Home GPU Comparison MTT S50 vs NVIDIA PG506 232

MTT S50 vs NVIDIA PG506 232

We compared two Desktop platform GPUs: 8GB VRAM MTT S50 and 24GB VRAM PG506 232 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

MTT S50 's Advantages
Released 2 years and 6 months late
Lower TDP (85W vs 165W)
NVIDIA PG506 232 's Advantages
Boost Clock1440MHz
More VRAM (24GB vs 8GB)
1536 additional rendering cores

Score

Benchmark

FP32 (float)
MTT S50
5.2 TFLOPS
PG506 232 +98%
10.32 TFLOPS
VS

Graphics Card

Oct 2023
Release Date
Apr 2021
Generation
Tesla
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16

Clock Speeds

1200 MHz
Base Clock
930 MHz
0 MHz
Boost Clock
1440 MHz
MHz
Memory Clock
1215 MHz

Memory

8GB
Memory Size
24GB
GDDR6
Memory Type
HBM2
256bit
Memory Bus
3072bit
GB/s
Bandwidth
933.1GB/s

Render Config

-
-
-
SM Count
56
2048
Shading Units
3584
TMUs
224
ROPs
96
Tensor Cores
224
RT Cores
-
L1 Cache
192 KB (per SM)
0 MB
L2 Cache
24 MB
-
-
-

Theoretical Performance

GPixel/s
Pixel Rate
138.2 GPixel/s
GTexel/s
Texture Rate
322.6 GTexel/s
TFLOPS
FP16 (half)
10.32 TFLOPS
5.2 TFLOPS
FP32 (float)
10.32 TFLOPS
GFLOPS
FP64 (double)
5.161 TFLOPS

Board Design

85W
TDP
165W
W
Suggested PSU
450 W
1x HDMI 2.0 2x DisplayPort 1.4
Outputs
No outputs
Power Connectors
8-pin EPS

Graphics Processor

SuDi
GPU Name
GA100
GPU Variant
-
MUSA
Architecture
Ampere
TSMC
Foundry
TSMC
7 nm
Process Size
7 nm
Unknown
Transistors
54.2 billion
Unknown
Die Size
826 mm²

Graphics Features

11
DirectX
N/A
4.6
OpenGL
N/A
3.0
OpenCL
3.0
1.3
Vulkan
N/A
CUDA
8.0
Shader Model
N/A

Related GPU Comparisons

Related News

© 2024 - TopCPU.net   Contact Us Privacy Policy