Apple Silicon roadmap leak conflicts with previous Mac Pro rumors
- According to a new report, Apple's silicon plans for the Mac Pro are in conflict with previous reports.
- It does, however, provide some rumored new details about Apple's future silicon roadmap, which includes the next MacBook Air.
Apple's new MacBook Pro lineup was undoubtedly overshadowed by the Apple silicon that powers the devices. Since Apple chip guru Johny Srouji detailed the new M1 Pro and M1 Max chips' capabilities, they've been the talk of the tech blogosphere. While Intel's new 12th-generation Alder Lake laptop chips appear to be competitive in terms of overall performance, their performance-per-watt appears to be around 50% lower than Apple’s.
Of course, this has sparked a lot of speculation about what Apple may have planned for the launch of the Apple Silicon-based Mac Pro. The Mac Pro will reportedly use the first-generation M1-based Mac chips, code-named Jade, fabricated on a 5nm node, according to a new report from The Information, as cited by Ars Technica. However, unlike Bloomberg reporter Mark Gurman's previous report, which predicted a four-die M1 configuration, this would only be a two-die M1 Max configuration. Hector Martin, a well-known developer who is porting Linux to the Mac, tweeted that the macOS driver codebase contains numerous references to multi-die configurations. He also said that the IRQ (interrupt request) controller “is very clearly engineered with a (currently unused) second half for a second die.” This isn't to say that Apple couldn't connect the dies with fabric interconnect outside the dies for a total of 40 CPU cores and 128 GPU cores.
When it comes to the performance of a Mac Pro with a two-die M1 Max configuration, there simply isn't enough GPU horsepower (CPU side is not a concern). A Radeon Pro W6800 Duo MPX Module, which uses two linked GPUs over AMD's Infinity Fabric interconnect to produce up to 30.2 TF of single-precision compute, is the best performing GPU in the current Mac Pro. Two M1 Max dies would have a total of 64 cores for a single-precision compute capacity of 20.8 TF. This is clearly insufficient, pointing to Gurman's four M1 Max die configuration, which would yield 41.6 TF of computing power. The current Mac Pro, however, can support up to two Radeon Pro W8600 Duo MPX Modules, each of which can produce up to 60.4 TF when used in tandem. While Apple may adopt a four M1 Max configuration (if not a two M1 Max configuration for an entry-level Mac Pro), one can speculate that it may also have to consider a discrete GPU option and/or something similar to the Afterburner accelerator it currently offers as an option. Apple has already proven that its Mac silicon can exceed people's expectations, so anything could be possible by the time the next Mac Pro arrives in 2022.
Apple's second-generation Mac chip will be a single die chip, according to The Information, and will be used in the rumored redesigned MacBook Air. This would most likely be the rumored M2 chip, which would be based on the A15 architecture found in the iPhone 13 series, as opposed to the A14 architecture found in the iPhone 12 series. According to reports, the entire second generation of M2-based chips has been completed.
Work on the third generation of M-series chips, which are said to be fabricated on a 3nm process, is also said to have begun. Our guess is that these will be Apple's first chips built on the new Armv9 architecture, based on Apple's customized implementation of it on the upcoming A16 SoC. Over the next two generations of SoCs, the new Armv9 instruction set will bring CPU increases of over 30%, making it the first new Arm architecture in a decade. You can bet Apple will take full advantage of this in a way that no one else has been able to do.
Instant Stories by Apple Scoop