M3 roadmap speculation hints at next Apple Silicon generation chips
The M3 line of Apple Silicon chips will have a fairly familiar roadmap, a report points out in a breakout of the processor family's expected configurations.
As part of the fall product release schedule, Apple is anticipated to introduce the M3 generation for Apple Silicon. With speculation of the new chip lineup in full flow, fueled by developer logs, a roadmap for M3 has been created.
Writing in his "Power On" newsletter for Bloomberg, Mark Gurman offers caution that the roadmap can still be different from what Apple actually launches, since what Apple tests internally could vary wildly compared to what the public gets to purchase.
For the M3, Gurman believes it will have eight CPU cores, split evenly between performance and efficiency cores, as well as 10 GPU cores.
The list of Macs set to use M3 will apparently include the base MacBook Pro, 13-inch MacBook Air, 15-inch MacBook Air, Mac mini, and iMac. The iPad Pro is also believed to get M3 as well.
The M3 Pro's base configuration is anticipated to have 12 CPU cores, again split evenly between performance and efficiency cores, and an 18-core GPU. The top configuration will use add two more performance cores, bringing the total to 14, as well as a 20-core GPU.
Gurman says the M3 Pro will be available in the 14-inch MacBook Pro, 16-inch MacBook Pro, and Mac mini.
The M3 Max will start with a base configuration of 16 CPU cores, using 12 performance and four efficiency cores, and a 32-core GPU. On the high end, the M3 Max will have the same 16-core CPU but a 40-core GPU.
The 14-inch and 16-inch MacBook Pro and the Mac Studio are the strongest candidates for the chips.
Doubling what the M3 Max offers, the M3 Ultra will use a 32-core CPU with 24 performance and eight efficiency cores, and either a 64-core GPU or an 80-core GPU. The Mac Studio and possibly an upgraded Mac Pro could use the chip.
Along with the core changes, Gurman writes that memory options could vary for the releases. Models of MacBook Pro with 36GB and 48GB of memory have apparently undergone testing.
Read on AppleInsider
Comments
The GPU core counts look like a small increase so I'd say they will increase transistor count per core:
https://wccftech.com/apple-a10-fusion-cores-bigger-than-competition/
"One reason why Apple is adamant is designing larger cores is because having more transistors per core helps when performance and efficiency per-watt metric is calculated. While this might not be a good approach when conserving space, clock efficiency greatly increases thanks to these decisions."
They might also have a strategy similar to Intel's tick-tock. 2nm won't be ready until late 2025/2026 so they have to make 3nm last for 2023/2024/2025. I doubt they would throw everything in with the first 3nm revision then have a small refresh in 2025. It's best to split it so that each refresh has a worthwhile improvement (~50% increase each time) so that M4 (2025) is 2x M2 performance.
But TSMC won't start using Nanosheet (GAAFET) transistors until N2 and N2P.
4+4+10
6+6+18
8+6+20
12+4+32
12+4+40
Why would the Pro have 6 eff cores while the Max has 4 eff cores? They are not following the existing plan of designing 1 chip, the Max, and "chopping" off part of it for the Pro, and using two for the Ultra? So, if the Pro has 6 eff cores, the Max would have 6 eff cores as well. It really should look like:
4+4+10
6+6+16 (M3 Pro binned)
8+6+20 (M3 Pro)
10+6+32 (M3 Max binned)
10+6+40 (M3 Max)
Or, something like this is more sensible, which assumed each CPU perf complex is 4 cores, each CPU eff complex is 4 cores. So they would be using the same CPU perf and eff complexes from M3 all the way to M3 Ultra. It just varies by number of complexes and binning.
4+4+10
6+4+16
8+4+20
10+4+32
12+4+40
The big hole in the current Pro vs Max SoC lineup is that the Max doesn't offer more CPU perf cores than the Pro. Adding 2 or 4 more perf cores for the Max would provide a better upsell. So, good to hear the rumors are intimating this. The 18 to 20 GPU cores in the Pro though? Doesn't make sense. 10% isn't much of difference. It needs to be about 15% difference to be able to upsell on order $200. The only way they could make +10% work is if they combine it with a CPU core increase.
