Apple’s new M1 Pro and M1 Max processors • Hacker News Board

Re: Apple’s new M1 Pro and M1 Max processors

#301

Post by azeirah » Mon, Oct 18, 2021, 7:35 PM UTC

mattfrommars wrote: post #272

I am in the market for a new laptop and bit skeptic for M1 Chips. Could anyone please tell me how is this not a "premium high performance Chromebook" ? Why should I buy this and not Dell XPS machine if I will be using it for web development/Android Development/C#/DevOps. Might soon mess with machine learning

For web dev it works very well, android development no clue, c# is primarily a windows-oriented development environment so probably not so great. For devops, well.. Mac is a linux-esque environment, it'll be fine?

I have the m1 air and the big advantages are the absurd battery life and the extremely consistent very high performance.

It's always cold, it's always charged, it's always fast.

I believe you can get both faster cpu and gpu performance on a laptop, but it costs a lot in battery life and heat which has a bigger impact on usability than I believed before getting this one.

Might want to add, this is my first ever apple laptop. I've always used Lenovo laptops, ran mostly windows/linux dual boots on my laptops and desktops over the years.

Re: Apple’s new M1 Pro and M1 Max processors

#302

Post by sharkjacobs » Mon, Oct 18, 2021, 7:35 PM UTC

This is roughly in line with what I expected, given the characteristics of the M1. It's still very power efficient and cool, has more CPU cores, a lot more GPU cores, wider memory controller, and presumably it has unchanged single core performance.

Apple clearly doesn't mean these to be a high performance desktop offering though because they didn't even offer an Mac Mini SKU with the new M1s.

But what I'm really curious about is how Apple is going to push this architecture for their pro desktop machines. Is there a version of the M1 which can take advantage of a permanent power supply and decent air flow?

Re: Apple’s new M1 Pro and M1 Max processors

#303

Post by PragmaticPulp » Mon, Oct 18, 2021, 7:35 PM UTC

slayerjain wrote: post #68

Based on the numbers it looks like the M1 Max is in the RTX 3070-3080 performance territory. Sounds like mobile AAA gaming has potential to reach new heights :D

> Based on the numbers it looks like the M1 Max is in the RTX 3070-3080 performance territory.

The slides are comparing to a laptop with a 3080 Mobile, which is not the same as a normal RTX 3080. A desktop 3080 is a power hungry beast and will not work in a laptop form factor.

The 3080 Mobile is still very fast as a benchmark, but the full-size 3080 Desktop is in another league of performance: https://www.notebookcheck.net/The-mobile-GeForce-RTX-3080-is...

Still very impressive GPU from Apple!

Re: Apple’s new M1 Pro and M1 Max processors

#304

Post by klelatti » Mon, Oct 18, 2021, 7:36 PM UTC

mohanmcgeek wrote: post #124

Disingenuous for Apple to compare these against 2017 Intel chips and call them 2x and 3.7x faster. I would love to see how they fare against 2021 Intel and amd chips.

How is it disingenuous - defined in my dictionary as not candid - when we know precisely which chips they are comparing against?

They are giving Mac laptop users information to try to persuade them to upgrade from their 2017 MacBook Pros and this is probably the most relevant comparison.

Re: Apple’s new M1 Pro and M1 Max processors

#305

Post by rdw » Mon, Oct 18, 2021, 7:36 PM UTC

Unklejoe wrote: post #211

kergonath wrote: post #192

Earlier quoted context omitted.

> The DDR being closer to the core may or may not allow the memory to run at higher speeds due to better signal integrity, but you can purchase DDR4-5333 today whereas the M1 uses 4266. My understanding is that bringing the RAM closer increases the bandwidth (better latency and larger buses), not necessarily the speed of the RAM dies. Also, if I am not mistaken, the RAM in the new M1s is LP-DDR5 (I read that, but it…

The overall bandwidth isn't affected much by the distance alone. Latency, yes, in the sense that the signal literally has to travel further, but that difference is miniscule (like 1/10th of a nanosecond) compared to overall DDR access latencies. Better signal integrity could allow for larger busses, but I don't think this is actually a single 512 bit bus. I think it's multiple channels of smaller busses (32 or 64 bit…

Your comment got me thinking, and I checked the math. It turns out that light takes ~0.2 ns to travel 2 inches. But the speed of signal propagation in copper is ~0.6 c, so that takes it up to 0.3 ns. So, still pretty small compared to the overall latencies (~13-18 ns for DDR5) but it's not negligible.

