Connect with us

EXTRAS

goodbye to low-end GPUs

According to the latest rumors coming from ASIA, NVIDIA would have decided ditch its GeForce MX line of GPUs. In case that doesn’t sound like much to you, we are talking about the line of low-end dedicated graphics cards for laptops.

Advertisement

The reason for this movement makes more than sense, and it is that they are graphics designed to consume very little energy. But of course, when it comes to talking about costs and power, it is clear that there are already much better options on the market. like integrated GPUs on AMD or Intel’s own CPUs. These GPUs offer everything necessary, and more, for any type of user. We talk from encoding or decoding the latest video formats, to running games (in the case of AMD).

NVIDIA GeForce MX could be dead by 2022

NVIDIA GeForce MX570

Specifically, the last GPU of this type that we saw was the NVIDIA GeForce MX570. This graph is essentially a GeForce RTX 2050 for laptops very limited. Curious, when this GPU was already a limited and stripped version from the GeForce RTX 3050.

Advertisement
Read  Radeon RX 6000 and below miss Hogwarts Legacy debut

The GeForce MX570 makes use of the graphics chip NVIDIA GA107 based on the Ampere architecture. This silicon has 2,048 CUDA Cores, 64 Tensor Core and 16 RT Core. Essentially the same specs as the RTX 3050 and RTX 2050 for laptops. The biggest difference, their frequency of 1.155MHz vs. 1,477 MHz for the RTX 2050 and 1,740 MHz for the RTX 3050.

Advertisement

This GPU stands out for being very limited with 2 GB memory at 12 Gbps attached to a memory interface 64 bit. The RTX 2050 doubles the bet with 4 GB at 14 Gbpswhile the RTX 3050 is like the 2050, but with a 128-bit memory interface for higher bandwidth.

Advertisement
Read  A Memoir Blue, analysis and opinions of the game for PC, PS4, Xbox One, PS5, Nintendo Switch and Xbox Series | 3D Games Reviews

All these adjustments make the GeForce MX570 have a power consumption of 15 to 25W. The GeForce RTX 2050 moves in the 30 – 45Wand the RTX 3060 in the 35 – 80W. When push comes to shove, you’re offering the same GPU configured differently to power three different GPUs.

More powerful iGPUs leave no room for a low-power GPU

GeForce GTX 1650 MaxQ vs Radeon RX 680M

While an NVIDIA GeForce MX570 is faster than an iGPU like the AMD Radeon 680M, the top-of-the-line Ryzen 6000 APU for laptops, it is expected to the Radeon RX 780M integrated in the Ryzen 7000 outperform dedicated GPU from NVIDIA. Although a future GeForce MX670 may be more powerful, it is not profitable. A slightly less powerful iGPU is preferable, in exchange for increase autonomy Of the device. And of course, also cost less money. Not only because of the additional chip itself, but also system desEsports Extrasand cooling.

Advertisement
Read  Drive Pilot is the first Level 3 autonomous driving system

When push comes to shove, such a basic dedicated GPU, albeit slightly more powerful, Has no sense seeing the great emphasis that is being given to iGPUs. Yes indeed, we are thinking of systems with x86 architecturebut we have already seen what the integrated graphics of the Apple Silicon, and Qualcomm is going the same way. In essence, the future depends on trying integrate all vital components (CPU, GPU and RAM) in the same silicon. It is more profitable and more efficient in every way.

Advertisement

Advertisement
Continue Reading
Advertisement
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *