GPU Comparison
GEFORCE
NVIDIA B200
CORE STATE
GB100
VRAM
90 GB
CLOCK SPEED
1965 MHz
TDP
1000 W
BUS WIDTH
4096 bit
ARCHITECTURE
Blackwell
PROCESS
5 nm
LAUNCH DATE
—
VS
GEFORCE
GeForce RTX 4070 SUPER
CORE STATE
AD104
VRAM
12 GB
CLOCK SPEED
2475 MHz
TDP
220 W
BUS WIDTH
192 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2024
PERFORMANCE BENCHMARKS
geekbench_opencl
3dmark_3dmark_steel_nomad_dx12
geekbench_vulkan
passmark_directx_10
passmark_directx_11
passmark_directx_12
passmark_directx_9
passmark_g2d
passmark_g3d
passmark_gpu_compute
DETAILED SPECIFICATIONS
SPECIFICATION
B200
RTX 4070 SUPER
Core Specs
Shading Units
18,944
7,168
-62.2%
Shaders
18,944
7,168
-62.2%
TMUs
592
224
-62.2%
ROPs
24
80
+233.3%
SM Count
148
56
-62.2%
Clocks
Base Clock
700 MHz
1980 MHz
Boost Clock
1965 MHz
2475 MHz
Memory Clock
2000 MHz
8 Gbps effective
1313 MHz
21 Gbps effective
Memory
Memory Size
90 GB
12 GB
VRAM (MB)
92,160
12,288
-86.7%
Memory Type
HBM3e
GDDR6X
Memory Bus
4096 bit
192 bit
Bandwidth
4.10 TB/s
504.2 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
48 MB
Performance
Pixel Rate
47.16 GPixel/s
198.0 GPixel/s
Texture Rate
1,163.3 GTexel/s
554.4 GTexel/s
FP32 (TFLOPS)
74.45 TFLOPS
35.48 TFLOPS
FP64 (TFLOPS)
37.22 TFLOPS (1:2)
554.4 GFLOPS (1:64)
FP16 (TFLOPS)
1,191.2 TFLOPS (16:1)
35.48 TFLOPS (1:1)
AI/RT
RT Cores
—
56
Tensor Cores
592
224
-62.2%
Power
TDP
1000 W
220 W
TDP (W)
1,000
220
-78.0%
Suggested PSU
1400 W
550 W
Power Connectors
—
1x 16-pin
Architecture
Architecture
Blackwell
Ada Lovelace
GPU Name
GB100
AD104
Generation
Server Blackwell
(Bxx)
GeForce 40
Process Size
5 nm
5 nm
Transistors
104,000 million
35,800 million
Die Size
—
294 mm²
Foundry
TSMC
TSMC
Density
—
121.8M / mm²
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
10.0
8.9
Shader Model
—
6.8
Physical
Slot Width
SXM Module
Dual-slot
Length
—
267 mm
10.5 inches
Height
—
112 mm
4.4 inches
Outputs
No outputs
1x HDMI 2.13x DisplayPort 1.4a
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x16
Other
Launch Price
—
599 USD
Production
Active
End-of-life
Predecessor
Server Hopper
GeForce 30
Successor
Server Rubin
GeForce 50