NVIDIA A40 PCIe vs Moore Threads MTT S2000
We compared two Desktop platform GPUs: 48GB VRAM A40 PCIe and 32GB VRAM Moore Threads MTT S2000 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA A40 PCIe 's Advantages
Boost Clock1740MHz
More VRAM (48GB vs 32GB)
Larger VRAM bandwidth (695.8GB/s vs 448.0GB/s)
6656 additional rendering cores
Moore Threads MTT S2000 's Advantages
Lower TDP (150W vs 300W)
Score
Benchmark
FP32 (float)
A40 PCIe
+251%
37.42 TFLOPS
Moore Threads MTT S2000
10.65 TFLOPS
Graphics Card
Oct 2020
Release Date
Unknown
Tesla
Generation
MUSA-Chunxiao
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1305 MHz
Base Clock
1300 MHz
1740 MHz
Boost Clock
-
1812 MHz
Memory Clock
1750 MHz
Memory
48GB
Memory Size
32GB
GDDR6
Memory Type
GDDR6
384bit
Memory Bus
256bit
695.8GB/s
Bandwidth
448.0GB/s
Render Config
-
Compute Units
32
84
SM Count
-
10752
Shading Units
4096
336
TMUs
256
112
ROPs
256
336
Tensor Cores
-
84
RT Cores
-
128 KB (per SM)
L1 Cache
-
6 MB
L2 Cache
4 MB
-
-
-
Theoretical Performance
194.9 GPixel/s
Pixel Rate
332.8 GPixel/s
584.6 GTexel/s
Texture Rate
332.8 GTexel/s
37.42 TFLOPS
FP16 (half)
21.30 TFLOPS
37.42 TFLOPS
FP32 (float)
10.65 TFLOPS
584.6 GFLOPS
FP64 (double)
166.4 GFLOPS
Board Design
300W
TDP
150W
700 W
Suggested PSU
450 W
3x DisplayPort 1.4a
Outputs
2x DisplayPort 1.2
8-pin EPS
Power Connectors
1x 8-pin
Graphics Processor
GA102
GPU Name
MTT Gen 1
-
-
-
Ampere
Architecture
MUSA-Chunxiao
Samsung
Foundry
TSMC
8 nm
Process Size
12 nm
28.3 billion
Transistors
Unknown
628 mm²
Die Size
Unknown
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.4
8.6
CUDA
-
6.6
Shader Model
6.5
