Discussion Apple Silicon SoC thread

[Eug]

M1
5 nm
Unified memory architecture - LP-DDR4
16 billion transistors

8-core CPU

4 high-performance cores
192 KB instruction cache
128 KB data cache
Shared 12 MB L2 cache

4 high-efficiency cores
128 KB instruction cache
64 KB data cache
Shared 4 MB L2 cache
(Apple claims the 4 high-effiency cores alone perform like a dual-core Intel MacBook Air)

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

16-core neural engine
Secure Enclave
USB 4

Products:
$999 ($899 edu) 13" MacBook Air (fanless) - 18 hour video playback battery life
$699 Mac mini (with fan)
$1299 ($1199 edu) 13" MacBook Pro (with fan) - 20 hour video playback battery life

Memory options 8 GB and 16 GB. No 32 GB option (unless you go Intel).

It should be noted that the M1 chip in these three Macs is the same (aside from GPU core number). Basically, Apple is taking the same approach which these chips as they do the iPhones and iPads. Just one SKU (excluding the X variants), which is the same across all iDevices (aside from maybe slight clock speed differences occasionally).

EDIT:



M1 Pro 8-core CPU (6+2), 14-core GPU
M1 Pro 10-core CPU (8+2), 14-core GPU
M1 Pro 10-core CPU (8+2), 16-core GPU
M1 Max 10-core CPU (8+2), 24-core GPU
M1 Max 10-core CPU (8+2), 32-core GPU

M1 Pro and M1 Max discussion here:

Page 78 - Discussion - Apple Silicon SoC thread

Page 78 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M1 Ultra discussion here:

Page 109 - Discussion - Apple Silicon SoC thread

Page 109 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M2 discussion here:

Page 127 - Discussion - Apple Silicon SoC thread

Page 127 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

Second Generation 5 nm
Unified memory architecture - LPDDR5, up to 24 GB and 100 GB/s
20 billion transistors

8-core CPU

4 high-performance cores
192 KB instruction cache
128 KB data cache
Shared 16 MB L2 cache

4 high-efficiency cores
128 KB instruction cache
64 KB data cache
Shared 4 MB L2 cache

10-core iGPU (but there is an 8-core variant)
3.6 Teraflops

16-core neural engine
Secure Enclave
USB 4

Hardware acceleration for 8K h.264, h.264, ProRes

M3 Family discussion here:

Page 215 - Discussion - Apple Silicon SoC thread

Page 215 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M4 Family discussion here:

Page 263 - Discussion - Apple Silicon SoC thread

Page 263 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

 

[Eug]

https://browser.geekbench.com/v6/cpu/7698811

4,04 Ghz

As expected, 8 GB RAM.

 

[FlameTail]

MacBook Pro Won't Switch to OLED Display Next Year, Two Analysts Say

While it has been rumored that the MacBook Pro will eventually get an OLED display, two analysts believe that change won't happen next year. ...

www.macrumors.com

Apple supply chain analyst Ming-Chi Kuo and display industry analyst Ross Young today both predicted that the 14-inch and 16-inch MacBook Pro models will continue to use mini-LED displays in 2025. This lines up with previous rumors that have claimed the MacBook Pro will switch to OLED display technology as early as 2026.

So I am guessing there will be no M5 Macbook Pros released next year then, and instead it will come in early 2026.

 

[ikjadoon]

Dan Nystedt

Apple makes up half of ARM's revenue??

Dan is even more specific: "The iPhone makes up close to half of Arm’s total revenue already." I find that impossible to reconcile with publicly-available data. Arm disclosed its major customer revenue shares during its IPO and its FY24 financial filings:

Arm Total Revenue Share from...	FY21	FY22	FY23	FY24
1st Largest Customer	20%	18%	24%	21%
2nd Largest Customer	12%	12%	11%	11%
3rd Largest Customer	10%	12%	9%	10%
4th Largest Customer	<10%	<10%	<10%	??

Of course, we don't know the names (though Arm admitted that Qualcomm is 10% of Arm's total revenue for FY24; Qualcomm's share was disclosed due to pending litigation).

Apple (which includes iPhones, Macs, iPads, wearables, etc.) as a whole cannot contribute more than 21%. There is no way the iPhone alone is close to 50% of Arm's total revenue. Dan is wildly off. Further, Apple uses the cheaper per-unit uArch licenses (predominantly; like most, Apple also licenses Arm microcontrollers, i.e., Cortex-M) and sells in lower volume than Android OEMs.

//

As expected, 8 GB RAM.

This score appears to show a pure clock bump, e.g., the numbers virtually align for 0.0% IPC improvement vs A17 Pro.

A18: 3114 / 4.04 GHz = ~771 Pts / GHz
A17 Pro: 2978 / 3.84 GHz = ~776 Pts / GHz

A coincidence or perhaps the "A18" is higher-clocked A17 Pro with one GPU core disabled? But GB scores & claimed peak frequency can be rather messy.

//

For easier searching on GB6, assuming the naming convention holds for the motherboards:

D47AP = A18 / iPhone 16
D48AP = A18 / iPhone 16 Plus
D93AP = A18 Pro / iPhone 16 Pro
D94AP = A18 Pro / iPhone 16 Pro Max

 

[FlameTail]

Arm Total Revenue Share from...	FY21	FY22	FY23	FY24
1st Largest Customer	20%	18%	24%	21%
2nd Largest Customer	12%	12%	11%	11%
3rd Largest Customer	10%	12%	9%	10%
4th Largest Customer	<10%	<10%	<10%	??

Any guesses as to who those other largest customers are?

 

[okoroezenwa]

MacBook Pro Won't Switch to OLED Display Next Year, Two Analysts Say

While it has been rumored that the MacBook Pro will eventually get an OLED display, two analysts believe that change won't happen next year. ...

www.macrumors.com

So I am guessing there will be no M5 Macbook Pros released next year then, and instead it will come in early 2026.

I’d imagine it’d just be that we’d get a 2025 M5 MBP and that’s the last of the miniLED displays. Or at least I hope that’s what they do. Then again the late 2019 - late 2021 MBP gap and the late 2020 to mid 2022 MBA gap happened so who knows.

 

[LightningZ71]

Any guesses as to who those other largest customers are?

My guesses, Samsung and Mediatek...

 

[igor_kavinski]

Early preview of Apple "Intelligence" shows it can make a lot of stuff up, still needs work

Apple is banking on Apple Intelligence to give its stagnant iPhone sales a boost. The technology received plenty of coverage during yesterday's Glowtime event, such as its...

www.techspot.com

As expected, nothing special. They are gonna be going on an AI experts hiring spree soon!

Good for OpenAI workers I suppose.

 

[jdubs03]

Dan is even more specific: "The iPhone makes up close to half of Arm’s total revenue already." I find that impossible to reconcile with publicly-available data. Arm disclosed its major customer revenue shares during its IPO and its FY24 financial filings:

Arm Total Revenue Share from...	FY21	FY22	FY23	FY24
1st Largest Customer	20%	18%	24%	21%
2nd Largest Customer	12%	12%	11%	11%
3rd Largest Customer	10%	12%	9%	10%
4th Largest Customer	<10%	<10%	<10%	??

Of course, we don't know the names (though Arm admitted that Qualcomm is 10% of Arm's total revenue for FY24; Qualcomm's share was disclosed due to pending litigation).

Apple (which includes iPhones, Macs, iPads, wearables, etc.) as a whole cannot contribute more than 21%. There is no way the iPhone alone is close to 50% of Arm's total revenue. Dan is wildly off. Further, Apple uses the cheaper per-unit uArch licenses (predominantly; like most, Apple also licenses Arm microcontrollers, i.e., Cortex-M) and sells in lower volume than Android OEMs.

//



This score appears to show a pure clock bump, e.g., the numbers virtually align for 0.0% IPC improvement vs A17 Pro.

A18: 3114 / 4.04 GHz = ~771 Pts / GHz
A17 Pro: 2978 / 3.84 GHz = ~776 Pts / GHz

A coincidence or perhaps the "A18" is higher-clocked A17 Pro with one GPU core disabled? But GB scores & claimed peak frequency can be rather messy.

//

For easier searching on GB6, assuming the naming convention holds for the motherboards:

D47AP = A18 / iPhone 16
D48AP = A18 / iPhone 16 Plus
D93AP = A18 Pro / iPhone 16 Pro
D94AP = A18 Pro / iPhone 16 Pro Max

Like you said, I’ll add: no IPC improvement only adds up if the performance core is the same in the A18 Pro as the A17 Pro and is clocked higher.

And that would be a shame because the M4 has 8.42% higher IPC than the M3 (spec2017 int/fp avg) at 4GHz.

And in particular because the A17 Pro (756.34 Pts/Ghz) actually performed better in Pts/GHz than the M3: 752.46 at 4.06GHz.
Similar story with the A16 Bionic and the M2.

To me the history suggests the score would include a clock speed increase and IPC increase. If so that would be max 15.87% more performance. That would lead to a score of ~3352, or 829.71 Pts/GHz.

