Since Apple announced the A12Z and its move away from x86, there have been questions about exactly how these ARM chips will work and what we can expect from them. The first benchmark results are starting to appear in Apple’s development kits, and as long as you take them with a mountain of salt, they’re pretty cool.
What we have to work with here is Geekbench. Geekbench tends to be a very solid test for Apple CPUs, but in this case, we are talking about Apple CPUs running the x86 version through emulation. Even if Geekbench favors Apple’s CPUs more than x86, running the app through an emulator will affect performance.
Also, note that the app only reports four cores. The A12Z is nominally an eight-core chip, with four large, four small. It is unclear if these development systems only use the “big” kernels, or if the app just doesn’t detect them correctly, or if this is a limitation of the emulator. Anyway, it is very early and these are early results.
Here is the data as it got to Geekbench 5.
We see single-threaded scores of 844 and a multi-threaded score of 2958, producing a scale factor of 3.5x. On the x86 side of the equation is the 13-inch MacBook Pro, with scores of 1218 and 4233. This also results in a scale factor of about 3.5x. Similarly, the 13-inch Macbook Pro is approximately 1.44 times faster than the A12Z in single-threaded and multi-threaded mode.
One thing to note is that emulation performance can vary drastically Depending on the application. Some programs can run with relatively small penalties, while other craters and die. Rosetta 2 is specifically designed to avoid those results, but historically, there’s a nasty corner case or two lurking in some emulator. Some applications are more difficult to emulate than others. But the result of this effect is that we don’t really know if that 1.44x advantage that the 13-inch MacBook has is the product of the emulator’s disability or if it’s a good look at the CPU performance. IPad Pro data suggests it might be the first.
If we assume that the A12X on the iPad Pro is a pretty good substitute for the A12Z, we can check the performance of ARM’s native Geekbench, albeit on iOS, not macOS. Here, we’re looking at 1120 single-core, 4650 multi-core, with a 4.16x scale factor. The 13-inch MacBook Pro is only 8 percent faster than the iPad Pro in a single thread, and 10 percent slower in multiple threads.
Frankly, that should send a frisson scary through Intel and AMD. The implication of these results is that the gap between the 13-inch Mac and the A12Z is largely the result of emulation. That’s not a guarantee, because operating system differences are important in situations like this, but it certainly seems like most of the penalty that comes with the A12Z is related to emulating the x86 code.
That fact should send a frisson of fear for AMD and Intel’s collective thorns. Apple’s annual record of offering new performance improvements is considerably better than Intel’s at the moment. AMD can make a much stronger argument for its own recent upgrade, thanks to Ryzen, but Excavator’s huge 1.52x IPC upgrade to Ryzen tilts the comparison a bit. To put it bluntly, AMD’s improvements over the past three years would be a little less impressive if Bulldozer hadn’t been such a horrible chip to start with.
We are in a strange situation right now. Intel has always been Apple’s top provider, but AMD is selling more efficient mobile CPUs today, making them the most obvious point of comparison. The 4900HS seems to get a single-core score of 1116 and a multi-threaded score of 7013. x86 MT is not, at least, in immediate danger, in absolute terms. Keep in mind that the 4900HS also consumes much more power than either Intel or Apple chips.
What we see here is not proof that Apple will release a MacBook ARM chip that rivals the best that Intel and AMD can offer, but it certainly puts a floor below expected performance, barring the unusual quirks of the emulator that Apple will pass the next few months crushing. X86 companies can ask their mobile CPU designers to grab an extra cup of coffee.
Final Note: These kits are not the CPUs that Apple will ship to customers and do not represent final performance.
Featured image from Apple.
Now read:
