Apple M3 Pro Chip to Pack Entry-Level Configuration of 12 CPU Cores and 18 GPU Cores on TSMC 3 nm Technology

[AleksandarK]

Thanks to the latest release of the Power On newsletter from Mark Gurman, we have additional information about Apple's upcoming M3 Pro chip. Currently in testing and reported on by an App Store developer, Apple is looking to upgrade the microarchitecture of the forthcoming chip and add additional cores to the system for more performance. As the report notes, the entry-level M3 Pro chip currently in testing will have 12 CPU cores, six for efficiency and six for performance tasks, with 18 graphics cores, all manufactured on TSMC's 3 nm node. The current baseline for M2 Pro is 10 CPU cores, where four are dedicated to efficiency, and six are dedicated to performance. The current generation entry-level M2 Pro also features a 16-core GPU, which is two cores fewer than the upcoming model.

Generally, the M3 Pro chip will boost integrated memory across the board, as the sample spotted in testing shows 36 GB of memory. The M2 Pro offered 32 GB in that memory tier, so a four GB increase is inbound there. Presumably, the 16 GB version (if it exists) and 64 GB version will also get memory bumps by going the M3 Pro route. Of course, we have to wait for more information as these chips become more widely available to developers.



View at TechPowerUp Main Site | Source

 

[Fouquin]

Gurman was wrong about M2 before release and he's wrong here too. Just saying.

 

[Nephilim666]

Will probably match zen 4/5 mobile chips at similar tdp. Apple are going to need coprocessors as complex instruction sets become the norm and machine learning gets rolled into consumer hardware.

 

[Minus Infinity]

Gurman was wrong about M2 before release and he's wrong here too. Just saying.

Please elaborate.

 

[Fouquin]

Please elaborate.

No. Unlike Gurman I don't leak specs.

 

[david salsero]

Probablemente coincidirá con los chips móviles zen 4/5 a un tdp similar. Apple necesitará coprocesadores a medida que los conjuntos de instrucciones complejos se conviertan en la norma y el aprendizaje automático se incorpore al hardware de consumo.

Mientras tanto, los OEM no están usando los mejores procesadores del mercado, calidad/precio/consumo, el AMD Zen 4 Phoenix increíble mientras miles de consolas portátiles ya están en el mercado con Zen 4, ¿qué pasa con AMD, TSMC y OEM que sí? No están poniendo en el mercado los famosos ultrabooks de 2,2 libras que prometieron con RDNA 3 + inteligencia artificial.
Ahora mismo una compra segura es: AMD ZEN 4 Phoenix = DDR5 + RDNA 3 + USB 4.0 + HDMI 2.1 + AI con arquitectura XDNA desarrollada por Xilinx y todo a 4nm, el único procesador que supera al M2 hasta que llega el M3 a 3nm pero a que precio........

 

[forman313]

No. Unlike Gurman I don't leak specs.

He didn't ask you to.

When pointing out errors in others, you should be specific and give details. A quote (with source ofc) is a bare minimum if your intention is to help and inform.
If this wasn't your intention, I'm more than happy to give you a couple of tips on how to do that better as well. All in the spirit of helping the fellow man.

You should only type in Caps, and add 4-7 exclamation marks at the end.

 

[Fouquin]

He didn't ask you to.

Yes, they did. Elaborating on how Gurman's report is wrong on M3 would require divulging information about M3 as a counter. That info is not public and I'm not about to make it so just to prove a point. Just know Gurman has seeded sources that don't have the full picture, and he's wrong on many counts.

 

[forman313]

Yes, they did. Elaborating on how Gurman's report is wrong on M3 would require divulging information about M3 as a counter. That info is not public and I'm not about to make it so just to prove a point. Just know Gurman has seeded sources that don't have the full picture, and he's wrong on many counts.

Im guessing they wanted you to elaborate on what he got wrong in the M2 report/predictions.

BTW: Are you allowed to deny claims/rumors if you are under NDA?

That being said, 36GB sounds a little strange. The memory configuration would probably require a major redesign of the whole package.
Unless it's just 4GB hooked up to function as a cache or dedicated VRAM. 4GB app-L4 iFinity Cache ofc.. It's obvious