A first benchmark of the upcoming Apple M1X processor has been leaked on the web, but the data should be taken with great caution.
Already a few weeks after the official launch of the new Apple M1 processors, which took place on November 10, 2020, on the Web have begun to turn the first rumors about the second model of the new family of SoC, which should equip the next 14 and 16-inch MacBook Pro and the new 27-inch iMac.
Now pops up a first benchmark of a fictitious Apple M1X, allegedly the new member of the M1 family, which looks very similar to the rumors so far circulated. This is a processor very similar to the current Apple M1, compared to which has differences only "quantitative" (more cores for the CPU and GPU) and not also "qualitative" (the architecture remains identical). The benchmark, which appeared on the CPU Monkey platform, should be taken with extreme caution since Apple has not yet made official the new processor and the new models of iMac and MacBook Pro. However, it is interesting to read the data of this test and, above all, the technical characteristics of the next Apple M1X (which is not at all sure that it will be called so).
Apple M1X: the technical characteristics
In comparison to Apple M1 the M1X appeared on CPU Monkey maintains the 5 nm production process, the architecture with unified memory and also the clock speed, equal to 3.2 GHz. It goes up to 35 Watts, however, the maximum heat production of this new SoC (the previous one, in models with active fan cooling, stops at 20 Watts).
The CPU goes up from 8 cores to 12 while the GPU goes up from 8 cores to 16. This information is consistent with what the market expects: the Apple M1 has proven to have a very good CPU, fast and efficient, and a good GPU. But the GPU is still far from the specific solutions of Nvidia and AMD and, on Mac Mini, MacBook Air and MacBook Pro 13 inches (i.e. the three current models with Apple M1) it is unfortunately not possible to add an external video card.
Slightly boosting the CPU and doubling the GPU capabilities, along with doubling the maximum amount of RAM (which on M1X would rise to 32 GB), are therefore exactly what users would like. Too good to be true? Perhaps, which is why the information published by CPU Monkey should be weighed carefully.
How fast is the Apple MX1
The fact that the benchmark appeared on CPU Monkey is to be taken with a grain of salt is confirmed by the same test results: despite the 4 extra CPUs, in fact, the results are identical to those of the Apple M1.
The GPU tests, however, show a performance gain far too high: the 32-bit iGPU test, for example, goes from 2,600 points to 5,200 points. A roundabout doubling, which would reflect the doubling of GPU processing units already described. But in real-world use, it's not that simple: a theoretical doubling is never matched by a doubling in practice, to the single point.