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

Apple M1 Max Geekbench Score

browser.geekbench.com

Post reply on HN
Print view View original

771–780 of 917 posts

Re: Apple M1 Max Geekbench Score

#771

Post by throwaway894345 »

chrisjc wrote: post #727
musicale wrote: post #606

Earlier quoted context omitted.

I wonder how well ARM64 is supported in the Docker community? Presumably x86 Docker images running in emulation would be even slower than ARM64 images running in a Linux VM. Though as you note they're completely different classes of hardware (500W vs. 26.5W), a more Apples-to-Apples (so to speak) comparison would be native Ubuntu on both systems, though as I understand it Ubuntu for M1 is not complete and may lack GP…

I don't think chip architecture was the main issue that the parent was alluding to. The same issues would arise if it was Docker running an x86 image on an x86 chip on a Mac. Docker doesn't run natively on OSX, unlike Linux where it does. As a result, Docker Desktop for Mac spins up a linux VM, and this is where penalties begin to be payed. I do wonder if Docker Desktop is smart enough to tailor the VM's architecture…

> I wonder how difficult it would be to support docker natively on OSX (BSD?) or if Apple has any intention on providing that support?

I think it would be pretty difficult. Mac would effectively have to provide a fully compatible Linux interface in its kernel. It would probably be more likely to build a better fs-event bridge to guest systems, but considering MacOS still doesn’t have ext4 support, I’m not counting on it.

Re: Apple M1 Max Geekbench Score

#772

Post by klelatti »

gsnedders wrote: post #764
klelatti wrote: post #737

Earlier quoted context omitted.

Frankly don't believe this. Who do you give majority of leading node capacity to? 1. Biggest long term partner. 2. Someone who competes with you as a manufacturer. Not a hard decision!

TSMC had long said they were happy to have Intel as a larger customer, including on leading edge nodes… if there was long-term commitment from Intel.

Of course but that’s not the same as giving them the majority of the leading edge at the expense of Apple which is what the article implies.

Re: Apple M1 Max Geekbench Score

#773

Post by hajile »

pja wrote: post #714
hajile wrote: post #309

Earlier quoted context omitted.

M1 only uses around 15-18w with all 8 cores active according to Anandtech's review (with E-cores using 1/10 the power, that's equivalent to just a little less than 4.5 big cores. I'd guess 30w for all cores would be the upper-end power limit for the cores. Intel's "TDP" is a suggestion rather than reality. Their chips often hit 60-80 watts of peak power before hitting thermals and being forced to dial back.

Anandtech measured the max TDP of the CPUs in the M1 as being around 28W, with the GPU adding ~17W to that figure. https://www.anandtech.com/show/16252/mac-mini-apple-m1-teste...

That figure is compute workloads.

Also, those measurements are “at the wall” and include RAM, SSD, and everything else.

At the very least, you must subtract most idle power, but you also need to subtract more when DRAM/SSD kicks up from low power state.

Re: Apple M1 Max Geekbench Score

#774

Post by thom »

cschmid wrote: post #306
m15i wrote: post #179

How will these chips do for training neural nets? The 64 GB RAM would be awesome for larger models, so I’m willing to sacrifice some training speed.

I know this doesn't answer your question, but can I just ask (out of curiosity) why you're training ML models on a laptop?

Is this much different than asking why someone would do _any_ work on a laptop?

Re: Apple M1 Max Geekbench Score

#775

Post by ody4242 »

blakesterz wrote: post #100

I just can't figure out what I'm missing on the "M1 is so fast" side of things. For years I worked* on an Ubuntu desktop machine I built myself. Early this year I switched to a brand new M1 mini and this this is slower and less reliable than the thing I built myself that runs Ubuntu. My Ubuntu machine had a few little issues every no and then. My Mini has weird bugs all the time. e.g. Green Screen Crashes when I have…

As someone who has an Intel NUC w. Ubuntu, a Thinkpad with Windows 10, and a Macbook Pro 2015, I have a totally different experience. I spend like 5% of my total effort on OSX to make things working, the remaining 95% is spent to make things working on Windows and Ubuntu...

