@DeltaWye@tiggi.es@NanoRaptor@bitbang.social Having done MPI stuff on a Powerbook G4 running a code that was made for and mostly run on BlueGene/L and BlueGene/P (because yay PPC), I can say Apple's memory management at that time kinda sucked.
@gsuberland@ryanc And however many delay slots worth of them due to microcode shenanigans, and the debug registers (you can use them to store keymat if coldboot and DMA attacks are in your threatmodel).
@0x2ba22e11@dashdsrdash@mhoye I'd argue tiny data fits in the CPU cache (now 100MB). Small data fits in RAM/SSD (They're basically the same size: single digit TB). Medium data fits on one disk (tens of TB). Big data needs distributed storage and processing. Very few things are big or even medium data.
@dangoodin Shame that even subscribing doesn't stop the ads and tracking. I pay for a several news site subscriptions and they all fail regularly when using an adblocker, even when I explicitly let the subscription cookie through. It seems news sites are more interested in tracking and malware than in selling subscriptions.
@cccpresser I've used plain old resistors for this too (10MOhm) not as good as the dedicated AlN thermal jumpers, but still 10x as good as 1.6mm of FR4 and much cheaper than dedicated thermal jumpers.
@azonenberg@whitequark@kirakira@jneen I've done matlab/octave/python version of the datasheet test vectors put through a channel simulation FIR for stuff in the past. Not perfect but gives you something to test that will at least exercise the filters.
@Athena@chaosfem.tw What if I explain to TÜV that LLM tainted compilers should make everything they compile be considered LLM output for the purposes of their review and V&V requirements. I'm sure everyone using rust for safety critical firmware would love the extra scrutiny (and associated costs).
@julie@merida.hair@MrNick2@mastodon.social Are the whole bed and support and extruder fully up to temperature before you z-calibrated? I spent a week fighting that in my old monoprice mini. The bed expanded with heat, and took a good 10min after the thermistor said it was at temp.
@annika@xoxo.zone You don't want prompt engineers. They're unstable and likely to break everything. You want nice steady delayed engineers. Maybe some nice water or something to thermalize them will help.
@electropict@mastodon.scot@TechConnectify@mas.to Hybrids want more displacement if possible. GM authored a few papers in the Volt1 era about how a hybrid works best (highest fuel economy) with the engine just below it's power band putting out maximum torque.
@azonenberg@ioc.exchange@funkylab@mastodon.social I'm holding on for the 10GbE Thunderscope. Figure it'll beat the absolute pants off the midrange (350MHz) fiber coupled probes because I can put a few channels (Gate, drain, a supply rail or two, maybe a current) on the thing swinging around at 70V/ns.