00:39:11 <hyc> hmm
04:47:52 <hv-bridge> <xx> this result is red hyc
04:48:02 <hv-bridge> <xx> wrong hashsum
17:53:04 <hyc> so apple M1 chip has 128KB L1I$? or 192K?
17:53:10 <hyc> either way, huge.
17:53:16 <hyc> at least 2x AMD, 4x Intel
18:28:37 <sech1> The high-performance cores have 192 kB of instruction cache and 128 kB of data cache and share a 12 MB L2 cache; the low-power ones: 128 kB instruction cache, 64 kB data cache, and a shared 4 MB L2 cache.
18:28:55 <sech1> So 4 fast RandomX threads and 2 slow RandomX threads
18:29:37 <sech1> https://www.extremetech.com/computing/317228-apples-new-m1-soc-looks-great-is-not-faster-than-98-percent-of-pc-laptops
18:29:46 <sech1> 8-wide decode, x86 can only dream about this
18:47:22 <gingeropolous> so are these neural engines just GPUs tailored to machine learning stuff?
18:48:19 <kinghat[m]> maybe one of you ppl might have some info? https://twitter.com/_trapexit/status/1327020162432765952
18:49:05 <kinghat[m]> arm c and c++ compiler things
19:52:49 <hyc> can't help him
19:53:01 <hyc> but it sounds like if he can get an old version of gcc built that'd be enough
19:53:32 <hyc> neural processors are pretty brain-dead. they use 8 bit floating point. I don't even know how that's a thing.
19:54:26 <hv-bridge> <xx> >neural
19:54:30 <hv-bridge> <xx> >brain-dead
19:54:34 <hv-bridge> <xx> is this a pun
20:22:11 <kinghat[m]> thanks for taking a look hyc
22:00:27 <jtgrassie> "tailored to machine learning stuff" <- right, because the majority of Mac laptop owners are doing machine learning stuff lol
22:08:22 <sech1> machine learning stuff is all the way in video streaming/chats
22:08:33 <sech1> noise cancelling stuff, background replacement etc
22:09:09 <moneromooo> human replacement