NVIDIA B100 vs Moore Threads MTT S4000
We compared two Desktop platform GPUs: 96GB VRAM B100 and 48GB VRAM Moore Threads MTT S4000 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA B100 's Advantages
Boost Clock1837MHz
More VRAM (96GB vs 48GB)
Larger VRAM bandwidth (4.10TB/s vs 768.0GB/s)
8704 additional rendering cores
Moore Threads MTT S4000 's Advantages
Lower TDP (450W vs 1000W)
Score
Benchmark
FP32 (float)
B100
+152%
62.08 TFLOPS
Moore Threads MTT S4000
24.58 TFLOPS
Graphics Card
Nov 2024
Release Date
Unknown
Server Blackwell
Generation
MUSA-Chunxiao
Desktop
Type
Desktop
PCIe 5.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
1665 MHz
Base Clock
1500 MHz
1837 MHz
Boost Clock
-
2000 MHz
Memory Clock
2000 MHz
Memory
96GB
Memory Size
48GB
HBM3e
Memory Type
GDDR6
4096bit
Memory Bus
384bit
4.10TB/s
Bandwidth
768.0GB/s
Render Config
-
Compute Units
64
132
SM Count
-
16896
Shading Units
8192
528
TMUs
512
24
ROPs
512
528
Tensor Cores
128
-
-
-
256 KB (per SM)
L1 Cache
-
50 MB
L2 Cache
4 MB
-
-
-
Theoretical Performance
44.09 GPixel/s
Pixel Rate
768.0 GPixel/s
969.9 GTexel/s
Texture Rate
768.0 GTexel/s
248.3 TFLOPS
FP16 (half)
49.15 TFLOPS
62.08 TFLOPS
FP32 (float)
24.58 TFLOPS
31.04 TFLOPS
FP64 (double)
384.0 GFLOPS
Board Design
1000W
TDP
450W
1400 W
Suggested PSU
850 W
No outputs
Outputs
4x DisplayPort 1.4a
-
Power Connectors
2x 8-pin
Graphics Processor
GB102
GPU Name
MTT Gen 2
-
-
-
Blackwell
Architecture
MUSA-Chunxiao
TSMC
Foundry
TSMC
5 nm
Process Size
12 nm
104 billion
Transistors
22 billion
Unknown
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
10.1
CUDA
-
N/A
Shader Model
6.5
