Apple's M1 Max GPU is as powerful as an Nvidia RTX 2080 desktop GPU and the Sony PS5 gaming console
Apple has launched its long-awaited and all-new 14-inch and 16-inch models. Unlike in the past, and thanks to the thermal efficiency of Apple’s new M1 Pro and M1 Max SoCs, both models can be fitted with the fastest silicon that Apple offers. In the past, a chip like the M1 Max with 10-core CPU and up to 32-core GPU might only have made it into the larger device.
We already have an insight into the single-core and multi-core performance of the new chips thanks to leaked GeekBench 5 scores. But what of the graphics performance? Now that we know the exact configurations of these chips, we can start to deduce the graphics performance of each chipset based on the performance of the GPU in the original M1 chip. As Apple highlighted in its keynote, its new chips keep the M1 designations because they are scaled up versions of the M1 architecture.
We’ve put together this short estimated GPU performance list based on the assumption that Apple has not clocked the new GPU cores higher in the M1 Pro and M1 Max chips or made any other changes. Based on this, we have come up with the following teraflops (TF) performance estimates:
M1 8-core = 2.6 TF
M1 Pro 14-core = 4.5 TF
M1 Pro 16-core = 5.2 TF
M1 Max 24-core = 7.8 TF
M1 Max 32-core = 10.4 TF
The performance of all variants is exceptional, particularly given how much less power they draw than competing chipsets from Intel, Nvidia and AMD. As Apple pointed out during its presentation, even at peak performance, the M1 Max is drawing up to 100 W less power than a high-end PC laptop with discrete GPU. Yet, the 32-core GPU in the M1 Max produces identical performance to a desktop class Nvidia RTX 2080 GPU (10.07 TF) and Sony’s latest PS5 gaming console (10.28 TF). It’s jaw dropping performance in notebooks that deliver up to a claimed 21 hours of battery life as well.