#
✕ nick:chadmed
#asahi · 2021-10-19 08:50
chadmedso it depends on what specific test its running. some of them seem infeasibly large to reside entirely in cache
#asahi · 2021-10-19 08:50
chadmedwell geekbench does a ton of different tests
#asahi · 2021-10-19 08:48
chadmeddo we know if the cores got a frequency bump? im assuming not since its the same cores on tsmc N5
#asahi · 2021-10-19 08:17
chadmedas in a dual-chip mini or 2x max-based skus?
#asahi · 2021-10-19 08:01
chadmedill have to flog off some hardware before i stuff my lab with any more but man am i disappointed that theres no mac mini
#asahi · 2021-10-19 05:10
chadmedintel's next architecture is a heterogeneous design so theyve gone down that route
#asahi · 2021-10-19 05:09
chadmedamd's already said theyre going to keep iterating on the zen design and maybe add some FFUs to it if they think itll be necessary
#asahi · 2021-10-19 04:13
chadmedsorry just got home from uni, have we expressed dismay at the lack of M1 Pro/Max mac minis yet?
#asahi · 2021-10-18 13:34
chadmedi have phsyiology lab time booked at 10am tomorrow, it is currently 11:30pm and the mac event is at 3am. my dilemma right now is do i make myself another green tea and stay up to watch it, or go to the lab well rested and ready to sit around collecting data?
#asahi · 2021-10-18 13:32
chadmed[22:44] <alyssa> do I have class today? idk! <-- me every day for ~26 weeks of the year
#asahi · 2021-10-18 04:37
chadmedGlanzmann: you can build u-boot so that it runs non-interactively, when was the last time you had to point your android phone to its device tree and kernel?
#asahi · 2021-10-18 04:35
chadmedyeah i dont see a reason for directly embedding the kernel unless you really, _really_ hate u-boot or something
#asahi · 2021-10-18 04:26
chadmeddamn konversation doesnt play nice with fcitx...
#asahi · 2021-10-18 03:22
chadmedoh youre trying to build on macos lmao yeah that'll do it
#asahi · 2021-10-18 03:06
chadmedthe M1 uses 16K pages so as long as that's set you should be fine. not sure on the status of sven's 4K patches rn.
#asahi · 2021-10-18 02:56
chadmedyou should run menuconfig or nconfig and set the kernel up properly. it will stop it from doing that. theres probably some new options in the kernel that arent in your .config or vise versa
#asahi · 2021-10-18 02:49
chadmedm1n1 has some problems with clang i think though so it might be worth investing the time into building an AArch64 gnu toolchain
#asahi · 2021-10-18 02:49
chadmedyeah just add LLVM=1 LLVM_IAS=1 after make
#asahi · 2021-10-18 02:43
chadmedSerentty: https://github.com/AsahiLinux/docs/wiki/Developer-Quickstart#direct-payload-booting
#asahi · 2021-10-18 02:40
chadmedSerentty: you can concatenate the device tree, kernel and m1n1 into a single macho binary and have m1n1 chainload the kernel. its in the quickstart, a bit further down i believe
#asahi · 2021-10-18 02:27
chadmedah right
#asahi · 2021-10-18 02:26
chadmedwould that mean a separate DT for each size macbook?
#asahi · 2021-10-18 02:18
chadmedSerentty: if youve already got m1n1 installed then the developer quickstart guide should still be accurate to follow on from. nothing about that process has changed materially, marcan just made a script to automate some of it
#asahi · 2021-10-17 12:33
chadmedwhat a rousing success the neoliberal experiment has been :-{
#asahi · 2021-10-17 12:04
chadmedyou could, feasibly, use "rt core" hardware to speed up any similar matrix operations, just as the "metal" hardware in the apple gpu can be used just fine for opengl/vulkan features that do the same maths. for others, you can just build a software model of the maths that the particular opengl/vulkan feature does and run it. we will not need to translate metal to vulkan on linux at all.
#asahi · 2021-10-17 12:01
chadmedconsider an "rt core" on an Ampere or Turing gpu. the core doesnt "trace rays" or anything, its just highly optimised for calculating the intersections of matrices, in the case of ray tracing (specifically BVH/culling) that would be the vector representing the ray emitted by the light source, and the matrix representing the coordinates of the bounding volume of whatever the ray is intersecting
#asahi · 2021-10-17 11:59
chadmednsklaus: the gpu doesnt "speak" metal, it "speaks" AGX. when we say that it implements "metal" features all that means is that the hardware has optimised blocks to do the maths associated with those functions.
#asahi · 2021-10-17 11:57
chadmedsomeone please remind me never to do things XD
#asahi · 2021-10-17 11:56
chadmeduh ok so i just realised that the event isnt until the 19th in UTC+10... aaaand its 10pm and i just had an energy drink thinking that the event was this evening...
#asahi · 2021-10-17 09:52
chadmedcounter strike is especially impressive because it contains an in-engine implementation of toGL, so its dx9-->toGL-->moltenvk-->metal all on rosetta 2
#asahi · 2021-10-17 09:50
chadmedeven more impressive is that some games are playable. gtav 60+ fps on a macbook air, csgo 100+fps on the same
#asahi · 2021-10-17 09:47
chadmedagain, not really something im gonna pretend to know how to answer. i am a dumb hardware jock and embarrassingly incompetent at anything software
#asahi · 2021-10-17 09:47
chadmedconsider also that there may be bugs with rosetta 2 that prevent the game engine from doing what it needs to do on the cpu
#asahi · 2021-10-17 09:46
chadmedbut these are definitely questions best directed to alyssa, whether or not moltenvk's level of support can be used as a yardstick to predict how complete our opengl/vulkan implementation will be is something i cant answer for you
#asahi · 2021-10-17 09:44
chadmedi mean yeah its probably moltenvk i just wanted to make a joke about how awfully optimised bethesda games are
#asahi · 2021-10-17 09:43
chadmedyoud be lucky to get fo4 to look anything but bland/incorrect even on a windows machine with directx
#asahi · 2021-10-17 09:42
chadmednsklaus: fallout 4 (or any creation engine game) probably isnt the best tool to use to test an API's stability just saying ;)
#asahi · 2021-10-17 09:37
chadmedoh theyll definitely be able to run, theyll just require some software emulation of functionality that vulkan/opengl have that metal does not. fortunately though, at the end of the day stuff like that is just abstractions of math concepts, and because all GPUs do virtually the same maths even though the hardware has no explicit support for vulkan/opengl its hardware should be able to handle most of both APIs fairly well
#asahi · 2021-10-17 09:35
chadmedif you think that's bad, just remember that x86 chips didnt have integrated FPUs until the the 486
#asahi · 2021-10-17 09:34
chadmednsklaus: yeah the gpu lacks some hardware implementations of opengl/vulkan functions that some games will assume. hopefully we get some software emulation of this missing functionality at some point but that wont be coming from official sources
#asahi · 2021-10-17 09:21
chadmedi have a cruddy A1708 macbook pro you can have, no touchbar :P
#asahi · 2021-10-17 09:19
chadmedgonna go to the gym and grab an energy drink in a little while so i can stay up for the event. 3am? childs play.
#asahi · 2021-10-17 09:18
chadmedthe touchbar being gone has been confirmed for a while now. good riddance imo
#asahi · 2021-10-17 09:14
chadmedwhich under similar conditions would guzzle down hundreds of Watts
#asahi · 2021-10-17 09:14
chadmedyeah its insane, the entire M1 mac mini pulls something like 30W at the wall under full load which is ridiculous considering it is performance competitive with PCs double its price
#asahi · 2021-10-17 09:12
chadmedit has traditionally been a pretty decent proxy for chip power consumption, but is becoming increasingly meaningless because of boosting algorithms and how TDP is calculated by chip designers
#asahi · 2021-10-17 09:11
chadmedthermal design power, basically how much thermal energy the chip is designed to disspiate (measured in W) under "normal" operating conditions
#asahi · 2021-10-17 09:10
chadmedyeah, and its likely that this will continue since thats how apple likes their platform. try not to think of the new macs as PC-equivalent, but rather idevices with a higher tdp, because thats essentially what the platform is
#asahi · 2021-10-17 09:08
chadmedadditionally, theres nothing stopping current ARM/RISC-V vendors from offering a platform like that. in fact, in the server space ARM machines commonly have ACPI or UEFI. theres just not been a need for any of that in the traditional non-x86 markets (i.e. mobile devices, SBCs, embedded, etc)
#asahi · 2021-10-17 09:06
chadmedi.e. you will have an $ISA platform with ACPI, UEFI, "BIOS" settings, etc

plus ›