I do wonder if there are nonlinearities that come in to play when it comes to these bottlenecks. Yes, by moving the RAM closer it's only reducing the latency by 0.2 ns. But, it's also taking 1/3rd of the time that it used to, and maybe they can use that extra time to do 2 or 3 transactions instead. Latency and bandwidth are inversely related, after all!

Re: Apple’s new M1 Pro and M1 Max processors

#306

Post by baybal2 » Mon, Oct 18, 2021, 7:37 PM UTC

azinman2 wrote: post #215

baybal2 wrote: post #119

Earlier quoted context omitted.

Actually no "extra" chip area in comparison to x86 based solution. They just throw away so much of cruft from the die like PCIE PHYs, and x86 legacy I/O with large area analog circuitry. Redundant complex DMA, and memory controller IPs are also thrown away. Clock, and power rails on the SoC are also probably taking less space because of more shared circuitry. Same with self-test, debug, fusing blocks, and other small…

This is very interesting and first time I’ve heard / thought about this. Wonder how much power efficiency comes from exactly these things?

PCI is quite power hungry when it works on full trottle.

The seemed power efficiency when PCIE was going 1.0 2.0 3.0 ... was due to dynamic power control, and link sleep.

On top of it, they simply don't haul memory nonstop over PCIE anymore, since data going to/from GPU is simply not moving anywhere.

Re: Apple’s new M1 Pro and M1 Max processors

#307

Post by mrtksn » Mon, Oct 18, 2021, 7:37 PM UTC

ChuckNorris89 wrote: post #264

mrtksn wrote: post #241

Earlier quoted context omitted.

Just today I got one 6s and one iPhone 7 screen repaired(6s got the glass replaced, the 7 got full assembly replaced) and battery of the 6s replaced at a shop that is not authorized by Apple. It cost me 110$ in total. Previously I got 2017 Macbook Air SSD upgraded using an SSD and an adapter that I ordered from Amazon. What’s that narrative that Apple devices are not upgradable or repairable? It simply not true. If a…

>Just today I got one 6s and one iPhone 7 screen repaired Nice, except doing a screen replacement on a modern iPhone like the 13 series will disable your FaceID making your iPhone pretty much worthless. >Previously I got 2017 Macbook Air SSD upgraded using an SSD and an adapter that I ordered from Amazon Nice, but on the modern Macbooks, the SSD is soldered and not replaceable. There is no way to upgrade them or repl…

Hardware people are crafty, they find ways to transfer and combine working parts. The glass replacement(keeping the original LCD) I got for the 6S is not a procedure provided by Apple. Guess who doesn’t care? The repair shop that bought a machine from China for separating and re-assembly of the Glass and LCD.

Screen replacement is 50$, glass replacement is 30$.

iPhone 13 is very new, give it a few years and the hardware people will leverage the desire of not spending 1000$ on a new phone when the current one works fine except for that broken part.

Re: Apple’s new M1 Pro and M1 Max processors

#308

Post by 19h » Mon, Oct 18, 2021, 7:37 PM UTC

btzo wrote: post #277

twic wrote: post #268

Earlier quoted context omitted.

The M1 Pro is faster. The M1 MAX is an M1, but with new control software which makes it keep crashing.

wrong

I assume it was a reference to the 737 Max..

Re: Apple’s new M1 Pro and M1 Max processors

#309

Post by willis936 » Mon, Oct 18, 2021, 7:38 PM UTC

sunaurus wrote: post #35

I would like to upgrade to Apple silicon, but I have no need for a laptop. I hope they put the M1 Max in a Mac Mini soon.

There were rumors that it was supposed to be today. Given that it wasn't, I now expect it to be quite a while before they do. I was really looking forward to it.

Re: Apple’s new M1 Pro and M1 Max processors

#310

Post by tehnub » Mon, Oct 18, 2021, 7:39 PM UTC

sk0g wrote: post #115

Now, if only Unreal Engine builds were available M1 native, I could get rid of my huge and heavy desktop entirely! Interestingly, some improvements to Rosetta were mentioned, extremely briefly.

Same wish here. Last I tried a few months ago, I was unable to compile UE4 as well. These machines would be great for UE4 dev if only the compatibility was there. I wonder if the politics between Epic and Apple has made efforts in this area a lower priority.