Notices by Ignas Kiela (ignaloidas@not.acu.lt)

I feel like people have been misunderstanding the new AMD Ryzen AI Max chips (and relatedly, the Framework Desktop). It's not a "laptop CPU with a GPU tacked on", it's a laptop GPU with a CPU tacked on. By all accounts, it's a GPU-first system - silicon area, memory bandwidth, etc. When comparing it with other AMD laptop chips, it becomes obvious that this is a GPU first, and a CPU second.

Which is why I dislike the criticism of Framework Desktop's soldered memory. Nobody bats an eye if GPU's memory is soldered (when was the last time that it wasn't?), but because people see the chip as a CPU, and thus see the soldered down ram as a regression.

You should in all regards look at the chip as a GPU first, if you're considering to buy something with it. Sure, it has CPU chops, but you can get those way cheaper, and without the limitations that being attached to a GPU brings - if you're not buying it for the GPU, then get something else. Comparisons with other laptop CPU-centric chips are IMO misguided, the comparison should be done with other GPU's first, and CPU's attached to them second.

@mikoto@akko.wtf Yeah, much more into that direction.

It looks cool, but I'm absolutely not in the market for one even if I have the budget, because I do jack shit with GPU's, and the CPU side is way more important to me. I don't game, I don't do LLMs, I don't do any kind of HPC that could go on a GPU, so I have zero reason to buy a system that's GPU first, and a CPU second.

@wolf480pl@mstdn.io @mikoto@akko.wtf I mean, consider that the ram is shared - IIRC you talked about 16GB being bare minimum on the CPU side, and for GPU's, on the higher end at least 10GB for that level of GPU makes sense.

And that the same chip is intended for "mobile workstations", where you do CAD and stuff, where all that RAM is very valuable. It makes sense beyond ML.

@wolf480pl@mstdn.io @mikoto@akko.wtf I'm talking about the chip more than the Framework desktop.

I think the chip makes the most sense in a laptop format, but is hard to fit into existing designs - I think if a second gen of this concept comes out, we'll see a lot more laptops with it, the designs for this gen haven't properly came up yet.

It still makes sense in a desktop format, but the usecases are way narrower - it makes sense when you need a GPU with a lot of memory and tight integration with a CPU, or a complete, fairly powerful system in a small size/thermal envelope - both use cases are something that there are markets for, but it's not universal, and I don't think it's amazing to buy as a gaming rig without needing it to be small/efficient - I haven't seen the marketing from Framework on whether they market it as a gaming rig in general. I see their product as more of "hey, we put this chip on a mITX mobo, and sell a case to go with it if you want", thing, than as a true expansion towards desktop form factor.

@novenary@akko.wtf @lispi314@udongein.xyz @mikoto@akko.wtf @wolf480pl@mstdn.io yet you don't see motherboards with modular VRMs, and those would be infinitely easier to implement than any high speed data busrespectfully, no. Power delivery networks for modern high-power chips are very complex, with high currents (60+A), and also fairly high frequencies because the power draw frequency generally matches the processors frequency. Dealing with tens to hundreds of amps together with fairly high frequency (even with filtering on the board, it's still in at least tens of megahertz) is tough for connectors, and you'd prefer not to have that problem where you don't strictly need to.

@lanodan@queer.hacktivis.me @wolf480pl@mstdn.io @mikoto@akko.wtf @xian@ak.kazuma.family @novenary@akko.wtf while cache kinda makes sense to have separate from a process perspective (SRAM doesn't really scale and other stuff), you really want a low-latency, high-bandwidth connection to it. It's over to 1TB/s in modern CPU's, and that's real hard to put away from the CPU without introducing a bunch of latency.

@sun@shitposter.world @clawfulneutral@soc.port0.org Honestly, I could see him being that concerned about his image, that appearing like he's the smartest guy in a company is more important than any competitive advantage patents would give.

@sun@shitposter.world @clawfulneutral@soc.port0.org tbh it wouldn't surprise me - there is absolutely zero information on who actually does shit at Tesla or SpaceX, like absolute zero. Not even project managers or sth. speaking in PR releases or some shit. Any other company, they would share at least some of the main names from inside, where as with Elon's companies it's all Elon. Dude at seems to want to look like he does everyting at the damn companies.

@sun@shitposter.world yeah, I feel like with rust, all of the worse code will end up getting exposed, and I guess some people just don't want that

@sun@shitposter.world yeah, but managing you can abstract over

you can't abstract over "the only reason this doesn't leak in other drivers is because it leaks on shutting down" https://vt.social/@lina/113051677686279824

@sun@shitposter.world I don't have a citation but I remember reading last year complaints from C developers that they were being harassed by Rust maintainers to document their code better and stop making so many changes because it added a lot of difficulty to maintaining their Rust bindings and they didn't like the work of having to figure out how to safely use the APIs by having to read and understand the C code. There are various technical and historical details for why the C part operates like this.Tbh that's just bad maintainership. Rust people try to write their interfaces in a way that you can't misuse them, and when trying to do that, they find code that's hard to impossible to not misuse, and the maintainers just don't want to fix their shit.

@azonenberg@ioc.exchange from copyright perspective, it contains software, there was a case about Intel's microcode and whether it can be copyrighted as it is on silicon or not (since hardware is not copyrightable)

@sapphire@shortstacksran.ch @icedquinn@blob.cat @Forestofenchantment@clubcyberia.co @sun@shitposter.world Have you fucking read what's in those articles?As no significant British incursions into American territory had occurred by that time, and Britain's war strategy was largely defensive, the election proceeded without disruption.[3]While the Confederacy seemed to have survival potential in summer 1864, it was visibly collapsing by election day in November.There's no end in sight for current war, unlike for the examples you picked

@sapphire@shortstacksran.ch @icedquinn@blob.cat @Forestofenchantment@clubcyberia.co @sun@shitposter.world with a fuck ton of risk reducing measures

it would be pretty fucking hard to reduce the risk of a drone strike to a polling center

@sapphire@shortstacksran.ch @icedquinn@blob.cat @Forestofenchantment@clubcyberia.co @sun@shitposter.world I think "bombs regularly falling on and near major cities" is wartime enough. If you cannot guarantee the safety of election goers, then you shouldn't run them.

@icedquinn@blob.cat @Forestofenchantment@clubcyberia.co @sun@shitposter.world neat attempt to put words into peoples mouths right there

@icedquinn@blob.cat @sun@shitposter.world @Forestofenchantment@clubcyberia.co neat, UK was an autocracy during WW1 then I guess https://en.wikipedia.org/wiki/War-time_electoral_pact

like, I don't see any problem in not doing an election during wartime if there's no pressure from the citizens - the only pressure right now is coming from outside

@sun@shitposter.world @icedquinn@blob.cat @Forestofenchantment@clubcyberia.co like 60% of people said that they shouldn't have elections during war, so it's not like there's much pressure from the citizens

@sun@shitposter.world @foone@digipres.club it's not really the data transfer protocol that's accounting for it, it's mostly just configuring the muxes with USB-PD. I don't think even USB4 which was built with Type-C in mind has any orientation specific stuff.

@gsuberland@chaos.social @phenidone@mstdn.social I'm honestly surprised that the imbalance is that big, that requires the resistance on the rest of the cables to be like 5x bigger than the ones with high current, at least from my playing with a circuit simulator. That's a huge difference.