5 ms·

> If Apple is ready to switch to ARM they must have some impressive CPU. Apples not usually one to dabble here and there, so if they switch it’s going to be a w

> If Apple is ready to switch to ARM they must have some impressive CPU. Apples not usually one to dabble here and there, so if they switch it’s going to be a wholesale ordeal. Apple has an ARM-based product on sale today that has a better screen, better battery life, and better performance, than the 2020 Macbook air for many workloads, including compiling and doing development work (it crushed the air in the benchmark sets that can be run there). That product is the 2020 ipad pro, and that is why there are so many posts about people actually trying to turn them into development machines. This product is one step away from being a real laptop: it is missing real MacOSX. The performance of the Intel-based Macbook Air has increased by 1.8-1.9x from 2012 to 2020 (that's 8 years). Apple already has internal A13 prototypes, and A14 is probably going to the prototype phase right now. I can't imagine the numbers working in favor of Intel. By 2021 Intel can probably deliver a 1.1-1.2x speed up tops for Macbook airs, while Apple can probably deliver a 2x speed up for 2021 with the A13 and another 2x one for 2023 with the A14. Being able to scale the same chip for iphones, ipads, and macbooks, reusing internal resources, and avoiding the "hassle" of having to deal with Intel. The only thing that's IMO in the air is what is going to be the discrete graphics story for macbook pro's and Mac pros. Its unclear whether it would make sense for AMD to deliver discrete gfx products that interface well with ARM, and the Apple-nvidia bridge burned long time ago. So I wonder whether Apple has openings for electrical engineers to work on discrete graphics verification & design, and driver developers. It would strongly hint that apple would design their own discrete gfx in house.

ianai 6y ago
It’s even got a keyboard stand case that almost turns it into a miniature iMac-like device.
- oneplane 6y ago
  Imagine a teeny tiny iMac, that'd be a fun look. I wonder if the size vs. needs has changed in general as the adoption of large screens seems to have slowed down a bit in my environment (probably because everyone that wanted one now has one).
ajross 6y ago
The A12 is clearly at parity now vs. Intel's existing mobile offerings, probably somewhat ahead given the long delays with 10nm. The question upthread is whether or not switching architectures makes financial sense, not whether it's a (mild) technical win. Switching to their own chips cuts Intel out of the loop, but as far as business risk that simply replaces one single source manufacturer with another (TSMC). It probably saves money per-part, which is good. But then Apple is still drowning in cash and immediate term savings really aren't much of a motivator. > By 2021 Intel can probably deliver a 1.1-1.2x speed up tops for Macbook airs, while Apple can probably deliver a 2x speed up for 2021 with the A15 and another one for 2023 with the A16. That's going to need some citation. Moore's law is ending for everyone, not just Intel. TSMC has pulled ahead of Intel (and Samsung has caught up) for sure, but progress is slowing. That kind of scaling just isn't going to happen for anyone.
- fluffything 6y ago
  > That's going to need some citation. So this is Geekbench 5 showing the difference between a 2012 macbook air i5 and a 2020 macbook air i5 - both base models and similar price: https://browser.geekbench.com/v5/cpu/compare/2613713?baseline=2613509 https://browser.geekbench.com/v5/cpu/compare/2613713?baselin... And this is a thread in the rust subreddit about compilation speed on macbooks where some users report the performance increase for different generations of macbook pros and macbook airs, if you want a more "realistic benchmark" to calibrate geekbench results: https://www.reddit.com/r/rust/comments/gypajc/macbook_pro_2020_compile_performance/ https://www.reddit.com/r/rust/comments/gypajc/macbook_pro_20... This is the ipad pro 2020 crushing the macbook air 2020: https://browser.geekbench.com/v5/cpu/compare/2612714?baseline=2613509 https://browser.geekbench.com/v5/cpu/compare/2612714?baselin... And this is the improvement from the previous generation ipad A10 to the ipad pro's A12Z - 2x speed up in single a generation: https://browser.geekbench.com/v5/cpu/compare/2613991?baseline=2613301 https://browser.geekbench.com/v5/cpu/compare/2613991?baselin... You are definetely right that Moore's law is hitting Intel hard. But AMD is still doing quite well, nvidia and "ati" are doing incredibly well, and Apple chips have been doing extremely well over the last couple generations. Maybe you are right, and Apple won't be able to deliver 2x speed ups in the next 2 generations. I'd expect that, just like for Intel, things won't abruptly change from one gen to another, but for this to happen over a longer period of time. Right now, only apple knows what perf their next 2 gens of chips are expected to deliver. The only thing we know is that Apple ARM chips are crushing their previous generation both for ipads and iphones year after year, and now they are betting on them for macbooks, and potentially mac pros, probably for at least the next 10-15 years.
zozbot234 6y ago
Discrete graphics are not dependent on the processor instruction set, they just need to interface with the PCI bus. All AMD has to do is deliver their cards, the drivers are FLOSS already.
- raverbashing 6y ago
  I assume the drivers might have a lot of specific x86 optimizations, so it's easy (but not too easy) I'm not sure how's the status of PCI-X on Arm machines at the moment as well. Do PCs still have the North and South bridge to interface to multiple devices?
  - kcb 6y ago
    All CPUs are pretty much SOCs now. The chipsets are basically just fancy PCI-E switches with peripherals attached like USB and SATA controllers.
bitL 6y ago
> The performance of the Intel-based Macbook Air has increased by 1.8-1.9x from 2012 to 2020 That's not really correct, the latest Air can be configured with a quad core CPU, making it not that much slower than a quad core 13" Pro (-25%).
dogma1138 6y ago
There are HPC GPU centric servers with ARM CPUs and NVIDIA GPUs on the market https://www.hpcwire.com/off-the-wire/oss-introduces-solution-incorporating-nvidia-gpus-with-nvlink-and-marvell-thunderx2-arm-based-servers/ https://www.hpcwire.com/off-the-wire/oss-introduces-solution... It actually makes quite a bit of sense more so than buying Intel/AMD CPUs if you think about it. The CPU does little to nothing in those machines using a bespoke ARM design will allow you to have as many PCIe lanes as your heart desires and also optimize other things such as memory access to make those solutions even more optimal than using off the shelf general purpose server platforms.
drfuchs 6y ago
Apple could also just buy AMD for about a third of their cash in the bank.
dwaite 6y ago
> The only thing that's IMO in the air is what is going to be the discrete graphics story for macbook pro's and Mac pros. My opinion is that the A13 graphics performance is not significantly behind on discrete graphics cards today, and we know apple has historically been able to scale up over 50% as they adopted the integrated graphics capabilities from iPhone to iPad Pro. I wouldn't rule a competitive iGPU from Apple out once you are operating at a higher TDP.
sudosysgen 6y ago
AMD is already work on Radeon chips for Samsung. In any case, I don't think you can compare ARM to Intel right now, when talking about switching chips. The question is ARM vs AMD, and with the amazing AMD chips coming out I'm not convinced that the ARM chips can be much of an improvement.
michaelmrose 6y ago
I don't think the complexity of chip performance can be captured by assigning a trajectory number to wholly different product lines and assuming we can use the trajectory we hand waved into existence to predict future performance. This is especially true when our sole means of comparison is a singular synthetic benchmark we are assigning as the arbiter of truth because its hard to compare actual applications.

ianai 6y ago

It’s even got a keyboard stand case that almost turns it into a miniature iMac-like device.

oneplane 6y ago
Imagine a teeny tiny iMac, that'd be a fun look. I wonder if the size vs. needs has changed in general as the adoption of large screens seems to have slowed down a bit in my environment (probably because everyone that wanted one now has one).

ajross 6y ago

The A12 is clearly at parity now vs. Intel's existing mobile offerings, probably somewhat ahead given the long delays with 10nm. The question upthread is whether or not switching architectures makes financial sense, not whether it's a (mild) technical win. Switching to their own chips cuts Intel out of the loop, but as far as business risk that simply replaces one single source manufacturer with another (TSMC). It probably saves money per-part, which is good. But then Apple is still drowning in cash and immediate term savings really aren't much of a motivator. > By 2021 Intel can probably deliver a 1.1-1.2x speed up tops for Macbook airs, while Apple can probably deliver a 2x speed up for 2021 with the A15 and another one for 2023 with the A16. That's going to need some citation. Moore's law is ending for everyone, not just Intel. TSMC has pulled ahead of Intel (and Samsung has caught up) for sure, but progress is slowing. That kind of scaling just isn't going to happen for anyone.

fluffything 6y ago
> That's going to need some citation. So this is Geekbench 5 showing the difference between a 2012 macbook air i5 and a 2020 macbook air i5 - both base models and similar price: https://browser.geekbench.com/v5/cpu/compare/2613713?baseline=2613509 https://browser.geekbench.com/v5/cpu/compare/2613713?baselin... And this is a thread in the rust subreddit about compilation speed on macbooks where some users report the performance increase for different generations of macbook pros and macbook airs, if you want a more "realistic benchmark" to calibrate geekbench results: https://www.reddit.com/r/rust/comments/gypajc/macbook_pro_2020_compile_performance/ https://www.reddit.com/r/rust/comments/gypajc/macbook_pro_20... This is the ipad pro 2020 crushing the macbook air 2020: https://browser.geekbench.com/v5/cpu/compare/2612714?baseline=2613509 https://browser.geekbench.com/v5/cpu/compare/2612714?baselin... And this is the improvement from the previous generation ipad A10 to the ipad pro's A12Z - 2x speed up in single a generation: https://browser.geekbench.com/v5/cpu/compare/2613991?baseline=2613301 https://browser.geekbench.com/v5/cpu/compare/2613991?baselin... You are definetely right that Moore's law is hitting Intel hard. But AMD is still doing quite well, nvidia and "ati" are doing incredibly well, and Apple chips have been doing extremely well over the last couple generations. Maybe you are right, and Apple won't be able to deliver 2x speed ups in the next 2 generations. I'd expect that, just like for Intel, things won't abruptly change from one gen to another, but for this to happen over a longer period of time. Right now, only apple knows what perf their next 2 gens of chips are expected to deliver. The only thing we know is that Apple ARM chips are crushing their previous generation both for ipads and iphones year after year, and now they are betting on them for macbooks, and potentially mac pros, probably for at least the next 10-15 years.

zozbot234 6y ago

Discrete graphics are not dependent on the processor instruction set, they just need to interface with the PCI bus. All AMD has to do is deliver their cards, the drivers are FLOSS already.

raverbashing 6y ago
I assume the drivers might have a lot of specific x86 optimizations, so it's easy (but not too easy) I'm not sure how's the status of PCI-X on Arm machines at the moment as well. Do PCs still have the North and South bridge to interface to multiple devices?
- kcb 6y ago
  All CPUs are pretty much SOCs now. The chipsets are basically just fancy PCI-E switches with peripherals attached like USB and SATA controllers.

bitL 6y ago

> The performance of the Intel-based Macbook Air has increased by 1.8-1.9x from 2012 to 2020 That's not really correct, the latest Air can be configured with a quad core CPU, making it not that much slower than a quad core 13" Pro (-25%).

dogma1138 6y ago

There are HPC GPU centric servers with ARM CPUs and NVIDIA GPUs on the market https://www.hpcwire.com/off-the-wire/oss-introduces-solution-incorporating-nvidia-gpus-with-nvlink-and-marvell-thunderx2-arm-based-servers/ https://www.hpcwire.com/off-the-wire/oss-introduces-solution... It actually makes quite a bit of sense more so than buying Intel/AMD CPUs if you think about it. The CPU does little to nothing in those machines using a bespoke ARM design will allow you to have as many PCIe lanes as your heart desires and also optimize other things such as memory access to make those solutions even more optimal than using off the shelf general purpose server platforms.

drfuchs 6y ago

Apple could also just buy AMD for about a third of their cash in the bank.

dwaite 6y ago

> The only thing that's IMO in the air is what is going to be the discrete graphics story for macbook pro's and Mac pros. My opinion is that the A13 graphics performance is not significantly behind on discrete graphics cards today, and we know apple has historically been able to scale up over 50% as they adopted the integrated graphics capabilities from iPhone to iPad Pro. I wouldn't rule a competitive iGPU from Apple out once you are operating at a higher TDP.

sudosysgen 6y ago

AMD is already work on Radeon chips for Samsung. In any case, I don't think you can compare ARM to Intel right now, when talking about switching chips. The question is ARM vs AMD, and with the amazing AMD chips coming out I'm not convinced that the ARM chips can be much of an improvement.

michaelmrose 6y ago

I don't think the complexity of chip performance can be captured by assigning a trajectory number to wholly different product lines and assuming we can use the trajectory we hand waved into existence to predict future performance. This is especially true when our sole means of comparison is a singular synthetic benchmark we are assigning as the arbiter of truth because its hard to compare actual applications.