Apple M1 Max Duo and Quadro, the engines of the new Mac Pro?

Apple SoCs M1 Max Duo and Max Quadro They have appeared on the horizon to motorize the most powerful machines of the Cupertino firm, the Mac Pro workstations.

Apple continues to make adequate progress in the transition from Intel’s x86 to ARM designs, with its own chips that have demonstrated their capacity, favored by a closed ecosystem that allow full integration with the operating system and, it is fair to recognize it, for the good work from a company with extensive experience in RISC architectures.

Those of Cupertino announced that would complete the transition in three years, renewing its entire catalog with ARM. The MacBook Air, 13-inch MacBook Pro, Mac mini, and 24-inch iMac already use M1 silicon, and Apple recently announced new improved SoCs for the 14-inch and 16-inch MacBook Pros. Only the larger iMac AIOs and the Mac Pro still using Intel would be left out. This will surely change next year.

Apple M1 Max Duo and Quadro

The Mac Pro is Apple’s workstation and as such, there is expectation to know how Apple can replace Intel’s Xeon processors and especially NVIDIA’s dedicated professional graphics, maintaining its high performance. It is clear that it is much easier to make a hardware change in an ultraportable like the Air than a work machine as powerful as this one.

The well-known Apple leaker, Mark Gruman, says Apple plans to power the Mac Pro using multi-die designs from its latest developments. Instead of going back to the drawing board and designing chips with higher specs, Why not just join two or four of the existing SoCs together?

That seems to be the idea and as their names indicate, the Apple M1 Max would have two SoCs and would be used in the basic configurations of the new Mac Pro, while the “Quadro” would use four Max for the high-end Mac Pro, mounting 40 CPU cores and 128 graphics cores.

Hector Martin, who is in the process of migrating Linux to the M1 Pro / Max, indirectly supports Gruman’s claims by the evidence that discovered in the macOS code. «The macOS drivers have a lot of references to the multiple matrices, and the IRQ driver on the M1 Pro / Max is clearly designed with a second half (currently unused) for a second matrix“Martin tweeted. Technically it is a second set of config / mask / software-gen / hw-state registers with entries currently inactive, but which can be enabled.

If the CPU capacity seems sufficient to maintain the performance of the Intel Xeon (ARM is used in servers and supercomputers such as the Japanese Fugaku, the most powerful on the planet), more doubts arise from the capacity to cover the graphical section and the certifications of professional CAD / CAM / architecture / engineering applications… that require this type of machine. We will see.

The new 27-inch iMac is also expected to use these types of multi-array designs, as employing a simple SoC would degrade its performance versus the current model with Intel hardware.

And there is always the possibility of new chips. It is speculated that the SoC M2 with 40 cores is in development for arrival in 2023, although it may exceed the time horizon set to complete the transition and hence the rumors point to bundled chips and not to completely new dies.

Apple is expected to unveil the new ‘all-in-one’ iMacs AIOs in March 2022. The release schedule for the Mac Pro is unknown and will be the ultimate gauge to verify the success of the transition to ARM and collateral damage for Intel (and in general for the x86 architecture) you will have this program if the rest of the manufacturers follow suit. Of course, at the moment only Apple and with a super controlled platform, seems capable of unleashing the full potential of ARM on PCs.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *