02:34:59 <mattexchangefx[m> Never depend on a single income to get financial freedom. I can teach you how you can earn from $5,000 to $6,000 weekly,without sending money to me or anyone, but you will pay me 10% commission when you receive your profit. if you are interested, click on the link to join the chat room 👇
02:34:59 <mattexchangefx[m> https://t.me/+TGide6vQ554mNgRY
11:21:33 <pauliouk> wow, that was like a spam advertisement/cold caller... but on IRC - haven't seen one of those in years :o
11:47:25 <selsta> it's from matrix
16:00:31 <hyc> hmmm https://twitter.com/storagereview/status/1552685051195834368
16:31:54 <hyc> meh. Ampere Altra Max has 128 cores but only 16MB L3 cache
16:32:41 <hyc> Ampere Altra has 80 cores and 32MB L3 cache
16:33:26 <hyc> so even the heavy duty server ARM designs are cache starved
16:33:33 <hyc> stupid
16:42:53 <garth> I find it a bit odd that my Ryzen 3950X is still one of the best miners despite having been released an eternity ago
17:48:08 <gonbatfire[m]> <garth> "I find it a bit odd that my..." <- I wonder if that may be a glimpse of the future
17:48:38 <gonbatfire[m]> I can totally see AMD/Intel (forcedfully) having to design their CPUs to not be good at randomx
17:49:39 <gonbatfire[m]> Like govs putting them a "can only do 10kh/s max" and companies having to design around that
17:49:41 <gonbatfire[m]> s/can/must/
17:50:34 <gonbatfire[m]> Then the actually good at rx CPUs, would be reserved for government miners
17:50:51 <gonbatfire[m]> s/at/RandomX/, s/rx//
17:51:52 <merope> That's not how it works
17:53:05 <merope> RandomX is based off of common cpu instructions. Making a new cpu "bad at RandomX" would basically imply making a new cpu that is slower and less efficient than the previous generation
17:53:48 <merope> Which makes absolutely no sense for a company trying to sell a product (assuming they don't resort to actual scamming)
17:54:31 <merope> Also, the key metric for pow miners is efficiency, not speed
19:47:07 <garth> Future headline: Monero Project adjusts RandomX PoW to unnerf newest CPUs
19:47:55 <hyc> no no. Monero Project releases CPU microcode update to unnerf newest CPUs
19:49:28 <hyc> there's probably enough space in the microcode ROM of a typical Intel CPU to erase all 32bit support and replace with dedicated randomX machine
19:50:33 <gonbatfire[m]> <merope> "RandomX is based off of common..." <- Intel can compete with AMD just fine while being terrible at RandomX
19:50:36 <gonbatfire[m]> Efficiency wise
19:52:57 <gonbatfire[m]> I'm not saying they can prevent their CPUs from running the miner but surely they can do some "optimizations" to reduce efficiency (and reserve the "unlocked" variants for government miners)
19:54:00 <hyc> right now it seems AMD is beating Intel on power efficiency. in laptops at least, I haven't paid attention to desktop in a while\
19:54:15 <gonbatfire[m]> Also govs have the pews pews, why couldn't they force them to make worse CPUs? They already pinned down TSMC
19:55:43 <gonbatfire[m]> hyc: I think that's just due to them having the latest manufacturing nodes 
19:55:53 <gonbatfire[m]> Intel is having a hard time with their fabs
19:55:53 <moneromooo> Why bother ? Instead, tell people they can not use monero or pew pew. Or tell people htey can only walk backwards or pew pew. Or any number of things.
19:56:21 <gonbatfire[m]> * I think that's just due to them having the latest manufacturing processes 
19:56:38 <gonbatfire[m]> moneromooo: There is just two x86 manufacturers 
19:57:12 <gonbatfire[m]> Very easy to capture compared to billions of people
19:57:30 <gonbatfire[m]> I'm sure they can reach a deal..
19:58:05 <gonbatfire[m]> What are you gonna do, build your own CPU? I wish.
20:01:26 <gonbatfire[m]> So yeah just wanted to point that, Zen 2 is better at mining and yet Zen 3 blows it out of the water on all other tasks, clearly some design choices can lead to achieve better efficiency despite worse mining.
20:01:26 <gonbatfire[m]> RX hashrate ≠ general perfomance
20:03:08 <gonbatfire[m]> * your own (better) CPU? I
20:05:07 <gonbatfire[m]> s/RX hashrate ≠ general perfomance/Better at RandomX ≠ Better CPU/
20:09:05 <gonbatfire[m]> * There is just two x86 companies 
21:21:39 <merope> <gonbatfire[m]> "Also govs have the pews pews..." <- Because doing so would directly kill the company's business
21:22:29 <merope> <gonbatfire[m]> "I'm not saying they can prevent..." <- Just look how well that worked out for nvidia and their crippled drivers
22:20:57 <gonbatfire[m]> > <@endor00:monero.social> Because doing so would directly kill the company's business
22:20:57 <gonbatfire[m]> Not sure, that's like saying govs are killing the food industry because they make them comply with FSA, govs regulate industries all the time
22:20:57 <gonbatfire[m]> "Anti-terrorism act: Don't sell too good(1) RandomX CPUs to consumers, only to gov™️ approved actors (Amazon n' shit)"
22:22:13 <gonbatfire[m]> > Just look how well that worked out for nvidia and their crippled drivers
22:22:13 <gonbatfire[m]> Nvidia did that as an afterthought, look at how AMD crippled their cards in mining instead, quite effective
22:22:59 <gonbatfire[m]> RDNA2 suck for mining, compared to RDNA1, it was designed to suck from the start, and yet it's vastly superior in all other aspects
22:23:08 <gonbatfire[m]> s/suck/sucks/
22:28:04 <gonbatfire[m]> To make it clear, I'm not advocating for a switch to PoS or some crap, just pointing what could become a very dangerous threat, but we would still be ∞ times better than with ASICs, fuck ASICs
22:41:28 <gonbatfire[m]> s/threat/trend/
22:53:50 <gonbatfire[m]> s/threat/trend/, s/fuck ASICs/lol/