GIGABYTE AORUS RTX 5090 XTREME WATERFORCE WB vs NVIDIA GeForce RTX 3050 Max-Q Refresh 4 GB

Comparison of GIGABYTE AORUS RTX 5090 XTREME WATERFORCE WB with 32 GB GDDR7 and 21,760 cores vs NVIDIA GeForce RTX 3050 Max-Q Refresh 4 GB with 4 GB GDDR6 and 2,048 cores.

Loading...

Performance Rating

GIGABYTE AORUS RTX 5090 XTREME WATERFORCE WB outperforms NVIDIA GeForce RTX 3050 Max-Q Refresh 4 GB by 1,365.62% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

GIGABYTE AORUS RTX 5090 XTREME WATERFORCE WB

46.5

GIGABYTE AORUS RTX 5090 XTREME WATERFORCE WB

46.5
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA GeForce RTX 3050 Max-Q Refresh 4 GB

3.2

NVIDIA GeForce RTX 3050 Max-Q Refresh 4 GB

3.2

Contents:

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

Memory

Memory Size

🔥 +700% 32 ГБ
4 GB

Memory Type

GDDR7 GDDR6

Memory Bandwidth

🔥 1.79 TB/s
176.0 GB/s

Memory Bus Width

512 бит 128 бит

ML Performance

FP16 (Half Precision)

🔥 +2,407% 115.5 TFLOPS
4.608 TFLOPS

BF16 (Brain Float)

No No

TF32 (TensorFloat)

No No

Compute Power

FP32 (Single Precision)

🔥 +2,407% 115.5 TFLOPS
4.608 TFLOPS

FP64 (Double Precision)

🔥 +2,407% 1.805 TFLOPS
0.072 TFLOPS

CUDA Cores

🔥 +962% 21,760
2,048

RT Cores

🔥 +962% 170
16

Architecture & Compatibility

GPU Architecture

Blackwell 2.0 Ampere

SM (Streaming Multiprocessor)

🔥 +962% 170
16

PCIe Version

PCIe 5.0 x16 PCIe 4.0 x8

ML Software Support

CUDA Version

🔥 12.0
8.6

Clocks & Performance

Base Clock

🔥 +166% 2,017
757

Boost Clock

🔥 +136% 2,655
1,125

Memory Clock

🔥 +27% 1,750
1,375

Power Consumption

Recommended PSU

950 W No

Power Connector

1x 16-pin None

TDP/TGP

575 W
🔥 -94% 35 W

Rendering

Texture Units (TMU)

🔥 +962% 680
64

ROP

🔥 +962% 170
16

L2 Cache

🔥 +4,700% 96 MB
2 MB

Additional

Slots

Dual-slot IGP

Release Date

Jan. 30, 2025 July 6, 2022

Display Outputs

1x HDMI 2.1b
3x DisplayPort 2.1b
Portable Device Dependent

Renting is cheaper than buying