GPU Comparison

NVIDIA

GEFORCE

NVIDIA B200

CORE STATE GB100

VRAM 90 GB

CLOCK SPEED 1965 MHz

TDP 1000 W

BUS WIDTH 4096 bit

ARCHITECTURE Blackwell

nm

PROCESS 5 nm

LAUNCH DATE —

VS

NVIDIA

GEFORCE

RTX A6000

CORE STATE GA102

VRAM 48 GB

CLOCK SPEED 1800 MHz

TDP 300 W

BUS WIDTH 384 bit

ARCHITECTURE Ampere

nm

PROCESS 8 nm

LAUNCH DATE 2020

PERFORMANCE BENCHMARKS

geekbench_opencl

345,482

193,937

geekbench_vulkan

N/A

164,462

passmark_directx_10

N/A

155

passmark_directx_11

N/A

191

passmark_directx_12

N/A

87

passmark_directx_9

N/A

245

passmark_g2d

N/A

913

passmark_g3d

N/A

22,577

passmark_gpu_compute

N/A

14,110

DETAILED SPECIFICATIONS

SPECIFICATION

B200

RTX A6000

Core Specs

Shading Units

18,944

10,752 -43.2%

Shaders

18,944

10,752 -43.2%

TMUs

592

336 -43.2%

ROPs

24

112 +366.7%

SM Count

148

84 -43.2%

Clocks

Base Clock

700 MHz

1410 MHz

Boost Clock

1965 MHz

1800 MHz

Memory Clock

2000 MHz 8 Gbps effective

2000 MHz 16 Gbps effective

Memory

Memory Size

90 GB

48 GB

VRAM (MB)

92,160

49,152 -46.7%

Memory Type

HBM3e

GDDR6

Memory Bus

4096 bit

384 bit

Bandwidth

4.10 TB/s

768.0 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

201.6 GPixel/s

Texture Rate

1,163.3 GTexel/s

604.8 GTexel/s

FP32 (TFLOPS)

74.45 TFLOPS

38.71 TFLOPS

FP64 (TFLOPS)

37.22 TFLOPS (1:2)

604.8 GFLOPS (1:64)

FP16 (TFLOPS)

1,191.2 TFLOPS (16:1)

38.71 TFLOPS (1:1)

AI/RT

RT Cores

—

84

Tensor Cores

592

336 -43.2%

Power

TDP

1000 W

300 W

TDP (W)

1,000

300 -70.0%

Suggested PSU

1400 W

700 W

Power Connectors

—

8-pin EPS

Architecture

Architecture

Blackwell

Ampere

GPU Name

GB100

GA102

Generation

Server Blackwell (Bxx)

Workstation Ampere (Ax000)

Process Size

5 nm

8 nm

Transistors

104,000 million

28,300 million

Die Size

—

628 mm²

Foundry

TSMC

Samsung

Density

—

45.1M / 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

Dual-slot

Length

—

267 mm 10.5 inches

Height

—

112 mm 4.4 inches

Outputs

No outputs

4x DisplayPort 1.4a

Bus Interface

PCIe 5.0 x16

PCIe 4.0 x16

Other

Launch Price

—

4,649 USD

Production

Active

End-of-life

Predecessor

Server Hopper

Quadro Turing

Successor

Server Rubin

Workstation Ada

View B200 Details View RTX A6000 Details