NVIDIA GeForce RTX 3070 vs AMD Radeon RX 580

VS

NVIDIA GeForce RTX 3070 AMD Radeon RX 580

We compared two Desktop platform GPUs: 8GB VRAM GeForce RTX 3070 and 8GB VRAM Radeon RX 580 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA GeForce RTX 3070 's Advantages

Released 3 years and 5 months late

Boost Clock has increased by 29% (1725MHz vs 1340MHz)

Larger VRAM bandwidth (448.0GB/s vs 256.0GB/s)

3584 additional rendering cores

AMD Radeon RX 580 's Advantages

Lower TDP (185W vs 220W)

Score

Benchmark

FP32 (float)

GeForce RTX 3070 +228%

20.31 TFLOPS

Radeon RX 580

6.175 TFLOPS

3DMark Time Spy

GeForce RTX 3070 +212%

13637

Radeon RX 580

4359

Shadow of the Tomb Raider 2160p

GeForce RTX 3070 +211%

53

Radeon RX 580

17

Shadow of the Tomb Raider 1440p

GeForce RTX 3070 +171%

95

Radeon RX 580

35

Shadow of the Tomb Raider 1080p

GeForce RTX 3070 +188%

144

Radeon RX 580

50

Battlefield 5 2160p

GeForce RTX 3070 +200%

81

Radeon RX 580

27

Battlefield 5 1440p

GeForce RTX 3070 +161%

141

Radeon RX 580

54

Battlefield 5 1080p

GeForce RTX 3070 +141%

188

Radeon RX 580

78

GTA 5 1440p

GeForce RTX 3070 +75%

105

Radeon RX 580

60

GeForce RTX 3070

VS

Radeon RX 580

Graphics Card

Sep 2020

Release Date

Apr 2017

GeForce 30

Generation

Polaris

Desktop

Type

Desktop

PCIe 4.0 x16

Bus Interface

PCIe 3.0 x16

Clock Speeds

1500 MHz

Base Clock

1257 MHz

1725 MHz

Boost Clock

1340 MHz

1750 MHz

Memory Clock

2000 MHz

Memory

8GB

Memory Size

8GB

GDDR6

Memory Type

GDDR5

256bit

Memory Bus

256bit

448.0GB/s

Bandwidth

256.0GB/s

Render Config

46

SM Count

-

Compute Units

36

5888

Shading Units

2304

184

TMUs

144

96

ROPs

32

184

Tensor Cores

-

46

RT Cores

-

128 KB (per SM)

L1 Cache

16 KB (per CU)

4 MB

L2 Cache

2 MB

Theoretical Performance

165.6 GPixel/s

Pixel Rate

42.88 GPixel/s

317.4 GTexel/s

Texture Rate

193.0 GTexel/s

20.31 TFLOPS

FP16 (half)

6.175 TFLOPS

20.31 TFLOPS

FP32 (float)

6.175 TFLOPS

317.4 GFLOPS

FP64 (double)

385.9 GFLOPS

Graphics Processor

GA104

GPU Name

Polaris 20

GA104-300-A1

GPU Variant

Polaris 20 XTX (215-0910038)

Ampere

Architecture

GCN 4.0

Samsung

Foundry

GlobalFoundries

8 nm

Process Size

14 nm

17.4 billion

Transistors

5.7 billion

392 mm²

Die Size

232 mm²

Board Design

220W

TDP

185W

550 W

Suggested PSU

450 W

1x HDMI 2.1 3x DisplayPort 1.4a

Outputs

1x HDMI 2.0b 3x DisplayPort 1.4a

1x 12-pin

Power Connectors

1x 8-pin

Graphics Features

12 Ultimate (12_2)

DirectX

12 (12_0)

4.6

OpenGL

4.6

3.0

OpenCL

2.1

1.3

Vulkan

1.2

8.6

CUDA

-

6.6

Shader Model

6.4

Related GPU Comparisons

1

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 4060 Ti 8 GB

2

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 4070

3

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 4070 SUPER

4

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 3060 Ti

5

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 4060

6

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 4060 AD106

7

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce RTX 3060

8

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce GTX 275 PhysX Edition

9

NVIDIA GeForce RTX 3070 vs ATI Radeon HD 2600 PRO AGP

10

NVIDIA GeForce RTX 3070 vs NVIDIA GeForce GT 230

Related News

1

The P-Cores and E-Cores in Arrow Lake Will Be Interleaved, