• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion Apple Silicon SoC thread

Page 474 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Eug

Lifer
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:


M5 Family discussion here:

 
Last edited:
ahh thats total system power draw, not just CPU power. The Max/Pro have a lot more I/O, bigger dies and higher busses. You can also see the single core power consumption being higher on the M4 Pro/Max vs base M4 in the chart too
Fair fair. Though I will note the relative difference is much more.
wait what?

View attachment 139635

Macworld got 9426???

View attachment 139636

iPhonedo got 9524

Yeah this one:
He gets 771 ST in CB 2026, but 8903 in MT.
Only 55 in Speedometer 3.1? Should be 60ish.

The numbers are all over the place.
I mean 771 vs Macworld getting 722? That’s almost a 7% difference.
 
Pardon me if this was already posted:

Mobile RTX 5080 on power vs M5 Max on battery running the CP2077 benchmark

Settings: RT maxed, no path tracing nor upscaling.


The MacBook resolution was 2336×1460 while PC was 2560x1440 (so macbook rendered 7% less pixel while being overall 15,5% slower (22.91 FPS vs 26.96 FPS), but this is an insanely good result for high level raytracing on battery.

One thing is benchmark graphs, the other thing is to listen to these side-by-side (one barely audible).
 
That stutter on the PC laptop... there's more to it than meets the eye. Great result by the Mac though.
Those pauses!

BTW, depending upon the settings, sometimes it's actually advantageous to run a benchmark with the machine unplugged, since battery charging can generate heat. OTOH, if set another way, sometimes an unplugged machine will run slower. So, It's strange that he ran one machine plugged in and the other not.
 
Curious to see dieshots.

I thought the new P-cores might improve nT power efficiency significantly, but that doesn't seem to be case.
I wouldn’t think there would be more efficiency to be gained fully loaded, both Apple’s previous cores have had efficiency no one could imagine up in laptop and desktop until November 2020. The e core efficiency is insane!

I think the new cores are for day to day efficiency gains by taking on more work that required the previous P cores to be running.

I think they will be like what Intel is trying to do with their e cores. Intel’s e cores had about double the SPEC scores of Apple’s, but using about 5 W versus Apple’s sub .6 watts. If that Chinese post is accurate Apple’s new cores will have about the same performance as a Mont, but at what wattage.

The other question I have is Apple has been using per cluster frequencies and power, if the new cores are really in one 12-core cluster, it would seem that would require more complexity in cluster powering.

Alex Ziskind’s video he indicated that not only could six cores stay off, but that they stayed off even during an extended compute demand.


Alex Ziskind
 
Here we go again… Back in 2020 when the M1 came out, we got a wave of videos just like this one, telling us how great 8 GB M1 Macs were and how you could edit 4K video and compose on pro music apps on them. And it was true…

…but about 6 months later there was another wave, a smaller wave but a wave nonetheless, of regret videos saying that 8 GB works but causes problems. People would report huge slowdowns on bigger projects and random crashes that simply didn’t occur on 16 GB machines. IOW, going from 8 GB to 16 GB on otherwise identically spec’d machines meant the difference between crashing and not crashing. And yes, this was way back in 2021 when this second wave of videos came out. We’re in 2026 now.

Obviously the 8 GB Neo isn’t targeted at pros and it can indeed be used for 4K video editing and stuff like that, but for someone who actually wants to do this stuff more than once in a blue moon, they’d be better off just getting at least a MacBook Air even if they’re just in high school. The MacBook Air also has several other advantages such as Display P3 wide colour support and better battery life.

Meanwhile, I’m trying to convince the wife that my kid really doesn’t need a new laptop this year. If we can delay the purchase until next year, I might just do that and hope that Apple actually does release that 12 GB A19 Pro MacBook Neo in spring 2027. However, it would also depend upon what Apple does with the Back-To-School promotion this summer.
 
Back
Top