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 5090

CORE STATE GB202

VRAM 32 GB

CLOCK SPEED 2407 MHz

TDP 575 W

BUS WIDTH 512 bit

ARCHITECTURE Blackwell 2.0

PROCESS 5 nm

LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

geekbench_opencl

120,027

336,524

geekbench_vulkan

114,648

374,916

3dmark_3dmark_steel_nomad_dx12

N/A

18,355

passmark_directx_10

N/A

226

passmark_directx_11

N/A

341

passmark_directx_12

N/A

185

passmark_directx_9

N/A

395

passmark_g2d

N/A

1,413

passmark_g3d

N/A

39,650

passmark_gpu_compute

N/A

26,756

Analysis: NVIDIA GB10 vs NVIDIA GeForce RTX 5090

the NVIDIA RTX 5090 and NVIDIA NVIDIA GB10 GPU deliver almost equivalent benchmark frame rates in the comparative analysis. The frame rate variations between them generally remain inside margin of error, rendering both a reasonable choice. The balance positions the decision interesting. Additional variables become particularly relevant. Value typically sways the decision.

Overall, this evaluation between the NVIDIA RTX 5090 and NVIDIA NVIDIA GB10 GPU finishes in a virtual draw. Performance variations are minor and doubtful to change actual gaming. Direct your decision on price, driver preferences, and specific functionality preferences. Each are solid picks. The draw is true.

DETAILED SPECIFICATIONS

SPECIFICATION

GB10

RTX 5090

Core Specs

Shading Units

6,144

21,760 +254.2%

Shaders

6,144

21,760 +254.2%

TMUs

384

680 +77.1%

ROPs

176 +266.7%

SM Count

170 +254.2%

Clocks

Base Clock

1665 MHz

2017 MHz

Boost Clock

2418 MHz

2407 MHz

Memory Clock

1067 MHz 8.5 Gbps effective

1750 MHz 28 Gbps effective

Memory

Memory Size

128 GB

32 GB

VRAM (MB)

131,072

32,768 -75.0%

Memory Type

LPDDR5X

GDDR7

Memory Bus

256 bit

512 bit

Bandwidth

273.2 GB/s

1.79 TB/s

Cache

L1 Cache

256 KB (per SM)

128 KB (per SM)

L2 Cache

50 MB

96 MB

Performance

Pixel Rate

116.1 GPixel/s

423.6 GPixel/s

Texture Rate

928.5 GTexel/s

1,636.8 GTexel/s

FP32 (TFLOPS)

29.71 TFLOPS

104.8 TFLOPS

FP64 (TFLOPS)

464.3 GFLOPS (1:64)

1.637 TFLOPS (1:64)

FP16 (TFLOPS)

29.71 TFLOPS (1:1)

104.8 TFLOPS (1:1)

AI/RT

RT Cores

170 +254.2%

Tensor Cores

384

680 +77.1%

Power

TDP

140 W

575 W

TDP (W)

140

575 +310.7%

Suggested PSU

300 W

950 W

Power Connectors

None

1x 16-pin

Architecture

Blackwell 2.0

GPU Name

GB20B

GB202

Generation

Server Blackwell (Bxx)

GeForce 50

Process Size

5 nm

Transistors

—

92,200 million

Die Size

—

750 mm²

Foundry

TSMC

Density

—

122.9M / mm²

API Support

DirectX

—

12 Ultimate (12_2)

OpenGL

—

4.6

Vulkan

—

1.4

OpenCL

3.0

CUDA

12.1

12.0

Shader Model

—

6.8

Physical

Slot Width

IGP

Dual-slot

Length

150 mm 5.9 inches

304 mm 12 inches

Height

51 mm 2 inches

137 mm 5.4 inches

Outputs

1x HDMI

1x HDMI 2.1b3x DisplayPort 2.1b

Bus Interface

PCIe 5.0 x16

Other

Launch Price

3,999 USD

1,999 USD

Production

Active

Predecessor

Server Hopper

GeForce 40

Successor

Server Rubin

—

View GB10 Details View GeForce RTX 5090 Details