Nvidia is reportedly discontinuing its entry-level MX mobile graphics chip. While this is a rumor, it makes a lot of sense (open in new tab), as it predicts the death of Nvidia's weakest notebook chip.

The MX line was not only Nvidia's cheapest and lowest-performing series of notebook GPUs, it was also the lowest power consumption option. However, as both Intel and AMD have improved the performance of their integrated graphics with each generation, the MX has been pinched. Both from below and above.

Meanwhile, Nvidia's upcoming RTX 4050 mobile chip is expected to maintain the minimum 35W power consumption profile of the previous RTX 3050 generation. Considering both these APUs and the 35W RTX 4050, the window in which any Nvidia MX chip can operate is very narrow.

Currently, the best MX line is the MX570 (open in new tab). This chip is based on the GA107 GPU and has 2,048 CUDA cores, a boost clock of 1,155 MHz, and a 64-bit memory bus. The peak shader performance of 4.7 TFLOPS represents a considerable shortfall compared to the 8.9 TFLOPS of the fastest Radeon 780M integrated graphics in AMD's latest notebook APUs, the Ryzen 7040 series (opens in new tab).

At 25W, the MX570 is lower power than something like the RTX 4050, but also higher power than the 15W rating of integrated Radeon graphics. From there, the MX line drops off rapidly.

The MX570 is the only member of the MX family to feature Ampere generation technology. The runner-up, the MX550, falls to the Turing-based TU117 chip, with just 1,024 CUDA cores and 2.7 TFLOPS of raw shader performance.

Incidentally, the mobile RTX 4050 is expected to use the upcoming GPU AD107 with 2,560 CUDA cores, 1,755 MHz boost clock, and 96-bit memory bus for a total shader power of 9 TFLOPS. Therefore, even though it is expected to deliver better real-world gaming performance than AMD APUs, it will not improve raw shader power over AMD APUs.

In any case, given the cost and complexity of adding a discrete GPU to a laptop, the advantages, if any, of the MX line versus modern integrated graphics are sure to be marginal. space for Nvidia's smallest mobile GPU. There is no room left.