AMD Radeon RX 7900M vs Apple M2 Pro 16-Core GPU vs Apple M2 10-Core GPU

AMD Radeon RX 7900M

The AMD Radeon RX 7900M is a mobile upper high-end graphics card based on the Navi 31 chip (RDNA 3 architecture) manufactured in 5nm. It features 4,608 cores (72 CUs - compute units) and 16 GB of fast GDDR6 graphics memory connected with a 256 Bit memory bus. The chip also includes 64 MB Infinity Cache. The clock speed of the cores will depend on the laptop according to AMD. The specified game clock speed is 1825 MHz (most likely at 180 W TGP). 

AMD compares the performance to the mobile RTX 4080 and shows a number of games where the 7900M is on average 7% faster. Raytracing is supported in hardware, but usually slower than the Nvidia counterpart. Nvidia also has the advantage of DLSS support in selected games.

The Navi 31 chip is manufactured at TSMC in the current 5nm process. The power consumption (TGP) is specified at 160 to 180 Watt (200 Watt including SmartShift from the CPU).

Apple M2 Pro 16-Core GPU

The Apple M2 Pro 16-Core-GPU is an integrated graphics card by Apple offering all 16 of the 19 cores in the M2 Pro 10-Core Chip.

The graphics card has no dedicated graphics memory but can use the fast LPDDR5-6400 unified memory with a 256 bit bus (up to 200 GBit/s).

According to Apple, the performance of the 19-core-GPU should be 30% higher than the old 16-core-GPU in the M1 Pro. Therefore, the cut down 16-core version should still beat the old full-featured 16-core M1 Pro slightly and the old 14-core more clearly.

The GPU is intended to use Apple Metal 2 API and could still be based on the older PowerVR architectures (last used in the Apple A10). A new feature in the MacBook Pro 14 of 2023 is the support for HDMI 2.1 and 8k output.

The Apple M2 Pro is manufactured in the second generation 5nm process at TSMC.

Apple M2 10-Core GPU

The Apple M2 GPU is an integrated graphics card offering 10 cores designed by Apple and integrated in the Apple M2 SoC. It uses the unified memory architecture of the M2 SoC (up to 24 GB LPDDR5-6400 with 100 GB/s bandwidth) and should offer 160 execution units.

According to Apple it offers a 25% higher performance at slightly higher power consumption. The theoretical performance is rated at 3.6 Teraflops and therefore 1 TFLOP higher than the M1 8-core GPU. In our benchmarks, the GPU was able to top the M1 with up to 40%, but still stays behind the 14-core GPU in the M1 Pro. Compared to other iGPUs from AMD and Intel, the M2 benefits greatly from the unified memory architecture and the high bandwidth.

The M2 chip is manufactured in the second generation 5nm process at TSMC (most likely N5P). The power consumption is around 13.5 Watt (in our tests under load of the MBP13) and up to 15W according to Apple.