NVIDIA GeForce RTX 3090 Ti vs NVIDIA GeForce RTX 4090 Mobile

VS

NVIDIA GeForce RTX 3090 Ti

NVIDIA GeForce RTX 4090 Mobile

We compared a Desktop platform GPU: 24GB VRAM GeForce RTX 3090 Ti and a Mobile platform GPU: 16GB VRAM GeForce RTX 4090 Mobile to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA GeForce RTX 3090 Ti 's Advantages

Boost Clock has increased by 10% (1860MHz vs 1695MHz)

More VRAM (24GB vs 16GB)

Larger VRAM bandwidth (1008GB/s vs 576.0GB/s)

1024 additional rendering cores

NVIDIA GeForce RTX 4090 Mobile 's Advantages

Released 1 years late

Lower TDP (80W vs 450W)

Score

Benchmark

FP32 (float)

GeForce RTX 3090 Ti +21%

40 TFLOPS

GeForce RTX 4090 Mobile

32.98 TFLOPS

3DMark Time Spy

GeForce RTX 3090 Ti +2%

22133

GeForce RTX 4090 Mobile

21586

Blender

GeForce RTX 3090 Ti

6200

GeForce RTX 4090 Mobile +24%

7692

OctaneBench

GeForce RTX 3090 Ti

649

GeForce RTX 4090 Mobile +10%

714

Cyberpunk 2077 1440p

GeForce RTX 3090 Ti +50%

87

GeForce RTX 4090 Mobile

58

GTA 5 2160p

GeForce RTX 3090 Ti

141

GeForce RTX 4090 Mobile +20%

170

GTA 5 1440p

GeForce RTX 3090 Ti +10%

187

GeForce RTX 4090 Mobile

170

GTA 5 1080p

GeForce RTX 3090 Ti +25%

226

GeForce RTX 4090 Mobile

180

GeForce RTX 3090 Ti

VS

GeForce RTX 4090 Mobile

Graphics Card

Jan 2022

Release Date

Jan 2023

GeForce 30

Generation

GeForce 40 Mobile

Desktop

Type

Mobile

PCIe 4.0 x16

Bus Interface

PCIe 4.0 x16

Clock Speeds

1560 MHz

Base Clock

1335 MHz

1860 MHz

Boost Clock

1695 MHz

1313 MHz

Memory Clock

2250 MHz

Memory

24GB

Memory Size

16GB

GDDR6X

Memory Type

GDDR6

384bit

Memory Bus

256bit

1008GB/s

Bandwidth

576.0GB/s

Render Config

-

84

SM Count

76

10752

Shading Units

9728

336

TMUs

304

112

ROPs

112

336

Tensor Cores

304

84

RT Cores

76

128 KB (per SM)

L1 Cache

128 KB (per SM)

6 MB

L2 Cache

64 MB

-

Theoretical Performance

208.3 GPixel/s

Pixel Rate

189.8 GPixel/s

625.0 GTexel/s

Texture Rate

515.3 GTexel/s

40.00 TFLOPS

FP16 (half)

32.98 TFLOPS

40.00 TFLOPS

FP32 (float)

32.98 TFLOPS

625.0 GFLOPS

FP64 (double)

515.3 GFLOPS

Board Design

450W

TDP

80-150W

850 W

Suggested PSU

-

1x HDMI 2.1 3x DisplayPort 1.4a

Outputs

Portable Device Dependent

1x 16-pin

Power Connectors

None

Graphics Processor

GA102

GPU Name

AD103

GA102-350-A1

GPU Variant

GN21-X11

Ampere

Architecture

Ada Lovelace

Samsung

Foundry

TSMC

8 nm

Process Size

5 nm

28.3 billion

Transistors

45.9 billion

628 mm²

Die Size

379 mm²

Graphics Features

12 Ultimate (12_2)

DirectX

12 Ultimate (12_2)

4.6

OpenGL

4.6

3.0

OpenCL

3.0

1.3

Vulkan

1.3

8.6

CUDA

8.9

6.6

Shader Model

6.8

Related GPU Comparisons

1

NVIDIA GeForce RTX 3090 Ti vs NVIDIA GeForce 510 OEM

2

NVIDIA GeForce RTX 3090 Ti vs ATI Radeon HD 4870

3

NVIDIA GeForce RTX 3090 Ti vs NVIDIA GeForce GTX TITAN

4

NVIDIA GeForce RTX 3090 Ti vs AMD Radeon R7 250E

5

NVIDIA GeForce RTX 3090 Ti vs ATI Radeon HD 2950 XTX

6

NVIDIA GeForce RTX 3090 Ti vs AMD Radeon R9 285X

7

NVIDIA GeForce RTX 3090 Ti vs AMD Radeon R9 360 OEM

8

NVIDIA GeForce RTX 3090 Ti vs ATI Radeon HD 4810

9

NVIDIA GeForce RTX 3090 Ti vs AMD Radeon R7 360 896SP

10

NVIDIA GeForce RTX 3090 Ti vs NVIDIA GeForce 8600 GT Mac Edition

Related News

1

AMD Unveils New RX 9070 XT Graphics Card Performance: Positioned Between RX 7900 GRE and RX 7900 XT

2

Double the Excitement! AMD FSR 4.0 to Launch Alongside RX 9070 XT Graphics Card

3

CPU-Z Minor Update: Support for Intel's Four New Products and New Memory

4

CXMT Reportedly Achieves 80% Yield in DDR5 Production

5

Intel Wildcat Lake Appears on Shipping List with FCBGA1516 Socket, Poised to Replace Alder Lake-N

6

The Commencement of ARM and Qualcomm Licensing Dispute Trial: Potential Major Impact on PC Market

7

NVIDIA Allegedly Equips RTX 5080 with 30Gbps Memory Speed, Other Models to Feature 28Gbps

8

Intel Reports High Return Rates for Devices Featuring Snapdragon X Series; Qualcomm Counters with 'Industry Standard' Claim

9

Intel Confirms Panther Lake Adopts 18A Process, ES0 Samples Already Sent to Customers

10

Complete Lineup of Core Ultra 200HX/H/U and Core 200H/U Revealed: 31 Models to Dazzle the Eyes