Re: Apple M1 Max Geekbench Score

#776

Post by PolCPP »

PragmaticPulp wrote: post #245
blakesterz wrote: post #100

I just can't figure out what I'm missing on the "M1 is so fast" side of things. For years I worked* on an Ubuntu desktop machine I built myself. Early this year I switched to a brand new M1 mini and this this is slower and less reliable than the thing I built myself that runs Ubuntu. My Ubuntu machine had a few little issues every no and then. My Mini has weird bugs all the time. e.g. Green Screen Crashes when I have…

> I just can't figure out what I'm missing on the "M1 is so fast" side of things. Two reasons: 1. M1 is a super fast laptop chip. It provides mid-range desktop performance in a laptop form factor with mostly fanless operation. No matter how you look at it, that's impressive. 2. Apple really dragged their feet on updating the old Intel Macs before the transition. People in the Mac world (excluding hackintosh) were stu…

From what i see the Clang difference mostly looks like its because it is a 10 core cpu (with 2 efficiency ones at that), vs a 16 core. (I'm seeing 104 klines/sec vs 180.5), the 5800x seems to hit 94-96klines/sec

Kinda wish they had a m1 max focused on extra cpu cores rather than gpu cores tbh

Re: Apple M1 Max Geekbench Score

#777

Post by chaoskanzlerin »

Mindwipe wrote: post #751
lanza wrote: post #558

Earlier quoted context omitted.

You're missing out on the fact that Apple didn't release a 12980hk or 5980hx competitor. These are 30 watt chips that trounce the competition's 65 watt (e.g. the 12980hk and 5980hx) and beyond chips. Hell, this Geekbench is faster than a desktop 125 watt 11900k. It's faster than a desktop 105 watt 5800x. Apple intentionally played to the competition here. They know AMD/Intel reach some performance level X and release…

> It might be a long time before it makes sense to buy a non-Apple laptop. A laptop that can do some light gaming is not a niche requirement, and ultimately Apple decided to completely turn it's back on that market with the ARM transition.

The architecture isn't to blame; it's the 32-bit stuff that breaks a fair amount of Steam games, and Vulkan/graphics engines lacking, both of which are...Apple being Apple

Re: Apple M1 Max Geekbench Score

#779

Post by hajile »

O_H_E wrote: post #566
hajile wrote: post #288

Earlier quoted context omitted.

M1 uses TSMC high-density rather than high-performance. They get 40-60% better transistor density and less leakage (power consumption) at the expense of lower clockspeeds. Also, a core is not necessarily just limited by power. There are often other considerations like pipeline length that affect final target clocks. The fact is that at 3.2GHz, the M1 is very close to a 5800X in single-core performance. When that 5800…

> M1 uses TSMC high-density rather than high-performance. Is that an actual product differentiation from TSMC? Or just observational + the fact that it's 5nm. I'm actually curious from a chip manufacturing standpoint.

There’s actually three major version on N3 (sorry, I don’t have time to look up the similar articles on N5/7, but I think wikichip has a couple).

https://www.techdesignforums.com/blog/2021/06/03/three-libra...

I remember seeing a picture of a partner list for a TSMC node with a couple dozen libraries that not only changed based on density, but also on type of chip being built.

Re: Apple M1 Max Geekbench Score

#780

Post by vijucat »

adrianN wrote: post #576
lanza wrote: post #558

Earlier quoted context omitted.

You're missing out on the fact that Apple didn't release a 12980hk or 5980hx competitor. These are 30 watt chips that trounce the competition's 65 watt (e.g. the 12980hk and 5980hx) and beyond chips. Hell, this Geekbench is faster than a desktop 125 watt 11900k. It's faster than a desktop 105 watt 5800x. Apple intentionally played to the competition here. They know AMD/Intel reach some performance level X and release…

> It might be a long time before it makes sense to buy a non-Apple laptop. ...if you only care about the things that Apple laptops are good at. Almost nobody needs a top-of-the-line laptop to do their tasks. Most things that people want to do with computers can be done on a machine that is five to ten years old without any trouble at all. For example I use a ThinkPad T460p, and while the geekbench scores for its proc…

> Battery life, screen quality, and ergonomics are the only thing most consumers need to care about.

You, Sir, are a very utilitarian consumer. Most consumers care about being in the "In" crowd, i.e., owning the brand that others think is cool. Ideally, it comes in a shiny color. That's it. The exact details are just camouflage.

Post reply on HN