Apple is an existential threat to the PC

spec2006_A14.png


That's at 5 Watt.
 
Yeah but I want to see real world perf and not a synthetic bench. I want to see how fast can run games, how fast can use PS, or render a video, etc.
 
There made so many points of higher CPU and GPU performance throughout. 3X here, 6X here, 2X here. Did you stop watching after the first 20 minutes?

If I'm not mistaken those comparisons were a fully enabled M1 against the base model dual core Macbook Air and quad core Macbook pro from 2018. So not exactly an apples to apples comparison. The performance is there yes, but 5x and 6x outside of ML perf is a bit of a stretch. Tigerlake and Renoir have certainly increased CPU and GPU performance substantially since 2018.
 
Im more interested in seeing the benchmarks for these than the recent Zen3 ones.
Esp the fanless laptop.
They stuck a fan in the mac mini :sly: God I used to have one, It was soooooo loud. Same as my recent Intel NUC, I had to downclock that to 80% speed otherwise its sounds like a vacuum cleaner
 
No eGPU support for M1: https://techcrunch.com/2020/11/10/macs-with-the-m1-chip-do-not-support-egpus/

What is also intrestin, they say only 8 GPU Cores. No Shader count no clock speed and the fans are happy? Realy!?

Your first Apple event?

Yeah but I want to see real world perf and not a synthetic bench. I want to see how fast can run games, how fast can use PS, or render a video, etc.

And I want a date with Charlize Theron.

I hope there will be better reviews than "how cool M1 is, it has a 16-core Neural Engine" reviews. Reviews with Cinebench results, Handbrake etc.
 
What is also intrestin, they say only 8 GPU Cores. No Shader count no clock speed and the fans are happy? Realy!?
This is what we have for now, summary taken from Eug:
8-core iGPU (but there is a 7-core variant, likely with one inactive core)
128 execution units
Up to 24576 concurrent threads
2.6 Teraflops
82 Gigatexels/s
41 Gigapixels/s
Also: 128-bit LPDDR4x (from Andrei at Anandtech)

They will reach users within a week, so if interested you’ll see a plethora of benchmarks very soon. Most of the cross platform ones will run under Rosetta 2 initially.
 
And MacBook Pro 13" too. I guess they had to flip the 13" otherwise - based not the numbers they gave -the 13" Air would have murdered the 13" MBP in performance. Hardware shipping next week. It'll be interesting to see how fast they are once they're out there. I imagine compatible benchmarks will be rare but it'll have to be application run benchmarks for a while.

I have ignored it, the whole event was so boring.

"Look, here is the M1, it has 8 CPU cores, 8-core GPU and 16-core Neural Engine."
"Look, here is MBA with M1, it has 8 CPU cores, 7-8-core GPU and 16-core Neural Engine."
"Look, here is Mm with M1, it has 8 CPU cores, 8-core GPU and 16-core Neural Engine and a fan."
"Look, here is MBP with M1, it has 8 CPU cores, 8-core GPU and 16-core Neural Engine and a fan."
 
Im more interested in seeing the benchmarks for these than the recent Zen3 ones.
Esp the fanless laptop.
They stuck a fan in the mac mini :sly: God I used to have one, It was soooooo loud. Same as my recent Intel NUC, I had to downclock that to 80% speed otherwise its sounds like a vacuum cleaner
The latest Mac Minis use a 65W desktop part from intel. This one will likely draw 10W or so in the Mini. The fan ensures it won’t thermally throttle.

Quick Edit: The MBP 13” has a 20% larger battery and 10-15% longer reported runtime (20h !!), implying that the fans aren’t used to raise clocks a lot.
 
Last edited:
The latest Mac Minis use a 65W desktop part from intel. This one will likely draw 10W or so in the Mini. The fan ensures it won’t thermally throttle.

Quick Edit: The MBP 13” has a 20% larger battery and 10-15% longer reported runtime (20h !!), implying that the fans aren’t used to raise clocks a lot.
Too early to tell, the battery life figures reflect what you’d get if workloads are dominated by stuff like browsing and video playback which would run mostly on the efficiency cores.

However, I do agree the clock difference, if there is one, will be minimal. Actually, I fully expect the clocks to be identical with the difference being simply that the MBA will throttle and the others won’t.
 
The latest Mac Minis use a 65W desktop part from intel. This one will likely draw 10W or so in the Mini. The fan ensures it won’t thermally throttle.

Quick Edit: The MBP 13” has a 20% larger battery and 10-15% longer reported runtime (20h !!), implying that the fans aren’t used to raise clocks a lot.

The battery life figures could be all with low power cores enabled only (as in "web browsing" and "video viewing"), so it probably won't tell us much about the clock rates of the high performance cores.
 
Rosetta 2 ? Do understand that this can trigger recompilations for CPU code just as we see for GPU code (shader recompilations) ?
 

I had posted this in the A14 thread but I expected as much. The M1 may also end up being used in the next iPad Pro update. The area and power cost for a PCIE 4.0 x8 connection for an eGPU which would be used only in a small number of M1 devices was probably not worth it. Another reason could have been driver maturity for the eGPU as this would be on AMD, who may not be inclined to invest much into the ARM ecosystem.
Your first Apple event?

Yea, just par for the course for Apple events (and Apple in general).
I hope there will be better reviews than "how cool M1 is, it has a 16-core Neural Engine" reviews. Reviews with Cinebench results, Handbrake etc.

