As reported by MacRumors, eager Apple developers are already posting benchmarks in Apple’s Developer Silicon Mac Developer Kits. These kits are based on the Mac mini chassis but include ARM-derived Apple silicon instead of Intel CPU.
Before going deeper, it is important to note some caveats. First, the CPU included in these developer kits may or may not reflect the CPUs included in future Apple Macs. These are not consumer products; They are development tools. Second, the benchmarks were made with Rosetta, which probably still has a lot of changes and optimizations to come. And third, the developers who leaked this information are violating Apple’s confidentiality agreements.
Developers who wanted to access the kit were required to pay an access fee of $ 500, agree to return the kit after one year, and agree not to write, review, share, or publicly display the unit without Apple’s prior written approval. So far, at least eight developers appear not to have read the fine print, judging by the loads on the Geekbench online leaderboard.
There is currently no way of knowing if these leaks are deliberate or accidental. At Ars, we often compare hardware that is not available to the general public and whose details are also under embargo, and we can confirm that you should pay close attention to what you are doing. Most modern benchmark utilities have an online leaderboard, with a “upload results” button directly in the utility, and in some cases, even upload default unless you specifically compel them not to.
Accidentally or not, the leaks give us additional insight into the potential performance of Apple’s new silicon Macs, though nothing conclusive. The developer transition kits are outfitted with what appears to be a variant of the A12Z SoC found in the latest iPad Pro models. These Geekbench database entries also report that the virtual CPU has four cores, rather than Eight, even though the A12Z, as we know it on the iPad Pro, is an eight-core CPU.
Four of the A12Z’s cores are fast, high-power cores, and the other four are slow, low-power cores used to increase battery efficiency when running tasks in the background. This setting is common in the ARM world but almost unknown on x86. So it’s not too surprising that an x86 emulation ignores the big / small configuration and reports itself as a simpler quad core configuration regardless of the underlying reality.
These tests appear to have run on Geekbench 5.2.0 for macOS x86 (64-bit), which means they ran on Rosetta, Apple’s tool for emulating ARM-based Apple silicon x86 Macs.
In terms of results, Apple’s silicon-equipped developer kits average 811 for single-threaded Geekbench and 2,781 for multi-threaded. That’s about 20 percent slower than the entry-level i3-1000ng4 with Macbook Air single-core results, and 38 percent faster than its multi-threaded results. However, high-end Macs produce much higher numbers.
The awesome thing about these leaked numbers is that they are not for Geekbench that runs natively in ARM mode. These tell us what legacy app emulation would look like on Apple’s silicon Macs, and early adopters of Apple’s new ARM-based Macs are likely to use Rosetta to run at least some apps, so it’s an idea potentially useful.
The performance characteristics of Rosetta 2 are not yet well known enough to significantly extrapolate the native performance of the A12Z, but if the leaked numbers are correct we can assume that it will be quite good.
All that said, there is still no indication that the A12Z is actually shipping on consumer Mac. Apple may have plans to introduce a very different chip when it hits the market with the new Macs, so while these benchmarks are an intriguing curiosity, it’s not certain that they’re representative of what we’ll see when the real deal hits later. . year. These kits were not designed to reflect the final hardware of Apple’s Silicon Macs.
