Discussion Apple Silicon SoC thread

Page 328 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Eug

Lifer
Mar 11, 2000
23,807
1,385
126
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:

Screen-Shot-2021-10-18-at-1.20.47-PM.jpg

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:


M1 Ultra discussion here:


M2 discussion here:


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:


M4 Family discussion here:

 
Last edited:

name99

Senior member
Sep 11, 2010
496
382
136
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

Senior member
Sep 11, 2010
496
382
136
what is 17,3? A18 or A18 Pro?

If we do the comparison vs A17 Pro
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)...
 
Last edited:

name99

Senior member
Sep 11, 2010
496
382
136
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

Platinum Member
Feb 8, 2020
2,742
4,664
136
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

Senior member
Oct 1, 2013
803
408
136
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.
 

jdubs03

Senior member
Oct 1, 2013
803
408
136
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.
 

ikjadoon

Senior member
Sep 4, 2006
218
470
146

To clarify, that is an A18 Pro score: D94AP → iPhone 16 Pro Max.

Dx7AP = Base
Dx8AP = Plus
Dx3AP = Pro
Dx4AP = Pro Max

D64AP - iPhone 13 Pro Max
D74AP - iPhone 14 Pro Max
D84AP - iPhone 15 Pro Max
D94AP - iPhone 16 Pro Max

For A18 scores, the motherboard should be D48AP (iPhone 16 Plus) or D47AP (iPhone 16).

Highest A18, 3420 / 8370: https://browser.geekbench.com/v6/cpu/7714502
Highest A18 Pro, 3409 / 8492: https://browser.geekbench.com/v6/cpu/7714134
 

mvprod123

Member
Jun 22, 2024
52
60
51
So we have proof that A18 has a new GPU uarch. The regular A18 with 5-core GPU matches the raw performance of the A17 Pro with 6-core GPU. A18 Pro 6-core gpu is faster than M1 8-core GPU. Both use LPDDR5X-7467, which is confirmed by Apple's claim of a 17% improvement in memory bandwidth. I'm curious about what changes Apple has made and whether FP32 has improved performance.
 

FlameTail

Diamond Member
Dec 15, 2021
3,855
2,297
106
So we have proof that A18 has a new GPU uarch. The regular A18 with 5-core GPU matches the raw performance of the A17 Pro with 6-core GPU. A18 Pro 6-core gpu is faster than M1 8-core GPU. Both use LPDDR5X-7467, which is confirmed by Apple's claim of a 17% improvement in memory bandwidth. I'm curious about what changes Apple has made and whether FP32 has improved performance.
So this means M4 Pro and M4 Max will also have this new GPU uarch.