The M6 Pro’s early Geekbench 7 scores look impressive, with nearly 5,000 points on the single-thread test and strong multi-thread scores. But it’s too early to get excited about the new Apple chips – it looks like what we’re seeing is a fake.
Geekbench creator John Poole told MacRumors that he found internal inconsistencies in the published results and concluded that the M6 Pro test may not be genuine. At the same time, it is not easy to see obvious problems with the claimed features or indicators at a glance.
A large number of processor cores does not in itself arouse suspicion. Thanks to the Fusion architecture Apple uses in the M5 Pro and M5 Max, the company has been able to equip both chips with the same number of CPU cores. The purported frequency of the tested M6 Pro also looks reasonable: it’s only slightly higher than the M5 Max’s performance core.
There’s another reason not to jump to conclusions. Mark Gurman previously reported that Apple may skip the M6 Pro and M6 Max and go directly to the M7 Pro and M7 Max. There is no official confirmation of this yet, but combined with John Poole’s doubts, the latest Geekbench results look even less convincing. Therefore, the M6 Pro tests found so far are best regarded as unconfirmed information.
Other publications on this topic
Liquid nitrogen would normally make smartphone cooling a farce, but Apple’s A20 Pro is suddenly almost ready to handle such extreme situations. In Geekbench 6, the increase is very small, which seems particularly unusual compared to competitors.
Today 14:09
0
TSMC will not stop at 1.4nm. The company has received approval to once again expand the Hsinchu Science and Technology Park, where it will produce more advanced chips in the future, according to Taiwanese media reports.
Today 14:03
0
Asus is preparing a new BIOS version for motherboards using AM5 sockets, which will improve the performance of DDR5 memory on Chinese manufacturer Changxin memory chips. The first beta version has appeared on many models on the X870, B850, X670 and B650 chipsets.
Today 14:02
0
