GPU Comparison

NVIDIA

GEFORCE

NVIDIA GeForce RTX 4090

CORE STATE AD102

VRAM 24 GB

CLOCK SPEED 2520 MHz

TDP 450 W

BUS WIDTH 384 bit

ARCHITECTURE Ada Lovelace

nm

PROCESS 5 nm

LAUNCH DATE 2022

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

9,223

N/A

geekbench_opencl

317,684

305,608

geekbench_vulkan

270,615

N/A

passmark_directx_10

224

N/A

passmark_directx_11

326

N/A

passmark_directx_12

150

N/A

passmark_directx_9

397

N/A

passmark_g2d

1,299

N/A

passmark_g3d

38,194

N/A

passmark_gpu_compute

26,613

N/A

DETAILED SPECIFICATIONS

SPECIFICATION

RTX 4090

H200 NVL

Core Specs

Shading Units

16,384

16,896 +3.1%

Shaders

16,384

16,896 +3.1%

TMUs

512

528 +3.1%

ROPs

176

24 -86.4%

SM Count

128

132 +3.1%

Clocks

Base Clock

2235 MHz

1365 MHz

Boost Clock

2520 MHz

1785 MHz

Memory Clock

1313 MHz 21 Gbps effective

1593 MHz 6.4 Gbps effective

Memory

Memory Size

24 GB

141 GB

VRAM (MB)

24,576

144,384 +487.5%

Memory Type

GDDR6X

HBM3e

Memory Bus

384 bit

6144 bit

Bandwidth

1.01 TB/s

4.89 TB/s

Cache

L1 Cache

128 KB (per SM)

256 KB (per SM)

L2 Cache

72 MB

50 MB

Performance

Pixel Rate

443.5 GPixel/s

42.84 GPixel/s

Texture Rate

1,290.2 GTexel/s

942.5 GTexel/s

FP32 (TFLOPS)

82.58 TFLOPS

60.32 TFLOPS

FP64 (TFLOPS)

1,290.2 GFLOPS (1:64)

30.16 TFLOPS (1:2)

FP16 (TFLOPS)

82.58 TFLOPS (1:1)

241.3 TFLOPS (4:1)

AI/RT

RT Cores

128

—

Tensor Cores

512

528 +3.1%

Power

TDP

450 W

600 W

TDP (W)

450

600 +33.3%

Suggested PSU

850 W

1000 W

Power Connectors

1x 16-pin

8-pin EPS

Architecture

Architecture

Ada Lovelace

Hopper

GPU Name

AD102

GH100

Generation

GeForce 40

Server Hopper (Hxx)

Process Size

5 nm

5 nm

Transistors

76,300 million

80,000 million

Die Size

609 mm²

814 mm²

Foundry

TSMC

TSMC

Density

125.3M / mm²

98.3M / mm²

API Support

DirectX

12 Ultimate (12_2)

—

OpenGL

4.6

—

Vulkan

1.4

—

OpenCL

3.0

3.0

CUDA

8.9

9.0

Shader Model

6.8

—

Physical

Slot Width

Triple-slot

Dual-slot

Length

304 mm 12 inches

267 mm 10.5 inches

Height

137 mm 5.4 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

1,599 USD

—

Production

End-of-life

Active

Predecessor

GeForce 30

Server Ada

Successor

GeForce 50

Server Blackwell

View GeForce RTX 4090 Details View H200 NVL Details