Joule |
21:51:35
benchmarking420:matrix.org:
|-----------|----------|----------------|------------|----------|--------------|---------------------|
21:51:37
benchmarking420:matrix.org:
| RandomX v1 | 15073.65 | 100.0% | 150.94 | 63.22e9 | 633.12e6 | 100.0% |
21:51:39
benchmarking420:matrix.org:
| RandomX v2 | 14003.05 | 90.0% | 139.40 | 88.10e9 | 877.05e6 | 130.0% |
21:51:41
benchmarking420:matrix.org:
**Config:** threads=32, affinity=0xFFFFFFFF, init=32
21:51:43
benchmarking420:matrix.org:
**Stability:** V1 crashes: 0/2, V2 crashes: 0/2
21:52:35
benchmarking420:matrix.org:
(didnt tweak nothing, lazy, not running headless, lazy, numbers could be better probably, didnt check from the wall yet, lazy)