12:14:56 RandomXv2 official pull request by sech1 https://github.com/tevador/RandomX/pull/317 (shared in #monero-pow) 15:22:32 plowsof: wait, +52% more operation mean 52% more performences ? 15:23:18 it's a different pow function 15:23:51 it's not directly equivalent but yeah, it does way more work (when measured in instructions, memory accesses) and at the same time higher hashrate on modern cpus compared to v1 15:24:17 look at https://github.com/SChernykh/RandomX/blob/v2/doc/design_v2.md#4-performance-impact benchmarks 15:24:28 look at VM+AES/s, that is absolute work 15:24:39 yeah I stumble on that, thanks 15:24:41 woah 15:24:49 or relative work/joule to compare 15:25:19 so while older gens might end up slower H/s they also do more work than v1 15:26:34 how does more instruction outputed result in lower hashrate 15:28:26 this look very good for zen5 but all other seem to receive more bad outlook 15:30:39 cause v2 does 384 program size instead of 256 15:30:41 for example 15:31:06 v2 hash is 1.5x more work than v1 hash 15:31:11 so don't look that v2 hashrate is lower 15:31:12 yeah, modern ones benefit more (we push their memory further) 15:31:23 multiply it by 1.5 first if you really want to compare hashrates 15:31:29 hmm 15:31:35 so I guess its not about hashrate, but higher share result ? 15:31:40 ^ they are pow functions, you can't compare each other directly. compare VM+AES/s or relative work/joule 15:31:42 no 15:31:50 hm 15:32:13 it's about work/joule (does more work per joule, aka more efficient) 15:33:41 it just so counter intuitive. Hashrate lower but does output more operation 15:33:55 Well, mine Bitcoin if you want more hashrate 15:33:57 cause you are comparing say 15:34:11 bitcoin sha hashrate to randomx hashrate 15:34:13 they do different things 15:34:15 same on v1 to v2 15:34:19 they do "different" things 15:34:21 just trying to warp my brain around, im non technical sorry 15:34:34 It's like dollar and euro 15:34:41 Similar things, but not 1:1 rate 15:34:53 imagine this. randomx 2x which is two randomx hashes 15:34:54 v1 hashes and v2 hashes 15:34:57 you have half hashrate 15:34:59 but 2x work 15:35:04 ahh > they do different things 15:35:19 this is why you can't compare hashrate directly 15:35:46 I see. Thanks for the explanation :)) 15:35:50 in this case it's like randomx 1.5x :') but with more specific changes that CPUs do well 15:35:58 you think it imght goes live next couple weeks ? 15:36:54 it needs a hardfork 15:37:17 ohh. So I guess might came with fcmp one 15:37:34 or around 15:38:26 exciting times 15:41:02 appreciate the hard work put behind everything of this 15:55:37 so modern CPUs are better but I can't build a new rig cause ram has 4x in price :( 16:02:22 If you undervolt and optimize for efficiency, a single RAM stick will do. 1 stick can handle up to 20-21 kh/s 16:04:43 you can use laptop memory :') 16:13:35 hmmm 16:30:09 nioc: good news is, ddr5 didnt affected that much the hashrate, on V1 at least 16:30:24 so now 1 kit of ram can become 2 rigs 16:30:49 should be interesting thing to test it out also, how ram effect hashrate on Zen5 with V2 16:47:02 Zen5 is still capped on bandwidth 16:47:28 so it shouldnt change ? 16:47:51 I have slow ram on a zen5 platform and it was also equivalent. The prefetch trick makes latency tuning less critical afaik (but you still benefit) 16:48:17 In my own implementation, it was like +2 KH/s over V2 without the memory prefetch 16:48:47 Without that there were no large differences across generations 17:08:36 thanks for the insight 18:27:58 [CCS Proposals] Lee Clagett opened merge request #637: Mark 2025 Q4 Month 1 Completed https://repo.getmonero.org/monero-project/ccs-proposals/-/merge_requests/637