A few days ago, Apple presented its own M1 SoC, intended for its new MacBook Air and Pro that have 8 cores and are manufactured with an advanced 5nm node. Apple had noted that this was the fastest low-power CPU in the world, although we didn’t have much detail on its graphics performance.
Apple’s M1 chip has superior graphics performance than the GTX 1050 Ti
Some recently discovered GFXBench 5.0 results were shared by an anonymous user, which makes it somewhat clearer how the GPU of Apple’s M1 SoC is shared.
The Apple M1 SoC was tested under the Metal API , which perhaps makes it difficult to find direct comparisons with other graphics cards on the market. One of the graphics cards that was registered using Metal is the GTX 1050 Ti, so this is good for a comparison.
Apple’s chip easily outperforms the GTX 1050 Ti in the Metal test. This shows the good work of Apple in the graphics of this SoC, beating a graphics card that has a TDP of 75W. It would have been interesting to be able to compare it with a more current GPU, such as the GTX 1650, which has a computing power of 2.9 teraflops, close to the 2.6 teraflops of the SoC M1 chip.
If we average the performance between all the benchmarks, the GPU of the Apple M1 gives an average of 275.07 FPS ; the Nvidia GeForce RTX 1050 Ti averages 199.83 FPS ; and the AMD Radeon RX 560 averages 177.34 FPS .
With this data in hand, we can say that the GPU of the Apple M1 is 37.6% faster than the Nvidia GeForce GTX 1050 Ti in this benchmark, and 55.1% faster than the AMD Radeon RX 560. The interesting thing will be to know here also energy consumption, which will be infinitely lower in the Apple GPU.
Even so, GFXBench 5.0 is not the best tool to make a comparison of this type, since it is designed more for smartphones than to test PC graphics cards, so these results must be taken with some caution.