05:55:11 <sech1> haha, someone actually did it: https://minero.cc/
05:55:19 <sech1> 0.7 h/s on 1 thread on my notebook
06:48:25 <Inge-> I was getting an awesome 6h/s on my phone back in the day - but only cn/r
07:04:07 <sech1> 1 h/s cn/r gives more $ than 1 h/s RandomX, so I have no idea why anyone would run randomx web-mining at all
07:04:21 <sech1> because web-mining cn/r give more hashes
07:05:14 <sech1> cn/r gives about 18% more per hash right now
08:25:09 <binaryFate> lol "Get revenue by running the hidden miner in the background of your Website."
13:36:34 <hyc> 3.6H/s on 8 threads
13:37:02 <hyc> it's not using 2GB RAM is it
13:37:54 <selsta> does not work on my phone
14:04:53 <sech1> hyc It's using light mode
14:12:58 <hyc> yeah that's what I figured
14:13:56 <hyc> I don't see any scenario where this is worth doing. even if you're posting longform articles
14:14:02 <hyc> and it takes several minutes to read them
14:14:09 <hyc> or you're posting lengthy videos
14:14:37 <hyc> even with several minutes per site visit, you're making dirt.
14:26:20 <sech1> watching videos online is one scenario
14:26:28 <sech1> but they certainly need to mine something else
15:42:20 <hyc> on an unrelated topic, would RandomX make a good memtest-like stress test? it would have to be modified to span across all physical memory
15:59:48 <sech1> It's already used for testing. It's the most sensitive test to detect Ryzen's Infinity Fabric instability.
15:59:59 <sech1> But it's not the best for memory testing, memtest is better
16:00:45 <hyc> yeah I suppose uniform bit patterns are more useful for memory testing
16:02:54 <sech1> It's really weird - my system can run for days mining RandomX, but it BSODs frequently when mining good old CN/R
16:03:01 <sech1> It's a bit beyond stable clocks
16:58:41 <tevador> FADD_R instruction in wasm: https://paste.debian.net/hidden/dc468caa/
16:59:47 <tevador> note that it's 1 instruction in x86...
17:23:44 <sech1> next we need someone make RandomX miner in Minecraft
17:34:49 <tevador> or better yet, wasm interpreter in Minecraft and run the wasm version there
17:34:54 <tevador> 1 hash per century?
17:45:18 <tevador> supportxmr is at 40% of the network hashrate
17:45:39 <xmrpow> 200mhs. That is crazy
17:46:22 <xmrpow> One guy just pointed 200mhs there.
17:46:47 <xmrpow> That clearly cant be a botnet
17:47:08 <tevador> nicehash is only ~50 MH/s
17:47:50 <tevador> maybe amazon started mining with spare VMs? lol
17:48:05 <xmrpow> just 150mhs missing ;)
17:48:47 <tevador> it must be some amateur who cannot setup their own pool
17:49:04 <xmrpow> That does not make sense...
17:49:20 <xmrpow> If you own that amount of hash, you have some money on the table...
17:49:34 <xmrpow> and even if you dont own the machines....
18:14:42 <xnbya> its 200MH from azure
18:14:47 <xnbya> was pointed at minexmr before
18:15:13 <xnbya> erdf in -pools is responsible for it it would seem
18:15:32 <sech1> 220 MH/s as of now
18:17:01 <xmrpow> xnbya: Do you know how M5M400 was able to get the info that it is on azure if the guy used a proxy?
18:17:45 <selsta> he didn’t before
18:17:53 <xmrpow> ah ok.
18:17:55 <xmrpow> Thank you
18:18:03 <xnbya> yeah he was directly connecting the miners before
18:18:31 <xnbya> I'm suprised the accounts havent been suspended by azure yet
18:18:37 <M5M400> me too
18:18:48 <selsta> I wonder if he is using stolen/education credits
18:19:02 <selsta> wouldn’t Azure be unprofitable?
18:19:36 <sech1> it would be like 10% profitable
18:19:44 <sech1> meaning you get back 10% of what you spend :D
18:19:49 <M5M400> selsta: I assume stolen, but can't prove it. so I watch and learn
18:21:03 <xmrpow> Shouldnt they recognize the load?
18:21:15 <M5M400> they should
18:21:18 <xmrpow> You can t hide 200mhs that easy
18:21:24 <sech1> well, it's office hours in the US now, let's see how alert they are
18:21:24 <M5M400> it's been going on for days
18:22:03 <xmrpow> M5M400: But not at that scale right?
18:24:42 <M5M400> only in bursts
18:24:47 <M5M400> few hours
18:27:06 <tevador> that's like 500000 vCPUs
18:29:51 <xnbya> was at ~38k connections earlier
18:30:03 <xnbya> so ~5kh per instance
18:31:30 <sech1> M$ had 1+ million servers in 2015, so it must be noticeable for them. I don't know how many servers they have now, but 500k vCPUs is a significant %
18:32:44 <M5M400> ~40k boxes
18:32:50 <M5M400> epycs it seems.
18:32:54 <M5M400> poorly configured
18:33:06 <moneromooo> Maybe they're just running windows.
18:33:22 <moneromooo> (sorry, it just feels so good)
18:33:32 <sech1> They do actually
18:34:02 <sech1> Windows on servers, IIS for websites...
18:34:09 <sech1> Nightmare
18:35:15 <M5M400> i can't really believe they don't care if someone just fully maxxes out 40k boxes
18:35:31 <M5M400> without a warning triggering anywhere
18:35:55 <xnbya> they are probably all on different accounts
18:36:39 <M5M400> unlikely
18:37:13 <M5M400> or does azure just enumerate public IPs across all accounts?
18:37:22 <M5M400> the netblocks are really tight
18:40:52 <tevador> 5 KH/s with an Epyc? must be heavily throttled
18:43:00 <M5M400> or just shit configured
18:59:39 <xmrpow> This guy cant mine with own credits: For 80 epyc cores it  would bring round about 5 Euro per day and it costs  round about 240 Dollar (L80s instance)
19:01:54 <xmrpow> sech1: so not even 10 % ... profit
19:02:51 <UkoeHB_> wouldn't it be -90% profit?
19:03:33 <M5M400> xmrpow: you are aware that azure and aws give away shitloads of credits to startups and research companies?
19:04:05 <xmrpow> UkoeHB: Yes should be -98%
19:04:26 <xmrpow> M5M400: Nope, never used them before
19:04:26 <M5M400> one of my first big miners back in the good ol' days was mining the shit out of AWS on totally legit credits. they got like 1M$ worth of credits shoved up their ass. he made like 100k$ worth of xmr out of it
19:05:29 <M5M400> he also applied for a variety of university clusters... had problems getting shit to compile on that unusual hardware
19:06:14 <M5M400> I agree that the most likely scenario is stolen credits. but simply assuming is not good enough
19:06:23 <xmrpow> true
19:08:04 <xmrpow> M5M400: You could point this guy at our pool. We are having lower fees :P
19:08:58 <M5M400> xmrpow: go ahead, ask him
19:09:21 <M5M400> I guess it'll be over soon though
19:09:26 <xmrpow> Just kidding ;)
19:10:37 <M5M400> i mean, he could totally set up a proxy to mine on a variation of small pools
19:10:44 <M5M400> simultanously
19:10:52 <xmrpow> he could....
19:11:04 <xmrpow> Maybe you can contact him ;)
19:12:42 <xmrpow> Just place him a message at his stats :-D
19:22:07 <M5M400> seems like it's over