Discussion Apple Silicon SoC thread

Page 321 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Eug

Lifer
Mar 11, 2000
23,871
1,438
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:



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:

FlameTail

Diamond Member
Dec 15, 2021
4,095
2,465
106
So we know Apple has been increasing clock speeds by substantial amounts in their recent CPU core generations.

Is this achieved by increasing the number of pipeline stages?
 

FlameTail

Diamond Member
Dec 15, 2021
4,095
2,465
106
Redesigned Mac Mini Incoming!
The new ‌Mac mini‌ will be the first major design change to the machine since 2010
The M4 generation of Apple silicon will purportedly be the first time that Apple has used the same chip generation across all of its Macs, with refreshes to add the M4 chip to the MacBook Air, MacBook Pro, iMac, Mac Studio, and Mac Pro planned to take place over the next year.
 
Reactions: Ghostsonplanets

poke01

Platinum Member
Mar 8, 2022
2,350
3,069
106

johnsonwax

Member
Jun 27, 2024
83
155
66
Redesigned Mac Mini Incoming!

Sounds like a MBP with screen, keyboard, battery removed. When you remove all the unnecessary bits, squeezing that logic board down to an AppleTV form factor doesn't seem unreasonable. Will need a fan.
 
Reactions: Ghostsonplanets

Doug S

Platinum Member
Feb 8, 2020
2,836
4,820
136
So similar in size to Apple TV but a bit taller and runs macOS. This could be a nice bring with me overseas desktop.

I had kind of assumed that Apple would redesign the Mini to ATV size for the M1, I was surprised to see it using the same form factor and 65W power supply as the x86 versions that preceded it. But I guess they have only so much bandwidth for redesigning Macs and clearly the Mini was well down in the pecking order.

But at the time I didn't know they were going to put a Pro in it, and that combined with the need for the USB-C ports to output power require a beefier power supply than what the base M1 was going to need.

I wonder if they will move to a power brick and power it via Magsafe/USB-C or keep the power supply internal? The advantage of an external power supply would be if you want to connect higher power USB-C/TB devices to it you could use a brick with more wattage. Also helps with cooling not having the power supply inside the case, and obviously makes it easier to make the thing smaller.
 
Reactions: poke01

Doug S

Platinum Member
Feb 8, 2020
2,836
4,820
136
Since there are rumours of M5 being server focused, what are the chances of Apple including SVE2?

I haven't seen such rumors, and it makes no sense given that Apple doesn't sell servers. They aren't going to design their consumer product around internal needs, that's crazy talk.

Anyway Apple doesn't need SVE2. SME requires SSVE, and while the performance of SSVE is underwhelming in M4 there's no reason it has to be. If they want what SVE2 provides they'll address SSVE's deficiencies.

Best of all since the SME/AMX unit is separate from the cores, if they wanted more "oomph" in either SME or SSVE in higher end models like the Max they could give it a bigger unit (which would come in handy for their internal server needs as well) That way they don't have to pay the cost for all that in all their cores like they would if they introduced some 512 bit SVE2 like some people think they should - including iPhones since they use the same P cores across all products!
 

Eug

Lifer
Mar 11, 2000
23,871
1,438
126
Since there are rumours of M5 being server focused
What rumours are those?
Indeed. All I've read is there may be some sort of server chip being developed. "M5 is server focused" is a rather big extrapolation.

Anyhow, the macOS Sequoia 15.1 beta apparently already has references to twelve M4 series Mac models, and (likely) two M5 series Mac models, judging by the identifiers.

That doesn't indicate when they will be released or what they are exactly, but in this context a fall 2024 release of M4 series Macs seems feasible. I had previously guessed early 2025, but some of the pundits are predicting earlier, and having my M4 iPad Pro already in hand (months ago in fact) supports that prediction.

As for M5 Macs, those macOS 15.1 references bode well for a 2025 release. Again, which models, I dunno.

BTW, the other prediction now is that the Mac mini will be redesigned, including with the the additional USB-C ports we had been predicting given additional Thunderbolt controllers which appear to be in M4. They're talking about a design that is much smaller and reminiscent of Apple TV, with at least 3 USB-C ports. That makes sense considering the current Mac mini is largely empty space inside, but I hope they don't compromise the thermals too much. I suppose the possibility of a fanless device isn't completely out of the question, although I still think both the M4 model and the M4 Pro model will have fans.
 
Last edited:

FlameTail

Diamond Member
Dec 15, 2021
4,095
2,465
106
So we know Apple has been increasing clock speeds by substantial amounts in their recent CPU core generations.

Is this achieved by increasing the number of pipeline stages?
That would allow for higher clocks, but at some other costs, like more branch mispredict penalties.


Apple paid a heavy price in terms of power for the large performance increase brought by M4.

+51%/+68% power increase.
 
Reactions: igor_kavinski

johnsonwax

Member
Jun 27, 2024
83
155
66
That would allow for higher clocks, but at some other costs, like more branch mispredict penalties.

View attachment 104965View attachment 104966
Apple paid a heavy price in terms of power for the large performance increase brought by M4.

+51%/+68% power increase.
I'm going to go out on a limb and say the 2023 iMac wasn't sitting in liquid nitrogen when the test was done, unlike I suspect the iPad Pro was. We don't really know how each of these is going to throttle. I'm going to go out on a limb and say the M4 iMac won't show either those performance numbers or those power numbers. And the relationship between the two is not linear.
 
Reactions: Nothingness

poke01

Platinum Member
Mar 8, 2022
2,350
3,069
106
View attachment 104965View attachment 104966
Apple paid a heavy price in terms of power for the large performance increase brought by M4.

+51%/+68% power increase.
Yep but only for ST. For MT ( when 3 or more cores are active) which runs at ~3.95GHz on the M4 remains at ~5 watts with 4% increase in INT. This is even more important as it will enable Apple to cram more cores into the M4 Max and enable higher core counts for their desktops.

With N3E having higher yields they could go this route and probably is why the M4 Ultra is planned this generation.
 

jdubs03

Senior member
Oct 1, 2013
951
594
136
I'm going to go out on a limb and say the 2023 iMac wasn't sitting in liquid nitrogen when the test was done, unlike I suspect the iPad Pro was. We don't really know how each of these is going to throttle. I'm going to go out on a limb and say the M4 iMac won't show either those performance numbers or those power numbers. And the relationship between the two is not linear.
You’d be correct. The M4 iPad Pro was cooled by liquid nitrogen in the Geekerwan test.

But I think in the below image it wasn’t:
 
Reactions: poke01

FlameTail

Diamond Member
Dec 15, 2021
4,095
2,465
106
This is even more important as it will enable Apple to cram more cores into the M4 Max and enable higher core counts for their desktops.

With N3E having higher yields they could go this route and probably is why the M4 Ultra is planned this generation
M4 Max = 16P + 4E?

Then M4 Ultra = 32P + 8E

That would be a huge uplift from the M2 Ultra (16P + 8E).Of course, there isn't an M3 Ultra (and will never be).
 

okoroezenwa

Member
Dec 22, 2020
105
112
86
M4 Max = 16P + 4E?

Then M4 Ultra = 32P + 8E

That would be a huge uplift from the M2 Ultra (16P + 8E).Of course, there isn't an M3 Ultra (and will never be).
I'd be surprised if they move from 12P for the Max, especially with how well it worked for the M3. I have wondered if in a world where the M4 Pro went back to 8P they still use 2 clusters of 6P or move to 2 clusters of 8P for the Max.
 

name99

Senior member
Sep 11, 2010
511
395
136
That would allow for higher clocks, but at some other costs, like more branch mispredict penalties.

View attachment 104965View attachment 104966
Apple paid a heavy price in terms of power for the large performance increase brought by M4.

+51%/+68% power increase.
This is your annual reminder that ENERGY and POWER are not the same thing. If you can't intelligently discuss whether the ENERGY required for a task went up, or how the energy-delay product changed then you have nothing useful to say. Period.


We went through this same BS with the M1 to M2 transition, and it seems that the main culprits at that time, the ones with hysterical click-bait headlines, have learned absolutely nothing since then...
 
Reactions: smalM

mikegg

Golden Member
Jan 30, 2010
1,847
471
136
M4 Max = 16P + 4E?

Then M4 Ultra = 32P + 8E

That would be a huge uplift from the M2 Ultra (16P + 8E).Of course, there isn't an M3 Ultra (and will never be).
Highly doubt they can cool 16P cores at the level Apple likes.

They'll likely add more E cores instead.

I think 12P + 6E. Total 18 cores.
 

FlameTail

Diamond Member
Dec 15, 2021
4,095
2,465
106
Highly doubt they can cool 16P cores at the level Apple likes.

They'll likely add more E cores instead.

I think 12P + 6E. Total 18 cores.
Poke said this:
Yep but only for ST. For MT ( when 3 or more cores are active) which runs at ~3.95GHz on the M4 remains at ~5 watts with 4% increase in INT. This is even more important as it will enable Apple to cram more cores into the M4 Max and enable higher core counts for their desktops.

With N3E having higher yields they could go this route and probably is why the M4 Ultra is planned this generation.
 

FlameTail

Diamond Member
Dec 15, 2021
4,095
2,465
106
Even if the 12P+4E core count is unchanged in M4 Max, I wonder if the amount of L2 cache pet cluster will be increased.

Although M3 Pro/Max increase the number of P-cores in a cluster to 6, the L2 cache size is unchanged!
Number of P cores in a clusterQuantity of L2 cache in a clusterL2 cache per core
M1 series412 MB3 MB
M2 series416 MB4 MB
M3 Pro/Max616 MB2.66 MB

This means the L2 cache per core in M3 Pro/Max is less than that of even the M1!
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |