Apple is an existential threat to the PC

I'm not sure what you are trying to argue. These notebooks are specifically for content creators and no other solution support hardware encode and decode of ProRes. It's faster than the 28-core Mac Pro with the $2,000 AfterBurner card.

The graphic card alone uses much more power than the entire M1 Max and would probably not be competitive on battery power anyway. No other mobile solution offers that much frame buffer (up to 64GB) or 400GB/s bandwidth.

It isn't just the node advantage but Apple have been making low power devices for a decade now and are able to leverage that know-how. Arm is just more power efficient than x86.

The price is absolutely warranted but I get your point, you don't need it and can already find alternatives that better suit you needs, great.

Indeed, they are a great replacement for the older Intel based macs. Replacement for windows laptops? nothings really changed in that landscape.

Geekbench 5 for M1 Max is out: https://browser.geekbench.com/v5/compute/3551790
That is on par with the RTX 3050 TI mobile: https://browser.geekbench.com/v5/compute/3552028

RTX 3080 mobile in the Razerblade 14" is ~2x faster in Geekbench 5.

Aha, i understood m1max was close to a 3080mobile, or that was what apple's own benches showed. Could have been optimal benchmarks perhaps, video encoding?
Ive never really laid much attention to geekbench, but forum members here told me its the holy grail to measure performance.
 
I thought openCL wasnt used anymore in Mac's
Actually if it is, it would be useful to see what kind of device capabilities the hardware reports, e.g. hardware thread size, shared memory etc. :)

I have no idea whether Metal reports any of that stuff.
 
I thought openCL wasnt used anymore in Mac's
OpenCL was deprecated back in 2018 in favour of Metal. Like OpenGL, then framework is still there but it has not had any enhancements to support modern hardware or other APIs that would otherwise accelerate its performance. Ergo, it probably not a relevant reference point to platforms where OpenCL is actively being developed.

That said, I have to say I'm dubious about Apple's GPU claims. Like all marketeers they will have found some metric that makes the hardware look great but which may not reflect general day-to-day performance. Just look at the battery life numbers, the high numbers are for watching videos on AppleTV (the software app for macOS) which is obviously massively optimised for both efficiency and performance.

We know what the current M1 Macs can do with their 8 core GPUs, the high-end M1MAX has four times as many cores, twice the memory bandwidth (as M1Pro - unsighted on how it compares to basic M1) and it has more high-performance CPU cores to push graphics. So in real terms I'm expecting anywhere between 4x and 10x the GPU performance of today's Macs. But it could be there are considerable improvements to the Metal API and graphics stack in macOS Monterey.

ninja edit: accordingly to Anandtech it seems the M1Pro GPU is a fair leap in terms of memory bandwidth over the basic M1 so perhaps in some memory-bound scenarios the M1Pro will be considerably faster then the 8 vs 16 GPU cure jump would suggest and therefore the M1MAX represents an even bigger jump. Regardless, I'll wait until how apps are games are actually running on it following thorough testing.

¯\_(ツ)_/¯
 
Last edited by a moderator:
Apple will likely have an understanding with TSMC to outbid for any 3nm capacity coming online for consumer PC products for at least 2 years (Qualcomm&co and Intel server products can get some space to make the monopoly dealings a little less obvious, they don't need the node advantage to push iphones).

AMD hopefully getting 5nm in 2024 and reducing the node disadvantage to a single node will be an improvement on the current situation, but the market is busted.

I thought Zen 4 was going to be on 5nm for 2022?

It is 5nm and 2022. Ridiculous to even think AMD would "hopefully" get 5nm by 2024 or that Apple would have exclusive or near exclusive rights to process for 2 years
 
It is 5nm and 2022. Ridiculous to even think AMD would "hopefully" get 5nm by 2024 or that Apple would have exclusive or near exclusive rights to process for 2 years

Given Apple have largely been waiting on Intel for significant performance/process/cost improvements before deploying them to Mac hardware, there is now also a greater expectation that Apple will be deploying updates to ARM-based Macs at a higher pace - because this is also be driven to need to advance traditional yearly-released devices like iPhone.

Macs could see updated perhaps annually. Apple are in transition right now, which requires a whole lot more effort than swapping one architecture CPU/GPU for a faster one, but once they've migrated to ARM completely, I wouldn't be surprised if Apple are updating Macs are a much higher cadence than Intel or PowerPC hardware, probably some on alternating six-month timetable (alternating iMac/MacMini/Mac Pro and Macbook/Pro/Air) which feels logical from an engineering focus.

There is clearly more performance leaps to be eked out of ARM compared to 80x86/x64 unless Intel are keeping some amazing R&D under wraps. Not impossible but also not expected.
 
It is 5nm and 2022. Ridiculous to even think AMD would "hopefully" get 5nm by 2024 or that Apple would have exclusive or near exclusive rights to process for 2 years

It's already known Intel won't do anything consumer on 3nm next year. Apple will have had a lock on N7+ period and ~2 years on 5nm. So yeah, it's not unlikely that for 2 years for consumer PCs there will be nothing on 3nm. It would just be a continuance of the two last nodes.

Just overpaying and doing the iPhone on 3nm even though it's not really necessary for their mobile market share, just to prevent competitors getting 3nm space in this extremely important period to profile the M1x would be a 100 billion dollar bargain for Apple. I could easily see them doing it.
 
Last edited:
Do you have a link for the best, most-detailed, description of these new chips?

There's about 23B transistors extra in M1 Max (57B) versus M1 Pro (33.7B) and it seems there are two key differences:
  1. GPU cores: 32 versus 16
  2. Off-die memory system: 400GB/s versus 200GB/s
So how does Apple spend so many transistors? 23B transistors is equivalent to pretty much all the transistors in Navi 21 (26.8B). All of 6700XT is only 17.2B transistors.

Looking at the die shot it appears that the NPU is doubled as well on the M1 Max.

So we have the GPU, display engine, media engine, neural engine, the extra 256-bit memory interface as well as the system level cache.

M1MAX.jpg
 
Looking at the die shot it appears that the NPU is doubled as well on the M1 Max.

So we have the GPU, display engine, media engine, neural engine, the extra 256-bit memory interface as well as the system level cache.
Good stuff.

Why would the display engine change (or why would there be more of it)? Similar question for media engine?
 
Good stuff.

Why would the display engine change (or why would there be more of it)? Similar question for media engine?
Monitor support is vastly extended. (Apple demonstrated Laptop screen (7MP) + 3x6k monitors + 4k television)
The media engine has hardware support for ProRes video - presenter lady said 30 simultaneous 4k streams or 7 simultaneous 8k streams, outperforming a top end (28-core Xeon) Mac Pro with Apples own $2000 Afterburner card.
 
Monitor support is vastly extended. (Apple demonstrated Laptop screen (7MP) + 3x6k monitors + 4k television)
The media engine has hardware support for ProRes video - presenter lady said 30 simultaneous 4k streams or 7 simultaneous 8k streams, outperforming a top end (28-core Xeon) Mac Pro with Apples own $2000 Afterburner card.

Which makes you wonder what the inevitable ARM Mac Pro will be capable of.
 
I notice not many games from 2019 onwards on this list
https://en.wikipedia.org/wiki/Metal_(API)
why is this? Sure gaming is very small on Mac, but is this the only reason?

Hmmm I couldnt find any benchmarks with metal in actual games, I could only find rosetta

The fact that you can boot Window on MacIntel have probably limitless the need, rentability of a native portage on Metal.
Only games, targetting iOs, need a Metal port. Now it's a different situation.
 
Yes you can boot into windows, but on the new M1 processors surely this is running through an emulator thus slowing everything down with that extra layer, Thus it won't give you the best gaming experience. Or am I wrong?
 
Yes you can boot into windows, but on the new M1 processors surely this is running through an emulator thus slowing everything down with that extra layer, Thus it won't give you the best gaming experience. Or am I wrong?
Actually you can’t boot into Windows anymore, Windows on ARM doesn’t support Mac hardware.
While you can convince the Apple Silicon macs to run Windows, (tenacious software guys can perform miracles) it’s not a very meaningful way to access the Windows game library IMHO. Buying a Windows box makes a ton more sense, if that is what you want. Boot Camp is dead.

Speaking as a geek - the M1 Pro Max is BY FAR the most powerful TBDR GPU ever produced. I would love to see graphics coded to take advantage of it!
 
Actually you can’t boot into Windows anymore, Windows on ARM doesn’t support Mac hardware.
While you can convince the Apple Silicon macs to run Windows, (tenacious software guys can perform miracles) it’s not a very meaningful way to access the Windows game library IMHO. Buying a Windows box makes a ton more sense, if that is what you want. Boot Camp is dead.

Speaking as a geek - the M1 Pro Max is BY FAR the most powerful TBDR GPU ever produced. I would love to see graphics coded to take advantage of it!

It’s entirely up to Microsoft to enable Windows for Arm as the Apple Silicon boot loader is not locked.

They have made great strides with Linux so far.
 
Back
Top