11:58:43
sech1:
RandomX v2 official pull request: https://github.com/tevador/RandomX/pull/317
12:13:07
plowsof:
🚀
13:38:02
gingeropolous:monero.social:
anybody got those newer intel chips? still can't convince myself to buy a dead socket...
13:38:23
gingeropolous:monero.social:
maybe i'll get a new laptop. this thinkpad is officially 10 yrs old now
13:40:57
sech1:
new panther lake laptops should be interesting: https://www.wired.com/story/intel-panther-lake-core-ultra-series-3-review/
13:42:00
gingeropolous:monero.social:
yeah im hoping montero does some more uppp and i can get one.
13:43:37
gingeropolous:monero.social:
and i can't be the only one willing to run these benchmarks...
13:43:54
gingeropolous:monero.social:
dang reddits all up in arms about the x9 yet no ones running these v2 benches
13:45:06
gingeropolous:
sech1, i put some new benchmarks up in the repo. are these numbers looking like they should? the old 3900x is not lookin great
13:45:54
sech1:
Looks like the prefetch tweak doesn't help Zen2 at all, helps Zen3 a little, helps Zen4 more and helps Zen5 a lot
13:47:08
sech1:
3900X hashrate didn't change (save for the measurement noise) at all compared to the previous test
13:47:22
sech1:
My 5950X test (Hetzner server) is in line with your 5900X test
14:14:54
gingeropolous:
maybe its because i have that 3900x undervolted so much or something. my 3950x shows 91%
14:16:57
gingeropolous:
eh im gonna run more tests on that. only 1 ran completed out of 2. that cpu might be on its last legs
14:22:13
sech1:
3950X got +600 h/s improvement