For reference:
M4 3P6E 4.4GHz - 3662; 832.27 Pts/GHz
M4 4P6E 4.4GHz - 3715; 844.32 Pts/GHz

I don’t think that benchmark run is representative of the A18/Pros true performance. If it is then they significantly broke with their history.

 

[igor_kavinski]

Apple Intelligence is going to be in beta for the time being. Funny how some users here claimed that Apple has been doing AI for years

 

[poke01]

Like you said, I’ll add: no IPC improvement only adds up if the performance core is the same in the A18 Pro as the A17 Pro and is clocked higher.

And that would be a shame because the M4 has 8.42% higher IPC than the M3 (spec2017 int/fp avg) at 4GHz.

And in particular because the A17 Pro (776 Pts/Ghz) actually performed better in Pts/GHz than the M3: 752.46 at 4.06GHz.
Similar story with the A16 Bionic and the M2.

To me the history suggests the score would include a clock speed increase and IPC increase. If so that would be max 15.87% more performance. That would lead to a score of ~3352, or 829.71 Pts/GHz.

For reference:
M4 3P6E 4.4GHz - 3662; 832.27 Pts/GHz
M4 4P6E 4.4GHz - 3715; 844.32 Pts/GHz

I don’t think that benchmark run is representative of the A18/Pros true performance. If it is then they significantly broke with their history.

That leaked bench doesn’t even make sense. The score is also being boosted by SME and clock speed and yet it’s so low.

An A17 Pro clocked at 3.78GHz scores 2900 and an A18 scores 3111 at 4.04GHz but with SME.

 

[ikjadoon]

This is a little tangential for the CPU sub-forum, but some brief speculation on the 16s:

1. What was the conclusion on the "8GB" M4 iPad Airs, e.g., does Apple Intelligence eat into the 8GB? I assume a off quirk with the lower-end M4 iPad Air. Perhaps different AI models on M4 devices vs A18 (Pro) devices.
2. That camera button, being a capacitive sensor: touch ID returning in a few years? Curious about cases, though: virtually all are making cutouts, but the UX seems worse unless the cutout is big enough.
3. Are the faster MagSafe chargers also Qi2.1 compatible?
   1. "In an interview with Android Authority, the WPC’s Marketing Director said that a Qi2.1 revision will bring faster wireless charging sometime in mid-2024." Source
4. Unknown if Thread 1.4? Thread 1.4 launched last week with key changes & Apple is a Thread board member.
   1. "The fixes coming with 1.4 are based on learnings from this larger deployment. 'Thread 1.4 represents our first strategic response to real-world use cases,' said Vividh Siddha, president of Thread Group and director of software engineering at Apple." Source
5. No BT 5.4, though the improvements seemed minor vs BT 5.3.
6. Wi-Fi 7 is a more important upgrade vs BT & Thread. But, like @Eug, we're also on Wi-Fi 5 and it works well even as most clients are Wi-Fi 6 capable.
7. Sustained SoC perf will be interesting, as I recall iPhones have some sub-optimal cooling setups from Andrei's reviews.
8. No tandem OLED: I'd be more interested in efficiency at lower brightnesses than its peak, but not this year.
9. Phones are a little thicker this year; hopeful the battery life improvements actually bear out.
10. Still waiting to see if Apple's camera (or lens coating) can fix the blue orbs during night photos & videos. Could Apple Intelligence fix that first? 😆

 

[jdubs03]

Whew that’s quite a list. I can tackle 2 right now. I highly doubt touch ID is coming back to the iPhone. Face ID works extremely well. My assumption is for the iPhone 17 Pros the cutout will be either the same or a little smaller. There is word that the 17 is finally when the Face ID module becomes transparent under the screen; so it would just be that front facing camera cut out at that point. I presume they would still keep the island fly out though. Hopefully they do.

 

[name99]

I'm a little worried about Apple modem rumoured for next year's iPhones though. Will it be able to keep up with Qualcomm's offering?


Heheh. My home WiFi is still 802.11ac. I have 6 Apple AirPort Extremes around my large oddly-shaped house, so I'm not inclined to upgrade the entire WiFi network as it would not provide much meaningful functional improvement in my house and would cost a lot to upgrade.

Read those two paragraphs side by side...

The Apple modem will be "comparable" to the QC modem. It will probably be superior along the dimensions Apple cares about (most importantly energy) and possibly slightly inferior in terms of peak bps.

Either way IT WILL NOT MATTER for anyone except the dick-measuring brigade, for precisely the reasons encapsulated in Eug's post. People are happy to insist that some network metric is the most important thing in the universe -- but revealed preference (when it's your own money you have to spend) shows that most people are perfectly happy with what they have right now...

 

[name99]

https://browser.geekbench.com/v6/cpu/7698811

4,04 Ghz

what is 17,3? A18 or A18 Pro?

If we do the comparison vs A17 Pro

iPhone17,3 vs iPhone 15 Pro Max - Geekbench

the main thing we see is the usual extreme noisiness (enough so that I wouldn't care too much about any detail) EXCEPT that we obviously have SME present (no surprise).

The more interesting question is whether there was enough time to implement the next steps (SVE present and/or SSVE that's performant enough to be useful)...

 

[name99]

Like you said, I’ll add: no IPC improvement only adds up if the performance core is the same in the A18 Pro as the A17 Pro and is clocked higher.

And that would be a shame because the M4 has 8.42% higher IPC than the M3 (spec2017 int/fp avg) at 4GHz.

And in particular because the A17 Pro (776 Pts/Ghz) actually performed better in Pts/GHz than the M3: 752.46 at 4.06GHz.
Similar story with the A16 Bionic and the M2.

To me the history suggests the score would include a clock speed increase and IPC increase. If so that would be max 15.87% more performance. That would lead to a score of ~3352, or 829.71 Pts/GHz.

For reference:
M4 3P6E 4.4GHz - 3662; 832.27 Pts/GHz
M4 4P6E 4.4GHz - 3715; 844.32 Pts/GHz

I don’t think that benchmark run is representative of the A18/Pros true performance. If it is then they significantly broke with their history.

Guys, we go through this EVERY FSCKING CYCLE. A few GB6 releases that have terrible numbers because the idiots running them are desperate to be FRIST!!! and run them even as the phone is downloading updates, recategorizing photos, etc.

We ALWAYS get these "new cores are disappointing" stories, then a week later it's clear that, no, the core is actually about 15% faster, just like Apple claimed.

Don't you ever learn?

 

[Doug S]

That leaked bench doesn’t even make sense. The score is also being boosted by SME and clock speed and yet it’s so low.

An A17 Pro clocked at 3.78GHz scores 2900 and an A18 scores 3111 at 4.04GHz but with SME.

Compare that result to iPhone 15 Pro Max. There are degradations in a bunch of benchmarks, both single and multi core. Clearly the device test was not performing optimally for whatever reason.

I saw decreases in compression, PDF, horizon, remover, and structure in single core. In multi it lost ground in everything except HTML5, photo, text, and object!

 

[jdubs03]

Guys, we go through this EVERY FSCKING CYCLE. A few GB6 releases that have terrible numbers because the idiots running them are desperate to be FRIST!!! and run them even as the phone is downloading updates, recategorizing photos, etc.

We ALWAYS get these "new cores are disappointing" stories, then a week later it's clear that, no, the core is actually about 15% faster, just like Apple claimed.

Don't you ever learn?

Fair fair. At least the history between the past few major releases were compared 🙃
Sets reasonable expectations.

 

[poke01]

iPhone17,2 - Geekbench

Benchmark results for an iPhone17,2 with an ARM processor.

browser.geekbench.com

This is A18 Pro

 

[xiewe3wq]

iPhone17,2 - Geekbench

Benchmark results for an iPhone17,2 with an ARM processor.

browser.geekbench.com

This is A18 Pro

Over 20% regression in file compression in single-core but no such regression in multi-core file compression? These results seem off.

 

[jdubs03]

Over 20% regression in file compression in single-core but no such regression in multi-core file compression? These results seem off.

Far off in a different solar system. Also i’ll be surprised that the A18 Pro does worse on single core than the A18; not like I trust either of those results.

 

[poke01]

Hey thats more like it

https://twitter.com/x/status/1833773661465018779

 

[jdubs03]

Hey thats more like it

https://twitter.com/x/status/1833773661465018779

Yea now that’s bang on.
I’d be curious of the MT score if they had 6 e-cores. Seems like it’d be pretty solid.

 

[Eug]

Hey thats more like it

https://twitter.com/x/status/1833773661465018779

This is M1 territory for multi-core. Yes there is that hardware acceleration boost that A18 Pro gets, but it’s still impressive.

iPhone17,2 vs iPad Air (5th generation) - Geekbench

 

[FlameTail]

This is M1 territory for multi-core.

The competition is already in M2 territory!

 

[mvprod123]

The competition is already in M2 territory!

And power consumption as well lol