NVIDIA Ampere-based GeForce MX570 tested in OpenCL API is (again) almost as fast as RTX 2050

Published: Apr 14th 2022, 11:04 GMT   Comments

GeForce MX 570 might not be such a bad choice

The GeForce MX570 has been announced in December last year, but even 4 months later, there are still very few details available on this SKU. 

A laptop detected as “HP Zhan 66 14-inch G5” has been spotted on Geekbench website after someone ran an OpenCL test. This graphics API is far from perfect for any GPU performance evaluation, especially when gaming is considered, yet synthetic tests are all we have for now.

NVIDIA’s MX 570 is based Ampere GA107 GPU with limited memory specs. Unlike most RTX cards, this model is limited to 2 GB 64-bit memory. NVIDIA initially informed the press that MX570 will have limited ray tracing capabilities, which later turned out to be a performance limitation due to lower clocks. In that regard, NVIDIA did not disable RT or Tensor cores, so MX570 still does support raytracing acceleration as well as DLSS.

NVIDIA GeForce MX 570 Specs, Source: Geekbench

In comparison to RTX 2050, the MX570 has the same GPU and the same number of CUDA cores (2048). However, it ships with lower TDP of 15-25W, while RTX model operates within 30 to 45W range. That alone should have an impact on performance. Interestingly, that is not exactly the case based on this and previous leaks:

NVIDIA GeForce MX 570 Performance, Source: Geekbench

In OpenCL test, the MX570 is almost as fast as RTX 2050. Unfortunately, Geekbench does not provide data on GPU power, which is quite important here. The RTX 2050 can score in 45986 (average Geekbench data) to 48617 range (Xiaomi Redmi Book Pro 15 2022). The MX 570, on the other hand, scores 44683 points.

NVIDIA GeForce RTX 2050/MX550, Source: Geekbench

NVIDIA previously announced that MX550/570 laptops will be available ‘this spring’. Given how frequently they now appear in leaked tests, it shouldn’t take much time before they start shipping with the first laptops.

GeForce MX 570 OpenCL test
Graphics Card SKUOpenCL Geekbench Score (average)
GeForce GTX 1660 Ti mobile53,231
GeForce GTX 1080 mobile46,870
GeForce RTX 205045,986
GeForce MX57044,638
GeForce GTX 1650 Ti mobile43,640
GeForce MX55036,084
GeForce GTX 1060 mobile29,934

It should be noted that MX570 had already appeared in some leaks, such as the official 3DMark benchmarks from Chinese MechRevo. The company did not list MX570 directly though because it was assumed that RTX 2050 and MX570 offer nearly the same performance.

NVIDIA Entry-Level GeForce Mobile GPUs
VideoCardz.comRTX 3050RTX 2050MX570GTX 1650
ArchitectureAmpereAmpereAmpereTuring
GPUGA107GA107GA107TU117
CUDA Cores
 
2048
 
2048
 
2048
 
1024
RT Cores
 
16
 
16
 
16
Tensor Cores
 
64
 
64
 
64
Max Boost Clock
 
1740 MHz
 
1477 MHz
 
1155 MHz
 
1020 MHz
Memory Configuration
 
4GB G6 128-bit
 
4GB G6 64-bit
 
2GB G6 64-bit
 
4GB G5 128-bit
Memory Clock
 
14 Gbps
 
14 Gbps
 
12 Gbps
 
8 Gbps
TGP
 
35-80W
 
30-45W
 
15-25W
 
15-30W
Launch DateJanuary 2021Spring 2022Spring 2022April 2019

Source: Geekbench via Tom’sHardware, @BenchLeaks




Comment Policy
  1. Comments must be written in English and should not exceed 1000 characters.
  2. Comments deemed to be spam or solely promotional in nature will be deleted. Including a link to relevant content is permitted, but comments should be relevant to the post topic. Discussions about politics are not allowed on this website.
  3. Comments and usernames containing language or concepts that could be deemed offensive will be deleted.
  4. Comments complaining about the post subject or its source will be removed.
  5. A failure to comply with these rules will result in a warning and, in extreme cases, a ban. In addition, please note that comments that attack or harass an individual directly will result in a ban without warning.
  6. VideoCardz has never been sponsored by AMD, Intel, or NVIDIA. Users claiming otherwise will be banned.
  7. VideoCardz Moderating Team reserves the right to edit or delete any comments submitted to the site without notice.
  8. If you have any questions about the commenting policy, please let us know through the Contact Page.
Hide Comment Policy
Comments