Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple M1 Ultra

apple.com

Post reply on HN
Print view View original

291–300 of 855 posts

Re: Apple M1 Ultra

#291

Post by dheera »

ChuckMcM wrote: post #276

In some ways I wish this processor was available from a CPU chip seller. As a compute engine it gets a lot "right" (in my opinion) and would be fun to hack on. That said, the idea that USB C/Thunderbolt is the new PCIe bus has some merit. I have yet to find someone who makes a peripheral card cage that is fed by USBC/TB but there are of course standalone GPUs.

> USB C/Thunderbolt is the new PCIe bus

Oh please hell no.

I have to unplug and plug my USB-C camera at least once a day because it gets de-enumerated very randomly. Using the best cables I can get my hands on.

File transfers to/from USB-C hard drives suddenly stop mid-transfer and corrupt the file system.

Don't ask me why, I'm just reporting my experiences, this is the reality of my life that UX researchers don't see because they haven't sent me an e-mail and surveyed me.

Never had such problems with PCIe.

Re: Apple M1 Ultra

#292

Post by jamesfmilne »

ChuckMcM wrote: post #276

In some ways I wish this processor was available from a CPU chip seller. As a compute engine it gets a lot "right" (in my opinion) and would be fun to hack on. That said, the idea that USB C/Thunderbolt is the new PCIe bus has some merit. I have yet to find someone who makes a peripheral card cage that is fed by USBC/TB but there are of course standalone GPUs.

https://www.sonnettech.com/product/thunderbolt/pcie-card-exp...

Re: Apple M1 Ultra

#293

Post by slmjkdbtl »

Curious about the naming here, terms like "pro" "pro max" "max", "ultra" (hopefully there's no "pro ultra" and "ultra max" in the future) is very confusing and hard to know which one is more powerful than which, or if it's a power-level relationship. Is this on purpose or it's just bad naming? Is there example of good naming for this kind of situation?

Re: Apple M1 Ultra

#294

Post by adfgadfgaery »

msoad wrote: post #76

At what point Apple will put those chips in their servers or sell server chips? It only makes sense for them to take this architecture to the cloud deployments

It doesn't make much sense to me. The M1 is designed to have memory in the same package as the processor. This leads to reduced latency and increased bandwidth. Moving to off-package memory might totally destroy its performance, and there is an upper limit on how much memory can go in the package.

The M1 Ultra is already a little light on memory for its price and processing power; it would have much too little memory for a cloud host.

Re: Apple M1 Ultra

#295

Post by nicoburns »

fivea wrote: post #261
kiratp wrote: post #230

Earlier quoted context omitted.

My workstation has a 3990x. Our "world" build is slightly faster on my M1 Max. https://twitter.com/kiratpandya/status/1457438725680480257 The 3990x runs a bit faster on the initial compile stage but the linking is single threaded and the M1 Max catches up at that point. I expect the M1 Ultra to crush the 3990x on compile time.

> The 3990x runs a bit faster on the initial compile stage but the linking is single threaded and the M1 Max catches up at that point. Isn't linking IO-bound?

https://github.com/rui314/mold would suggest otherwise. Massive speedups by multithreading the linker. I think traditional linkers just aren't highly optimised.

Re: Apple M1 Ultra

#297

Post by aneutron »

marcan_42 wrote: post #172
kasperni wrote: post #72

Earlier quoted context omitted.

> I wonder if they'll just do 4x M1 Max for that. Unlikely, M1 Ultra is the last chip in the M1 family according to Apple [1]. "M1 Ultra completes the M1 family as the world’s most powerful and capable chip for a personal computer.”" [1] https://www.apple.com/newsroom/2022/03/apple-unveils-m1-ultr...

I've been saying 4x M1 Max is not a thing and never will be a thing ever since the week I got my M1 Max and saw that the IRQ controller was only instantiated to support 2 dies, but everyone kept parroting that nonsense the Bloomberg reporter said about a 4-die version regardless... Turns out I was right. The Mac Pro chip will be a different thing/die.

Bloomberg brought the Supermicro hit pieces. I personally can't take them seriously anymore. Not after the second article with 0 fact checking and sad attempt at an irrelevant die shot. And their word is certainly irrelevant against one of people who are working (and succeeding) at running linux on M1.

Re: Apple M1 Ultra

#298

Post by droopyEyelids »

dheera wrote: post #291
ChuckMcM wrote: post #276

In some ways I wish this processor was available from a CPU chip seller. As a compute engine it gets a lot "right" (in my opinion) and would be fun to hack on. That said, the idea that USB C/Thunderbolt is the new PCIe bus has some merit. I have yet to find someone who makes a peripheral card cage that is fed by USBC/TB but there are of course standalone GPUs.

> USB C/Thunderbolt is the new PCIe bus Oh please hell no. I have to unplug and plug my USB-C camera at least once a day because it gets de-enumerated very randomly. Using the best cables I can get my hands on. File transfers to/from USB-C hard drives suddenly stop mid-transfer and corrupt the file system. Don't ask me why, I'm just reporting my experiences, this is the reality of my life that UX researchers don't se…

Friendly reminder that USB-C is a form factor, and thunderbolt is the actual transfer protocol.

Sounds like you're listing the common complaints with usb-3 over usb-c peripherals, which are not a suitable replacement for PCIe. Thunderbolt is something different, more powerful & more reliable.

Re: Apple M1 Ultra

#299

Post by LegitShady »

polyrand wrote: post #258

I think the GPU claims are interesting. According to the graph's footer, the M1 Ultra was compared to an RTX 3090. If the performance/wattage claims are correct, I'm wondering if the Mac Studio could become an "affordable" personal machine learning workstation (which also won't make the electricity bill skyrocket). If Pytorch becomes stable and easy to use on Apple Silicon [0][1], it could be an appealing choice. [0]…

I always take such claims with a grain of salt anyways. It usualy on one specific benchmark. I wait for better benchmarks always instead of trusting the marketing

Re: Apple M1 Ultra

#300

Post by yurishimo »

Thaxll wrote: post #235
maronato wrote: post #101

Earlier quoted context omitted.

This hasn't been true since M1's release

A CPU like the 5900x is better than the m1 and cost 400$.

That's one part of the equation though, not to mention it's a desktop chip. I can get a laptop with an M1 Max with hours of battery life running full tilt.

Your $400 CPU needs at least another $1000 in parts just to boot (and those aren't even the parts you likely want to pair with it).

Your cost comparison is silly. Nobody compares singular CPUs to entire machines.

Post reply on HN