GPU Comparison

NVIDIA

GEFORCE

NVIDIA GB10

CORE STATE GB20B

VRAM 128 GB

CLOCK SPEED 2418 MHz

TDP 140 W

BUS WIDTH 256 bit

ARCHITECTURE Blackwell 2.0

PROCESS 5 nm

LAUNCH DATE 2025

NVIDIA

GEFORCE

GeForce RTX 3090

CORE STATE GA102

VRAM 24 GB

CLOCK SPEED 1695 MHz

TDP 350 W

BUS WIDTH 384 bit

ARCHITECTURE Ampere

PROCESS 8 nm

LAUNCH DATE 2020

PERFORMANCE BENCHMARKS

geekbench_opencl

120,027

173,788

geekbench_vulkan

114,648

53,927

3dmark_3dmark_steel_nomad_dx12

N/A

5,118

passmark_directx_10

N/A

182

passmark_directx_11

N/A

220

passmark_directx_12

N/A

110

passmark_directx_9

N/A

268

passmark_g2d

N/A

1,063

passmark_g3d

N/A

26,645

passmark_gpu_compute

N/A

15,356

Analysis: NVIDIA GB10 vs NVIDIA GeForce RTX 3090

Looking at the test figures for NVIDIA RTX 3090 and NVIDIA's NVIDIA GB10 graphics, the story of competitive throughput emerges. Neither graphics card excels consistently throughout most scenarios. The swapping of advantages indicates inherent equality. Consumers confront a actual decision between them. Each cards deliver capable gaming throughput.

For buyers uncertain between NVIDIA RTX 3090 and NVIDIA's NVIDIA GB10 graphics, rest confident that each option delivers equivalent graphics throughput. This score numbers supports not a clear leader. Make your determination on additional elements and experience excellent graphics performance either way. Either are winners in this respective sense. Select comfortably.

DETAILED SPECIFICATIONS

SPECIFICATION

GB10

RTX 3090

Core Specs

Shading Units

6,144

10,496 +70.8%

Shaders

6,144

10,496 +70.8%

TMUs

384

328 -14.6%

ROPs

112 +133.3%

SM Count

82 +70.8%

Clocks

Base Clock

1665 MHz

1395 MHz

Boost Clock

2418 MHz

1695 MHz

Memory Clock

1067 MHz 8.5 Gbps effective

1219 MHz 19.5 Gbps effective

Memory

Memory Size

128 GB

24 GB

VRAM (MB)

131,072

24,576 -81.3%

Memory Type

LPDDR5X

GDDR6X

Memory Bus

256 bit

384 bit

Bandwidth

273.2 GB/s

936.2 GB/s

Cache

L1 Cache

256 KB (per SM)

128 KB (per SM)

L2 Cache

50 MB

6 MB

Performance

Pixel Rate

116.1 GPixel/s

189.8 GPixel/s

Texture Rate

928.5 GTexel/s

556.0 GTexel/s

FP32 (TFLOPS)

29.71 TFLOPS

35.58 TFLOPS

FP64 (TFLOPS)

464.3 GFLOPS (1:64)

556.0 GFLOPS (1:64)

FP16 (TFLOPS)

29.71 TFLOPS (1:1)

35.58 TFLOPS (1:1)

AI/RT

RT Cores

82 +70.8%

Tensor Cores

384

328 -14.6%

Power

TDP

140 W

350 W

TDP (W)

140

350 +150.0%

Suggested PSU

300 W

750 W

Power Connectors

None

1x 12-pin

Architecture

Blackwell 2.0

Ampere

GPU Name

GB20B

GA102

Generation

Server Blackwell (Bxx)

GeForce 30

Process Size

5 nm

8 nm

Transistors

—

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

CUDA

12.1

8.6

Shader Model

—

6.8

Physical

Slot Width

IGP

Triple-slot

Length

150 mm 5.9 inches

336 mm 13.2 inches

Height

51 mm 2 inches

140 mm 5.5 inches

Outputs

1x HDMI

1x HDMI 2.13x DisplayPort 1.4a

Bus Interface

PCIe 5.0 x16

PCIe 4.0 x16

Other

Launch Price

3,999 USD

1,499 USD

Production

Active

End-of-life

Predecessor

Server Hopper

GeForce 20

Successor

Server Rubin

GeForce 40

View GB10 Details View GeForce RTX 3090 Details