I'd expect Anandtech at least to go in depth into the performance and power efficiency claims. When they finally get around to it of course (Still waiting on that Ampere review...)
I have ignored it, the whole event was so boring.

"Look, here is the M1, it has 8 CPU cores, 8-core GPU and 16-core Neural Engine."
"Look, here is MBA with M1, it has 8 CPU cores, 7-8-core GPU and 16-core Neural Engine."
"Look, here is Mm with M1, it has 8 CPU cores, 8-core GPU and 16-core Neural Engine and a fan."
"Look, here is MBP with M1, it has 8 CPU cores, 8-core GPU and 16-core Neural Engine and a fan."

Haha yeah those were pretty much my thoughts exactly!
The latest Mac Minis use a 65W desktop part from intel. This one will likely draw 10W or so in the Mini. The fan ensures it won’t thermally throttle.

Quick Edit: The MBP 13” has a 20% larger battery and 10-15% longer reported runtime (20h !!), implying that the fans aren’t used to raise clocks a lot.

Not necessarily. Simply keeping a chip cooler does reduce its power consumption. However, in this case I expect you're probably right. Given the use case, I expect the fan on the MBP is more to ensure that it can sustain higher clocks/not throttle like the MBA for more demanding workloads. Web browsing and video playback are not really demanding on the CPU and should be running more on the efficiency cores. As such the battery life should pretty much scale with battery capacity (and screen brightness obviously).
 
I agree that we need to see actual hardware reviews to determine exactly how these Macs perform, particularly under load. For those interested in graphics programming, here is the new developer video for the M1:
Dev video.
It includes a sneak peek at Baldurs Gate 3. :D
 
It's fun to read the fooLnotes on Apple's homepage, https://www.apple.com/macbook-pro-13/

  1. Testing conducted by Apple in October 2020 using preproduction 13-inch MacBook Pro systems with Apple M1 chip, as well as production 1.7GHz quad-core Intel Core i7-based 13-inch MacBook Pro systems with Intel Iris Plus Graphics 645, all configured with 16GB RAM and 2TB SSD. Tested with Unity 2018.3.6f1 using Book of the Dead demo, at 1440x900 resolution. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Pro.
  2. Testing conducted by Apple in October 2020 using preproduction 13-inch MacBook Pro systems with Apple M1 chip and 16GB of RAM using select industry-standard benchmarks. Comparison made against the highest-performing integrated GPUs for notebooks and desktops commercially available at the time of testing. Integrated GPU is defined as a GPU located on a monolithic silicon die along with a CPU and memory controller, behind a unified memory subsystem. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Pro.
  3. Testing conducted by Apple in October 2020 using preproduction 13-inch MacBook Pro systems with Apple M1 chip, as well as production 1.7GHz quad-core Intel Core i7-based 13-inch MacBook Pro systems with Intel Iris Plus Graphics 645, all configured with 16GB RAM and 2TB SSD. Tested with prerelease Shapr3D 3.45.0 using a 288.2MB model. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Pro.
  4. Testing conducted by Apple in October 2020 using preproduction 13-inch MacBook Pro systems with Apple M1 chip, as well as production 1.7GHz quad-core Intel Core i7-based 13-inch MacBook Pro systems with Intel Iris Plus Graphics 645, all configured with 16GB RAM and 2TB SSD. Tested with Shadow of the Tomb Raider 1.0.1 using the built-in benchmark, at 1440x900 resolution, with medium settings. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Pro.

The entry level MBP had the Iris Plus Graphics 645 - MBP 13'' with Intel Core i5-8257U (Magic Keyboard), https://en.wikipedia.org/wiki/MacBook_Pro#Technical_specifications_5

So if the M1 GPU is so powerfull, why didn't Apple compare it with Ice Lake based MBP 13'' (Iris Plus Graphics G7, LPDDR4X-3733)?
 
I had posted this in the A14 thread but I expected as much. The M1 may also end up being used in the next iPad Pro update. The area and power cost for a PCIE 4.0 x8 connection for an eGPU which would be used only in a small number of M1 devices was probably not worth it. Another reason could have been driver maturity for the eGPU as this would be on AMD, who may not be inclined to invest much into the ARM ecosystem.

AMD does not develop or maintain the drivers for macOS. Apple takes responsibility for those things ...
 
Apple develops the drivers for AMD's mobile GPUs on MacOS?

They do indeed and it's also part of the reason why they dropped support for Nvidia HW on macOS since they couldn't get good control of the software stack from them. AMD and Intel have open source drivers and documentation where Apple driver engineers can get their information from and write metal drivers for them ...

I imagine Apple not offering eGPU support is out pettiness on their behalf rather than AMD being involved. They probably don't want to maintain drivers for another vendor so they don't like the idea that their effort was benefitting another corporation seeing cash flowing to them at their end. Apple still doesn't control AMD's HW stack so they are attempting to solve this issue by replacing them with their own in-house designed GPUs ...
 
AMD does not develop or maintain the drivers for macOS. Apple takes responsibility for those things ...
I am not sure if this is entirely true. Yes, it is bundled with macOS, and AMD offers no standalone distribution. Perhaps the part interfacing with Darwin & IOKit is indeed developed by Apple engineers. (Or potentially on-site/dedicated AMD support engineers, not surprising for huge accounts like Apple or Tier 1 OEMs)

But there seem to be many portions of it being contributed by AMD’s cross-platform driver codebase — codename leaks regularly coming out of macOS that are irrelevant to Mac were a clue, so as the known existence of cross-platform components like the DC/DAL in amdgpu.
 
Last edited:
Back
Top