Apple announced this week that its new M2 chip will "enable console-quality gaming (opens in new tab)." So Apple claims, but is it true?

Naturally, from this Apple we are talking about, no further details have been provided. Apple has not said anything about which game console they have in mind, nor about other factors such as the game in question, resolution, refresh rate, etc.

But we'll take Apple's word for it and compare the new chip to the latest gaming consoles. In terms of pure graphics rendering, the current pinnacle of consoles is Microsoft's Xbox Series X (opens in new tab). Its AMD RDNA 2-derived GPU delivers 12.1 single precision FP32 TFLOPS.

Now, TFLOPS is only one measure of the performance of the pixel processing shader core. There are many other important metrics that contribute to the important end result, the frame rate, such as texture processing and memory bandwidth.

Similarly, comparing GPUs across software platforms and APIs is tricky. 10TFLOPs on a console is not exactly the same as 10TFLOPs running on DirectX on Windows or Metal on MacOS.

However, as a rough indication of the gaming graphics performance of a given GPU, this is a reasonable guideline. With all this housekeeping out of the way, how do these new M2 chips stack up?

At the top of the M2 chip stack is the M2 Max, which has 38 GPU cores, up from 32 GPU cores in the old M1 Max. In all, it is rated at 13.6 TFLOPS. Additionally, M2 Max has adequate bandwidth in and out of the GPUs. The shared memory architecture allows the GPU and CPU to reach a combined bandwidth of 400 GB/s.

This is roughly the same level as the Xbox Series X, which peaks at 560 GB/sec for 10 GB of its 16 GB memory allocation and 336 GB/sec for the remaining 6 GB.

Of course, the fastest PC GPUs blow all this out of the water: the Nvidia RTX 4090 (open in new tab) has just under 83 TFLOPS and just over 1TB/sec of memory bandwidth; the RTX 4080 (open in new tab)? 49 TFLOPS and 717 GB/sec.

Perhaps the PC GPU with the closest performance in terms of raw TFLOPS is AMD's Radeon RX 6700 XT (open in new tab), with 13.2 TFLOPS supported by 384 GB/sec bandwidth.

On the other hand, these are discrete GPUs and are definitely not a fair comparison to the M2 family, which is a full SoC with the GPU integrated with the CPU and other functions. Perhaps a fairer comparison would be AMD's latest laptop APU, codenamed "Phoenix," announced at CES (open in new tab). It features an integrated RDNA 3-based GPU that achieves 4.3 TFLOPS.

In short, these new M2 chips have some pretty impressive graphics hardware, especially for an integrated GPU, and are broadly comparable to the latest consoles and even some decent desktop GPU hardware.

Apple's chip also seems to be quite special with respect to efficiency. Apple claims up to 22 hours of battery life for the latest 16-inch MacBook Pro with M2 Max. This may be somewhat optimistic, but for a laptop with the performance of a desktop 6700 XT, that's spectacular battery life, even cut in half; PC gaming laptops can't even come close to that.

Discussion of hardware performance is almost pointless, as there remains a rather pathetic selection of games made to run on MacOS and Apple silicon. Buying an M2 Mac for gaming will result in almost complete disappointment. However, one must grudgingly admit that the new M2 chip does have great graphics performance.