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

401–410 of 917 posts

Re: Apple M1 Max Geekbench Score

#401

Post by munro »

m15i wrote: post #313
munro wrote: post #274

Here's the link to the MacBookPro18,2 OpenCL benchmark: * M1 Max OpenCL https://browser.geekbench.com/v5/compute/3551790 [60,167 OpenCL Score] Comparing to my current MacBook Pro (16-inch Late 2019) & my Hetzner AX101 server: * MacBook Pro (16-inch Late 2019) vs M1 Max - CPU https://browser.geekbench.com/v5/cpu/compare/10496766?baseli... [single 163.6%, multi 188.8%] * MacBook Pro (16-inch Late 2019) vs M1 Max - Open…

Anyone know if that “Device Memory 42.7 GB” is fixed or can be increased?

The MacBook Pro M1 Max can be ordered with either 32 GB or 64 GB of unified memory. The geekbench report shows 64 GB of memory (maxed), not sure why only 42.7 GB is usable by OpenCL--so I guess we have to assume that's the max, unless there's some software fix to get it up to 64 GB.

Re: Apple M1 Max Geekbench Score

#402

Post by als0 »

sydthrowaway wrote: post #399
GeekyBear wrote: post #249

Earlier quoted context omitted.

Tick - New cores; Tock - Scaling up the number of those same cores I think most of the work went into the uncore portions of the SOC this time.

uncore?

Parts of the SoC that are not the main CPUs e.g. power management controllers, display controllers, etc.

Re: Apple M1 Max Geekbench Score

#403

Post by alexcnwy »

eecc wrote: post #393
camillomiller wrote: post #272

Earlier quoted context omitted.

You need to consider the larger target group of professionals. It's really GPU capabilities that blow everything away. If you don't plan to use your MacBook Pro for video/photo editing or 3D modeling, then a M1 Pro with the same 10-core CPU and 16-core Neural Engine has all you need and costs less. Unless I'm missing something I don't think there much added benefit from the added GPU cores in your scenario, unless yo…

Well, the MAX has double the memory bandwidth of a PRO, but I cannot see workloads other than the ones you mentioned where it would make a significant improvement. Perhaps ML but that’s all proprietarized on CUDA so it’s unlikely. Perhaps Apple could revive OpenCL from the ashes?

Pro only supports 2 external displays which is why I ordered max

Re: Apple M1 Max Geekbench Score

#404

Post by hajile »

andy_ppp wrote: post #360
hajile wrote: post #294

Earlier quoted context omitted.

1/3 the performance, but 1/10 the power. Not adding more was a mistake IMO. Maybe next time...

Really? I mean if it gets me 10-14h coding on a single charge that’s awesome…

The A15 efficiency cores will be in the next model. They are A76-level performance (flagship-level for Android from 2019-2020), but use only a tiny bit more power than the current efficiency cores.

At that point, their E-cores will have something like 80% the performance of a Zen 1 core. Zen 1 might not be the new hotness, but lots of people are perfectly fine with their Threadripper 1950X which Apple could almost match with 16 E-cores and only around 8 watts of peak power.

I suspect we'll see Apple joining ARM in three-tiered CPUs shortly. Adding a couple in-order cores just for tiny system processes that wake periodically, but don't actually do much just makes a ton of sense.

Re: Apple M1 Max Geekbench Score

#405

Post by jjcon »

DCKing wrote: post #383
jjcon wrote: post #284

Earlier quoted context omitted.

> GPU capabilities that blow everything away Compared to previous macs and igpus - an nvidia gpu will still run circles arounnd this thing

Not so sure about that "running circles around". While the M1 Max will not beat a mobile RTX 3080 (~same chip as desktop RTX 3070), Apple is in the same ballpark of its performance [1] (or is being extremely misleading in their performance claims [2]). Nvidia very likely has leading top end performance still, but "running circles around this thing" is probably not a fair description. Apple certainly has a credible cl…

Definitely impressive in terms of power efficiency if Apples benchmarks (vague as they are) come close to accurate. Comparing the few video benchmarks we are seeing from the M1Max to leading Nvidia cards I'm still seeing about 3-5x the performance across plenty of workloads (Id consider anything >2x running circles).

https://browser.geekbench.com/v5/compute/3551790

Re: Apple M1 Max Geekbench Score

#407

Post by alwillis »

shados wrote: post #255
throwaway894345 wrote: post #234

Earlier quoted context omitted.

That's Docker for Mac versus native Docker. Docker only runs on Linux, so Docker for Mac spins up a linux VM to run your containers. When you mount your Ruby or Python projects into your containers, Docker for Mac marshals tons of filesystem events over the host/guest boundary which absolutely devastates your CPU. Docker for Mac is really just bad for your use case. No idea what's going on with the thumb drive, bluet…

> Beyond that, it's a little silly to compare a desktop (presumably many times larger, ~500+W power supply, cooling system, etc) with a Mac mini (tiny, laptop chip, no fans, etc). You're totally right, but I see a lot of folks around me making absolutely bonker claims on these M1 devices. If I believed everything I'm told, I'd be expecting them to run complex After Effect rendering at 10000 frames per second.

I think that’s disingenuous. I haven’t seen any claims that haven’t been backed up by benchmarks and specific use cases.

What most people keep missing is the M1 jumped out way ahead on performance per watt on their first attempt at making their own processor.

I’ve seen the M1 Mac for less than $700. In that price range, there’s not much competition when it comes to computation, GPU performance, etc. and comes fairly close to much higher priced x86 Macs and PCs.

That’s why people are excited about the M1. You generally can’t edit 4k (and 8k) video in real-time on a machine in this price range—and what people happily paid 5x that price to get this level of performance just a few years ago.

Re: Apple M1 Max Geekbench Score

#408

Post by LegitShady »

megablast wrote: post #44
mhh__ wrote: post #36

Earlier quoted context omitted.

Impressive but unbelievably expensive considering what will actually be run on it. I kind of want one, but the 5800x for example is about 10x less expensive than a specced out MacBook Pro

Are you comparing the price of a chip to a laptop??

can't really get it any other way.

The issue to me is that because apple is the only one who can put these in computers this will have no real effect on PC component pricing. Apple makes, apple puts in apple computers, can't run windows, if you're in another ecosystem it's not even a real option.

Re: Apple M1 Max Geekbench Score

#409

Post by lostlogin »

jiveturkey wrote: post #214
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…

Probably more to do with bug sir than the hardware. Catalina is pretty solid.

I hadn’t heard that one. I had some pain, but I can laugh now.

Re: Apple M1 Max Geekbench Score

#410

Post by derefr »

jjcon wrote: post #284
camillomiller wrote: post #272

Earlier quoted context omitted.

You need to consider the larger target group of professionals. It's really GPU capabilities that blow everything away. If you don't plan to use your MacBook Pro for video/photo editing or 3D modeling, then a M1 Pro with the same 10-core CPU and 16-core Neural Engine has all you need and costs less. Unless I'm missing something I don't think there much added benefit from the added GPU cores in your scenario, unless yo…

> GPU capabilities that blow everything away Compared to previous macs and igpus - an nvidia gpu will still run circles arounnd this thing

> an nvidia gpu will still run circles arounnd this thing

Not for loading up models larger than 32GB it wouldn't. (They exist! That's what the "full-detail model of the starship Enterprise" thing in the keynote was about.)

Remember that on any computer without unified memory, you can only load a scene the size of the GPU's VRAM. No matter how much main memory you have to swap against, no matter how many GPUs you throw at the problem, no magic wand is going to let you render a single tile of a single frame if it has more texture-memory as inputs than one of your GPUs has VRAM.

Right now, consumer GPUs top out at 32GB of VRAM. The M1 Max has, in a sense, 64GB (minus OS baseline overhead) of VRAM for its GPU to use.

Of course, there is "an nvidia gpu" that can bench more than the M1 Max: the Nvidia A100 Tensor Core GPU, with 80GB of VRAM... which costs $149,000.

(And even then, I should point out that the leaked Mac Pro M1 variant is apparently 4x larger again — i.e. it's probably available in a configuration with 256GB of unified memory. That's getting close to "doing the training for GPT-3 — a 350GB model before optimization — on a single computer" territory.)

Post reply on HN