GeForce 8400 GS PCI vs B100
We compared two Desktop platform GPUs: 512MB VRAM GeForce 8400 GS PCI and 96GB VRAM B100 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
GeForce 8400 GS PCI 's Advantages
Lower TDP (25W vs 1000W)
B100 's Advantages
Released 16 years and 11 months late
Boost Clock1837MHz
More VRAM (96GB vs 512GB)
Larger VRAM bandwidth (4.10TB/s vs 5.328GB/s)
16888 additional rendering cores
Score
Benchmark
FP32 (float)
GeForce 8400 GS PCI
22
B100
+282081%
62080
Graphics Card
Dec 2007
Release Date
Nov 2024
GeForce 8
Generation
Server Blackwell
Desktop
Type
Desktop
PCI
Bus Interface
PCIe 5.0 x16
Clock Speeds
-
Base Clock
1665 MHz
-
Boost Clock
1837 MHz
333 MHz
Memory Clock
2000 MHz
Memory
512MB
Memory Size
96GB
DDR2
Memory Type
HBM3e
64bit
Memory Bus
4096bit
5.328GB/s
Bandwidth
4.10TB/s
Render Config
1
SM Count
132
8
Shading Units
16896
4
TMUs
528
4
ROPs
24
-
Tensor Cores
528
-
L1 Cache
256 KB (per SM)
16 KB
L2 Cache
50 MB
Theoretical Performance
2.268 GPixel/s
Pixel Rate
44.09 GPixel/s
2.268 GTexel/s
Texture Rate
969.9 GTexel/s
-
FP16 (half)
248.3 TFLOPS
22.40 GFLOPS
FP32 (float)
62.08 TFLOPS
-
FP64 (double)
31.04 TFLOPS
Board Design
25W
TDP
1000W
200 W
Suggested PSU
1400 W
1x S-Video
1x DMS-59
Outputs
No outputs
None
Power Connectors
-
Graphics Processor
G98
GPU Name
GB102
G98-400-U2
GPU Variant
-
Tesla
Architecture
Blackwell
TSMC
Foundry
TSMC
65 nm
Process Size
5 nm
0.21 billion
Transistors
104 billion
86 mm²
Die Size
Unknown
Graphics Features
11.1 (10_0)
DirectX
N/A
3.3
OpenGL
N/A
1.1
OpenCL
3.0
N/A
Vulkan
N/A
1.1
CUDA
10.1
4.0
Shader Model
N/A
