Digital Trends can earn a commission when you shop through links on our website.
Apple’s big claims about the performance of the M1 Pro and M1 Max seem legitimate as more and more benchmarks keep popping up. This time a Redditor found GFXBench scores for the M1 Max that put it around 25% faster than the mobile Nvidia RTX 3070. This is in line with Apple’s charts from its presentation.
Reddit users sent to school sent the first M1 Max GFXBench scores in the r / hardware subreddit. Obviously, it would be wise to be skeptical of these results for several reasons. Initially, the submission to GFXBench was made using Apple’s Metal API, compared to results from Nvidia and AMD using OpenGL on Windows.
That is, compare M1 Max Results for the RTX 3080 Mobile and Radeon RX 6800M (the same GPUs Apple used in its presentation) sees the Apple Silicon either beat the other two GPUs or come within striking distance of the other two GPUs when looking at the offscreen results.
Comparing the Aztec Ruins Normal Tier’s offscreen results to other GPUs is pretty overwhelming. The M1 Max appears to be more powerful than several desktop graphics cards, including the RTX 2080 Super, Radeon RX 6800 XT, and even the Nvidia Titan V. The M1 Max doesn’t win in every benchmark like the Car Chase and Aztec Ruins High Tier, however the fact that it can even compete with desktop components is a game changer.
These aren’t the only benchmarks that prove the M1 Max’s performance. We reported about it a few days ago early Geekbench results for the M1 Max. In this benchmark, the M1 Max achieved 1,749 in the single-core test and 11,542 in the multi-core test. Compared to Apple’s own M1 silicon in the 13-inch MacBook Pro, the M1 Max is almost 60% faster in multi-core performance.
It must be said that brute force alone is not the only accomplishment here. These increases in performance are possible despite the 70% lower power consumption of the M1 Pro and Max compared to Intel’s offerings. Remember that the maximum wattage available to charge the new 16-inch MacBook Pro is 140 watts. The desktop RTX 2080 Super alone has a 250W TGP.
As mentioned earlier, it is important to view these numbers with a healthy amount of skepticism. There is an advantage to using Apple’s Metal API as it is tailored specifically to Apple hardware, while OpenGL is more generalized for hardware compatibility reasons.
However, this also means that apps specifically designed for Apple Silicon could achieve much greater performance advantages. Apple has drawn attention not only to Intel, but also to AMD and Nvidia, and it will be interesting to see how the companies battle Apple’s massive vertical integration advantage.