NVIDIA PG506 242 vs Moore Threads MTT S1000M
We compared two Desktop platform GPUs: 24GB VRAM PG506 242 and 4GB VRAM Moore Threads MTT S1000M to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA PG506 242 's Advantages
Boost Clock1440MHz
More VRAM (24GB vs 4GB)
Larger VRAM bandwidth (933.1GB/s vs 224.0GB/s)
2560 additional rendering cores
Moore Threads MTT S1000M 's Advantages
Released 1 years and 3 months late
Lower TDP (35W vs 165W)
Score
Benchmark
FP32 (float)
PG506 242
+403%
10.32 TFLOPS
Moore Threads MTT S1000M
2.048 TFLOPS
Graphics Card
Apr 2021
Release Date
Jul 2022
Tesla
Generation
MUSA-Chunxiao
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x8
Clock Speeds
930 MHz
Base Clock
1000 MHz
1440 MHz
Boost Clock
-
1215 MHz
Memory Clock
1750 MHz
Memory
24GB
Memory Size
4GB
HBM2
Memory Type
GDDR6
3072bit
Memory Bus
128bit
933.1GB/s
Bandwidth
224.0GB/s
Render Config
-
Compute Units
8
56
SM Count
-
3584
Shading Units
1024
224
TMUs
64
96
ROPs
64
224
Tensor Cores
-
-
-
-
192 KB (per SM)
L1 Cache
-
24 MB
L2 Cache
1024 KB
-
-
-
Theoretical Performance
138.2 GPixel/s
Pixel Rate
64.00 GPixel/s
322.6 GTexel/s
Texture Rate
64.00 GTexel/s
10.32 TFLOPS
FP16 (half)
4.096 TFLOPS
10.32 TFLOPS
FP32 (float)
2.048 TFLOPS
5.161 TFLOPS
FP64 (double)
32.00 GFLOPS
Board Design
165W
TDP
35W
450 W
Suggested PSU
200 W
No outputs
Outputs
2x DisplayPort 1.2
8-pin EPS
Power Connectors
None
Graphics Processor
GA100
GPU Name
MTT Gen 1
-
-
-
Ampere
Architecture
MUSA-Chunxiao
TSMC
Foundry
TSMC
7 nm
Process Size
12 nm
54.2 billion
Transistors
Unknown
826 mm²
Die Size
Unknown
Graphics Features
N/A
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.4
8.0
CUDA
-
N/A
Shader Model
6.5
