Enlarge /. Developers don't waste time getting their hands dirty with the new A12Z ARM Developer Transition Kits.

As reported by MacRumors, eager Apple developers are already publishing benchmarks for developer transition kits for Macs with Apple silicon. These kits are based on the Mac Mini chassis, but contain ARM-derived Apple silicon instead of Intel CPUs.

Before we delve into it, it is important to note some limitations. First, the CPU included in these developer kits may or may not reflect the CPUs included in future Apple Macs. These are not consumer goods; They are developer tools. Secondly, the benchmarks were performed with Rosetta, which is likely to have many changes and optimizations. And third, the developers who released this information violate Apple's nondisclosure agreements.

Developers wishing to access the kit must pay an access fee of $ 500, agree to return the kit after one year, and not publicly review, review, approve, or view the device without Apple's prior written permission . At least eight developers don't seem to have read the fine print yet, as measured by uploads to Geekbench's online leaderboard.

There is currently no way of knowing whether these leaks are intentional or accidental. At Ars, we often get benchmarking hardware that is not available to the general public and whose details are also embargoed – and we can confirm that you need to be careful about what you do. Most modern benchmark utilities have an online leaderboard with a "Upload Results" button built right into the utility. In some cases, they are even uploaded by default, unless you specifically force it to.

Accidentally or not, the leaks give us some additional information about the potential performance of the new Macs with Apple silicon, although nothing conclusive. The developer transition kits are equipped with a variant of the A12Z SoC, which can be found in the latest iPad Pro models. In these Geekbench database entries, the virtual CPU is also reported as a four-core CPU and not as an eight-core CPU – although the A12Z as we know it in the iPad Pro is an octa-core CPU.

Four of the A12Z cores are fast, high-performance cores, and the other four are slow, low-performance cores that are used to increase battery efficiency in background tasks. This configuration is common in the ARM world, but almost unknown in x86. It is therefore not surprising that an x86 emulation ignores the large / small configuration and reports as a simpler four-core setup regardless of the underlying reality.

These tests appear to have been run in Geekbench 5.2.0 for macOS x86 (64-bit). This means that they were run in Rosetta, Apple's tool for emulating x86 Macs on ARM-based Apple silicon.

The developer kits equipped with Apple silicon contain an average of 811 for single-threaded geekbench and 2781 for multi-threaded. That's about 20 percent slower than the single-core results of the Macbook Air with i3-1000ng4 and 38 percent faster than the multithreaded results. However, high-end Macs produce much higher numbers.

The impressive thing about these leaked numbers is that they are not suitable for Geekbench that run natively in ARM mode. These tell us what legacy app emulation on Apple Silicon Macs could look like – and it's likely that early users of Apple's new ARM-based Macs will use Rosetta to run at least some apps. So this is a potentially useful finding.

The performance features of Rosetta 2 are not yet well known enough to sensibly extrapolate the A12Z's native performance. However, if the leaked numbers are correct, we can assume that they are quite good.

Nevertheless, there is still no indication that the A12Z is ​​actually delivered on consumer Macs. Apple may have plans to introduce a completely different chip when it comes to actually launching the new Macs. Although these benchmarks are a fascinating oddity, they are not surely representative of what we will see when the actual deal arrives later this year. These kits are not designed to reflect the final hardware of Apple Silicon Macs.

LEAVE A REPLY

Please enter your comment!
Please enter your name here