#asahi · 2022-04-28 12:13
negglesthe functioning of a nerve cell is incredibly well characterized but i suspect I know where you're going with this argument and I'm not interested in having it#asahi · 2022-04-28 12:11
neggleswhat we don't know is how to make them *think* :P#asahi · 2022-04-28 12:11
neggleshey that's not true, we do know how nerves work, simulating a network of neurons is quite simple #asahi · 2022-04-28 12:08
negglesbut that's not really a question that can be answered until someone starts working on it#asahi · 2022-04-28 12:08
negglesprobably harder in this case#asahi · 2022-04-28 12:07
negglesI completely agree, but ylebars[m] specifically mentioned needing openCL :)#asahi · 2022-04-28 12:06
negglesmps: oh for sure, but not if you want openCL accelerated by the NNPU/GPU/etc...#asahi · 2022-04-28 12:04
negglesit's anyone's guess#asahi · 2022-04-28 12:04
neggleshonestly who knows how long it'll take?#asahi · 2022-04-28 12:01
negglesreverse-engineering and writing drivers for mostly-undocumented hardware is *hard*#asahi · 2022-04-28 12:01
negglesdon't expect to be making even remotely close to full use of the hardware on Linux for at least a year or two, though.#asahi · 2022-04-28 12:00
negglesbut you can probably just buy one and return it if you don't like it / it's too loud.#asahi · 2022-04-28 12:00
negglescheck what the policy is locally in your country#asahi · 2022-04-28 11:59
negglesapple's return policy is fairly lenient, you can return one within 2 weeks or so for a full refund in most countries#asahi · 2022-04-28 11:58
negglesand every computer that's not built to be purely passive will make noise under high load. at low load, it's effectively inaudible#asahi · 2022-04-28 11:57
negglesif your definition of 'noticeable' is 15dB(A), then, yes, because it's impossible to get down to a level that low outside of an anechoic chamber#asahi · 2022-04-24 01:18
negglesisn’t it still lacking support for 16k pages?#asahi · 2022-04-19 00:25
negglesyay#asahi · 2022-04-18 23:51
neggles(i.e. the instructions definitely know better than me)#asahi · 2022-04-18 23:49
negglesI would've thought doing the resize from recovery would've been the way to go? but I still don't actually have an AS machine heh#asahi · 2022-04-18 23:15
negglesif it goes back into recovery and the volume still exists you’ll want to run first aid on it probably#asahi · 2022-04-18 23:15
negglesAndyzzstuMak[m]: I’d just smother it with a pillow (hold power till off) then turn it back on and see what it does#asahi · 2022-04-18 23:09
negglesat this point it’s “wait as long as you feel is justified, then cross your fingers and reboot it”#asahi · 2022-04-18 23:08
negglesi doubt it would actually go to sleep#asahi · 2022-04-18 23:08
negglessleep would, if anything, make it worse#asahi · 2022-04-18 23:08
negglesthe recovery volumes should all be fine in theory#asahi · 2022-04-18 23:08
neggleswelp. leave it as long as you feel is reasonable then hard poweroff#asahi · 2022-04-18 23:05
negglesand now for the words everyone hates to hear: do you have a backup?#asahi · 2022-04-18 23:05
negglesmaybe a race condition of sorts, MC trying to do a read/write which won’t complete but is also blocking resize completion#asahi · 2022-04-18 23:02
neggles^#asahi · 2022-04-18 23:00
negglesdoesn’t mean it can’t still be doing it though#asahi · 2022-04-18 23:00
negglestpw_rules: hours if you’re unfortunate enough to be running it on an Intel mac with a HDD :P#asahi · 2022-04-18 22:59
neggless/nonzero/significant#asahi · 2022-04-18 22:59
negglesi’m afraid I don’t really know either, but yeah turning it off during a resize has a nonzero chance of going horribly wrong#asahi · 2022-04-18 22:57
negglesyeah it definitely shouldn’t take 2 hours#asahi · 2022-04-18 12:52
negglesthat's either a bad cable/bad connection, or a bug in the driver for the M1 Max studio's front USB-only ports#asahi · 2022-04-18 12:51
negglesyeah, the ~200mbps from the front USB-C port#asahi · 2022-04-18 12:50
negglesIOW, "connecting a 6Gbps-at-best-but-not-really SATA SSD to a 10Gbps USB port doesn't make it faster"...#asahi · 2022-04-18 12:49
negglesand hidden towards the end is an admission of the problem and that their results are invalid... > However, a Sabrent+Samsung NVMe conforming to USB 3.2 was found to connect at 10 Gb/s, and returned a read rate of 911 MB/s and write of 973 MB/s.#asahi · 2022-04-18 12:48
negglesif all the SSDs you test with are in SATA drive enclosures, you're never going to see north of 5Gbps throughput regardless of how much bandwidth the port has#asahi · 2022-04-18 12:48
neggles> A range of different external SSDs was used, using Crucial and Samsung SATA SSDs. In personal testing, these were mounted in basic UGREEN and similar cases claimed to be compatible with USB 3.1 Gen 2.#asahi · 2022-04-18 12:46
neggleshmm. the author of this article has been confused by USB-IF, and has also managed to make the overwhelming majority of their results entirely moot.#asahi · 2022-04-02 02:07
negglesbluetail[m] / ChaosPrincess: `dot_clean -m /path/to/volume` :P