GPU Comparison

NVIDIA

GEFORCE

NVIDIA GeForce RTX 3070

CORE STATE GA104

VRAM 8 GB

CLOCK SPEED 1725 MHz

TDP 220 W

BUS WIDTH 256 bit

ARCHITECTURE Ampere

nm

PROCESS 8 nm

LAUNCH DATE 2020

VS

NVIDIA

GEFORCE

H200 NVL

CORE STATE GH100

VRAM 141 GB

CLOCK SPEED 1785 MHz

TDP 600 W

BUS WIDTH 6144 bit

ARCHITECTURE Hopper

nm

PROCESS 5 nm

LAUNCH DATE 2024

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12

3,162

N/A

geekbench_opencl

123,479

305,608

geekbench_vulkan

120,664

N/A

passmark_directx_10

150

N/A

passmark_directx_11

182

N/A

passmark_directx_12

85

N/A

passmark_directx_9

247

N/A

passmark_g2d

1,001

N/A

passmark_g3d

22,214

N/A

passmark_gpu_compute

11,195

N/A

DETAILED SPECIFICATIONS

SPECIFICATION

RTX 3070

H200 NVL

Core Specs

Shading Units

5,888

16,896 +187.0%

Shaders

5,888

16,896 +187.0%

TMUs

184

528 +187.0%

ROPs

96

24 -75.0%

SM Count

46

132 +187.0%

Clocks

Base Clock

1500 MHz

1365 MHz

Boost Clock

1725 MHz

1785 MHz

Memory Clock

1750 MHz 14 Gbps effective

1593 MHz 6.4 Gbps effective

Memory

Memory Size

8 GB

141 GB

VRAM (MB)

8,192

144,384 +1662.5%

Memory Type

GDDR6

HBM3e

Memory Bus

256 bit

6144 bit

Bandwidth

448.0 GB/s

4.89 TB/s

Cache

L1 Cache

128 KB (per SM)

256 KB (per SM)

L2 Cache

4 MB

50 MB

Performance

Pixel Rate

165.6 GPixel/s

42.84 GPixel/s

Texture Rate

317.4 GTexel/s

942.5 GTexel/s

FP32 (TFLOPS)

20.31 TFLOPS

60.32 TFLOPS

FP64 (TFLOPS)

317.4 GFLOPS (1:64)

30.16 TFLOPS (1:2)

FP16 (TFLOPS)

20.31 TFLOPS (1:1)

241.3 TFLOPS (4:1)

AI/RT

RT Cores

46

—

Tensor Cores

184

528 +187.0%

Power

TDP

220 W

600 W

TDP (W)

220

600 +172.7%

Suggested PSU

550 W

1000 W

Power Connectors

1x 12-pin

8-pin EPS

Architecture

Architecture

Ampere

Hopper

GPU Name

GA104

GH100

Generation

GeForce 30

Server Hopper (Hxx)

Process Size

8 nm

5 nm

Transistors

17,400 million

80,000 million

Die Size

392 mm²

814 mm²

Foundry

Samsung

TSMC

Density

44.4M / mm²

98.3M / mm²

API Support

DirectX

12 Ultimate (12_2)

—

OpenGL

4.6

—

Vulkan

1.4

—

OpenCL

3.0

3.0

CUDA

8.6

9.0

Shader Model

6.8

—

Physical

Slot Width

Dual-slot

Dual-slot

Length

242 mm 9.5 inches

267 mm 10.5 inches

Height

112 mm 4.4 inches

111 mm 4.4 inches

Outputs

1x HDMI 2.13x DisplayPort 1.4a

No outputs

Bus Interface

PCIe 4.0 x16

PCIe 5.0 x16

Other

Launch Price

499 USD

—

Production

End-of-life

Active

Predecessor

GeForce 20

Server Ada

Successor

GeForce 40

Server Blackwell

View GeForce RTX 3070 Details View H200 NVL Details