00:01:06 <Inge-> Is the jury still out, or was there a conclusion about 3950x that it tops out around 15KH/s?
00:33:06 <nioc> Inge-: I saw one report at 16 KH/s
00:33:21 <nioc> this was using 28 threads, 7 threads per CCX gives better results because of the L3 cache
00:33:27 <nioc> so although you get 33% more cores vs 3900X you can only use 17% more threads
03:44:08 <p3rL> *shrugs* https://i.imgur.com/tELTwb4.png
11:40:00 <Inge-> nioc: limits of memory sustem, most likely
11:41:33 <Inge-> Look, a motherboard for the Up-to 64-core arm v8 https://www.tomshardware.com/news/huawei-introduces-desktop-pc-motherboard-for-kunpeng-920-armv8-processors
14:43:50 <hyc> cool. too bad the US has bannedt huawei
14:46:42 <hyc> hm, their desktop offering is weak though, only up to quadcore, only DDR4-2400, only PCIe3.0
14:47:04 <hyc> strange to limit memory controller and PCIe when their server chips already support DDR4-3200 and PCIe4.0
16:46:09 <cohcho> Does it looks like a fake for you: https://github.com/tevador/randomx-sniffer/issues/1#issuecomment-562083663?
16:46:20 <cohcho> look like*
16:53:02 <sech1> it's always "shield and weapon" competition
16:53:16 <sech1> one can detect the sniffer and stop mining while sniffer is working
16:53:36 <hyc> that means you must poll for sniffer existence quite frequently though
17:06:51 <sech1> or you can just hook all processes, replace GetThreadContext() code and "fix" registers for your thread
17:07:01 <sech1> but that counts as OS modification though
17:07:07 <sech1> but very simple to implement
17:07:23 <sech1> and doesn't require kernel driver
17:17:09 <tevador> I'm calling BS
17:19:13 <cohcho> bull...?
17:20:49 <tevador> Joined GitHub on December 2, 2019
17:43:33 <kinghat> https://www.phoronix.com/scan.php?page=article&item=epyc-rome-power
18:23:02 <hyc> quite a perf/watt boost
18:24:47 <tevador> no RandomX tests, disappointing
18:25:03 <hyc> have you suggested them?
20:26:11 <Inge-> I wonder how the 32-core threadripper compares to the 32-core epyc
21:02:57 <gingeropolous> total miners 27247
22:26:47 <sech1> 4GB computers can't mine RandomX: https://www.reddit.com/r/MoneroMining/comments/e6nfoz/speed_na_na_na_hs_what_did_i_do_wrong/
22:26:51 <sech1> one more proof
22:28:40 <Inge-> sech1: I'm not sure I understand why 3950x hashes faster if you free up some MB of L2 cache from the CCD (ref reddit), isn't it 64MB cache for 32 threads as per spec?
22:36:01 <tevador> mining with 4GB is possible on linux
23:25:32 <Inge-> Doesn't this indicate that the threadripper would be more hash/s/watt than a 3900x by a fair bit? https://www.phoronix.com/scan.php?page=article&item=amd-linux-3960x-3970x&num=12