For the GPU core count, I'm guessing they must be significantly redesigned cores with more transistors each. In the mobile chips, A12 was 6-core:
https://en.wikipedia.org/wiki/Apple_A12
2p/4e CPU, 4 GPU cores, total 7b transistors
A16 is also 6-core CPU (2p/4e), 5-core GPU but 16b transistors:
https://en.wikipedia.org/wiki/Apple_A16
They must have roughly doubled the transistor count per core.
Since this is a new node, they will have redesigned the cores and in later revisions, they can increase the number of cores. 3nm allows 70% density increase so each core can increase that much, it may end up a little less. Lets say each GPU core has 40% more transistors and 10% more cores, then it's over 50% increase. Plus they are supposed to be faster at the same power on top of this.
I'm expecting 50% gain for CPU/GPU from M2->M3 at a lower power. Then with M4 they can add more cores at a higher power draw. 20-30TFLOPs GPU in a thin and light laptop where the fan barely comes on will be really nice.
So, 6 efficiency cores are possible imo. It all depends on the prioritization of performance or runtime in the MBP models, or whatever metric they are using. The quibble I had was that the Max chips should have the same or more efficiency cores than the Pro chips. That part of the rumor doesn't make sense if Apple is doing the same thing they did with prior Pro and Max chips. The specific number of eff cores could mean the Max only has 2 more perf cores as the total number cores could be gleaned from GB testing. 10 versus 12 performance cores will impact upsell.
You probably only should expect 20% level performance gains per CPU core at the same power consumption, with a good portion of it from higher clocks. The days of 50% gain is over. For GPU cores, I can see 100% for some benchmarks if hardware raytracing is there, but aggregate benchmarks are going to be 30% maybe.
M2 Max top-end is 12-core CPU (8p/4e). M3 Max top-end is being reported as 16-core (12p/4e). 12p/8p = 1.5x.
They will still want to push ahead of Intel, the i9-13900HK is just ahead of M2 Max, albeit with significantly more power usage:
https://www.notebookcheck.net/M2-Max-vs-i9-13900HK_14975_14715.247596.0.html
A 50% jump would make M3 Max comparable to the higher-end mobile i9 and Ryzen 9 and put it in the top 5 mobile CPUs. Ryzen 9 is on 5nm too and the following is over 2x M2 Max performance:
https://www.amd.com/en/products/apu/amd-ryzen-9-7945hx
Notebookcheck mobile CPUs
I think both CPU and GPU will gain at least 50% with M3. It's unusual in the industry to gain that much but Apple Silicon isn't following the industry trend of slowing down hardware releases.
I don't know if they are wanting to match x86 multi-core CPU performance, and obviously, they don't want to match Nvidia/AMD GPU performance as they aren't going to ship a 200 to 400 W chip. There is some sort of wholistic, integrated performance they are targeting with the Max chip. Something like it has to fit inside a 15mm thick laptop, the CPU can use no more than 30 Watts, and the GPU can't use no more than 60 W. If an ASIC can drive good performance at low power for Apple's Pro customers, they'd add it. Fit as much as performance inside those requirements. So, given this, TSMC 3nm will only buy them about 20% per core CPU and 30% per core GPU performance.
If they made no changes to the core designs from the M2, a 12+6+28 type design could be done for a Max chip, but I presume most of the additional transistors is going to increase per core performance, increasing execution pipeline stages for more Hz, etc. Hence only modest increases in core counts. Cache sizes are supposed going to have smallest increase as SRAM cache isn't scaling that well from 5nm to 3nm.
x86 systems are obviously free to hit 100+ W for the CPU, 200+ W for the GPU and have >25mm thick laptop form factors to play with. >25mm thick laptops with 5 hr runtimes.