Gainward RTX 3090 Phantom GS vs NVIDIA GeForce RTX 5070 Ti SUPER

Comparison of Gainward RTX 3090 Phantom GS with 24 GB GDDR6X and 10,496 cores vs NVIDIA GeForce RTX 5070 Ti SUPER with 16 GB GDDR7 and 8,960 cores.

Loading...

Performance Rating

Gainward RTX 3090 Phantom GS outperforms NVIDIA GeForce RTX 5070 Ti SUPER by 4.15% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

Gainward RTX 3090 Phantom GS

20.8

Gainward RTX 3090 Phantom GS

20.8
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA GeForce RTX 5070 Ti SUPER

20.0

NVIDIA GeForce RTX 5070 Ti SUPER

20.0

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Additional

Memory

Memory Size

🔥 +50% 24 ГБ
16 ГБ

Memory Type

GDDR6X GDDR7

Memory Bandwidth

936.2 GB/s 896.0 GB/s

Memory Bus Width

384 бит 256 бит

ML Performance

FP16 (Half Precision)

38.73 TFLOPS
🔥 +13% 43.94 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

38.73 TFLOPS
🔥 +13% 43.94 TFLOPS

FP64 (Double Precision)

0.6052000000000001 TFLOPS
🔥 +13% 0.6866 TFLOPS

CUDA Cores

🔥 +17% 10,496
8,960

RT Cores

🔥 +17% 82
70

Architecture & Compatibility

GPU Architecture

Ampere Blackwell 2.0

SM (Streaming Multiprocessor)

🔥 +17% 82
70

PCIe Version

PCIe 4.0 x16 PCIe 5.0 x16

ML Software Support

CUDA Version

8.6
🔥 12.0

Clocks & Performance

Base Clock

1,395
🔥 +65% 2,295

Boost Clock

1,845
🔥 +33% 2,452

Memory Clock

1,219
🔥 +44% 1,750

Power Consumption

TDP/TGP

350 W 350 W

Recommended PSU

750 W 750 W

Power Connector

3x 8-pin 1x 16-pin

Rendering

Texture Units (TMU)

🔥 +17% 328
280

ROP

🔥 +17% 82
70

L2 Cache

🔥 6 MB
48 MB

Additional

Slots

Triple-slot Dual-slot

Release Date

Sept. 1, 2020 July 17, 2025

Display Outputs

1x HDMI 2.1
3x DisplayPort 1.4a
1x HDMI 2.1b
3x DisplayPort 2.1b

Renting is cheaper than buying