NVIDIA GeForce RTX 4070 GDDR6 vs NVIDIA B100
We compared two Desktop platform GPUs: 12GB VRAM GeForce RTX 4070 GDDR6 and 96GB VRAM B100 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4070 GDDR6 's Advantages
Boost Clock has increased by 35% (2475MHz vs 1837MHz)
Lower TDP (200W vs 1000W)
NVIDIA B100 's Advantages
More VRAM (96GB vs 12GB)
Larger VRAM bandwidth (4.10TB/s vs 480.0GB/s)
11008 additional rendering cores
Score
Benchmark
FP32 (float)
GeForce RTX 4070 GDDR6
29.15 TFLOPS
B100
+112%
62.08 TFLOPS
Graphics Card
Aug 2024
Release Date
Nov 2024
GeForce 40
Generation
Server Blackwell
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
1920 MHz
Base Clock
1665 MHz
2475 MHz
Boost Clock
1837 MHz
2500 MHz
Memory Clock
2000 MHz
Memory
12GB
Memory Size
96GB
GDDR6
Memory Type
HBM3e
192bit
Memory Bus
4096bit
480.0GB/s
Bandwidth
4.10TB/s
Render Config
-
-
-
46
SM Count
132
5888
Shading Units
16896
184
TMUs
528
64
ROPs
24
184
Tensor Cores
528
46
RT Cores
-
128 KB (per SM)
L1 Cache
256 KB (per SM)
36 MB
L2 Cache
50 MB
-
-
-
Theoretical Performance
158.4 GPixel/s
Pixel Rate
44.09 GPixel/s
455.4 GTexel/s
Texture Rate
969.9 GTexel/s
29.15 TFLOPS
FP16 (half)
248.3 TFLOPS
29.15 TFLOPS
FP32 (float)
62.08 TFLOPS
455.4 GFLOPS
FP64 (double)
31.04 TFLOPS
Board Design
200W
TDP
1000W
550 W
Suggested PSU
1400 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
No outputs
1x 16-pin
Power Connectors
-
Graphics Processor
AD104
GPU Name
GB102
AD104-2XX-A1
GPU Variant
-
Ada Lovelace
Architecture
Blackwell
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
35.8 billion
Transistors
104 billion
294 mm²
Die Size
Unknown
Graphics Features
12 Ultimate (12_2)
DirectX
N/A
4.6
OpenGL
N/A
3.0
OpenCL
3.0
1.3
Vulkan
N/A
8.9
CUDA
10.1
6.8
Shader Model
N/A
