NVIDIA B100 vs NVIDIA P104 101
We compared two Desktop platform GPUs: 96GB VRAM B100 and 4GB VRAM P104 101 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA B100 's Advantages
Released 6 years and 10 months late
Boost Clock has increased by 9% (1837MHz vs 1683MHz)
More VRAM (96GB vs 4GB)
Larger VRAM bandwidth (4.10TB/s vs 256.3GB/s)
14336 additional rendering cores
NVIDIA P104 101 's Advantages
Lower TDP (125W vs 1000W)
Score
Benchmark
FP32 (float)
B100
+620%
62.08 TFLOPS
P104 101
8.617 TFLOPS
Graphics Card
Nov 2024
Release Date
Jan 2018
Server Blackwell
Generation
Mining GPUs
Desktop
Type
Desktop
PCIe 5.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1665 MHz
Base Clock
1506 MHz
1837 MHz
Boost Clock
1683 MHz
2000 MHz
Memory Clock
2002 MHz
Memory
96GB
Memory Size
4GB
HBM3e
Memory Type
GDDR5
4096bit
Memory Bus
256bit
4.10TB/s
Bandwidth
256.3GB/s
Render Config
-
-
-
132
SM Count
20
16896
Shading Units
2560
528
TMUs
160
24
ROPs
64
528
Tensor Cores
-
-
-
-
256 KB (per SM)
L1 Cache
48 KB (per SM)
50 MB
L2 Cache
2 MB
-
-
-
Theoretical Performance
44.09 GPixel/s
Pixel Rate
107.7 GPixel/s
969.9 GTexel/s
Texture Rate
269.3 GTexel/s
248.3 TFLOPS
FP16 (half)
134.6 GFLOPS
62.08 TFLOPS
FP32 (float)
8.617 TFLOPS
31.04 TFLOPS
FP64 (double)
269.3 GFLOPS
Board Design
1000W
TDP
125W
1400 W
Suggested PSU
200 W
No outputs
Outputs
No outputs
-
Power Connectors
1x 8-pin
Graphics Processor
GB102
GPU Name
GP104
-
GPU Variant
GP104-101-A1
Blackwell
Architecture
Pascal
TSMC
Foundry
TSMC
5 nm
Process Size
16 nm
104 billion
Transistors
7.2 billion
Unknown
Die Size
314 mm²
Graphics Features
N/A
DirectX
12 (12_1)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
10.1
CUDA
6.1
N/A
Shader Model
6.4
