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
RTX A1000
CORE STATE
GA107
VRAM
8 GB
CLOCK SPEED
1462 MHz
TDP
50 W
BUS WIDTH
128 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2024
PERFORMANCE BENCHMARKS
geekbench_opencl
3dmark_3dmark_steel_nomad_dx12
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
B200
RTX A1000
Core Specs
Shading Units
18,944
2,304
-87.8%
Shaders
18,944
2,304
-87.8%
TMUs
592
72
-87.8%
ROPs
24
32
+33.3%
SM Count
148
18
-87.8%
Clocks
Base Clock
700 MHz
727 MHz
Boost Clock
1965 MHz
1462 MHz
Memory Clock
2000 MHz
8 Gbps effective
1500 MHz
12 Gbps effective
Memory
Memory Size
90 GB
8 GB
VRAM (MB)
92,160
8,192
-91.1%
Memory Type
HBM3e
GDDR6
Memory Bus
4096 bit
128 bit
Bandwidth
4.10 TB/s
192.0 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
2 MB
Performance
Pixel Rate
47.16 GPixel/s
46.78 GPixel/s
Texture Rate
1,163.3 GTexel/s
105.3 GTexel/s
FP32 (TFLOPS)
74.45 TFLOPS
6.737 TFLOPS
FP64 (TFLOPS)
37.22 TFLOPS (1:2)
105.3 GFLOPS (1:64)
FP16 (TFLOPS)
1,191.2 TFLOPS (16:1)
6.737 TFLOPS (1:1)
AI/RT
RT Cores
—
18
Tensor Cores
592
72
-87.8%
Power
TDP
1000 W
50 W
TDP (W)
1,000
50
-95.0%
Suggested PSU
1400 W
250 W
Power Connectors
—
None
Architecture
Architecture
Blackwell
Ampere
GPU Name
GB100
GA107
Generation
Server Blackwell
(Bxx)
Workstation Ampere
(Ax000)
Process Size
5 nm
8 nm
Transistors
104,000 million
8,700 million
Die Size
—
200 mm²
Foundry
TSMC
Samsung
Density
—
43.5M / mm²
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
10.0
8.6
Shader Model
—
6.8
Physical
Slot Width
SXM Module
Single-slot
Length
—
163 mm
6.4 inches
Height
—
69 mm
2.7 inches
Outputs
No outputs
4x mini-DisplayPort 1.4a
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x8
Other
Production
Active
Active
Predecessor
Server Hopper
Quadro Turing
Successor
Server Rubin
Workstation Ada