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
TITAN V CEO Edition
CORE STATE
GV100
VRAM
32 GB
CLOCK SPEED
1455 MHz
TDP
250 W
BUS WIDTH
4096 bit
ARCHITECTURE
Volta
PROCESS
12 nm
LAUNCH DATE
2018
PERFORMANCE BENCHMARKS
geekbench_opencl
passmark_g2d
passmark_g3d
DETAILED SPECIFICATIONS
SPECIFICATION
B200
TITAN V CEO Edition
Core Specs
Shading Units
18,944
5,120
-73.0%
Shaders
18,944
5,120
-73.0%
TMUs
592
320
-45.9%
ROPs
24
128
+433.3%
SM Count
148
80
-45.9%
Clocks
Base Clock
700 MHz
1200 MHz
Boost Clock
1965 MHz
1455 MHz
Memory Clock
2000 MHz
8 Gbps effective
848 MHz
1696 Mbps effective
Memory
Memory Size
90 GB
32 GB
VRAM (MB)
92,160
32,768
-64.4%
Memory Type
HBM3e
HBM2
Memory Bus
4096 bit
4096 bit
Bandwidth
4.10 TB/s
868.4 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
6 MB
Performance
Pixel Rate
47.16 GPixel/s
186.2 GPixel/s
Texture Rate
1,163.3 GTexel/s
465.6 GTexel/s
FP32 (TFLOPS)
74.45 TFLOPS
14.90 TFLOPS
FP64 (TFLOPS)
37.22 TFLOPS (1:2)
7.450 TFLOPS (1:2)
FP16 (TFLOPS)
1,191.2 TFLOPS (16:1)
29.80 TFLOPS (2:1)
AI/RT
Tensor Cores
592
640
+8.1%
Power
TDP
1000 W
250 W
TDP (W)
1,000
250
-75.0%
Suggested PSU
1400 W
600 W
Power Connectors
—
1x 6-pin + 1x 8-pin
Architecture
Architecture
Blackwell
Volta
GPU Name
GB100
GV100
Generation
Server Blackwell
(Bxx)
GeForce 10
Process Size
5 nm
12 nm
Transistors
104,000 million
21,100 million
Die Size
—
815 mm²
Foundry
TSMC
TSMC
Density
—
25.9M / mm²
API Support
DirectX
—
12 (12_1)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
10.0
7.0
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 2.03x DisplayPort 1.4a
Bus Interface
PCIe 5.0 x16
PCIe 3.0 x16
Other
Production
Active
End-of-life
Predecessor
Server Hopper
GeForce 900
Successor
Server Rubin
GeForce 20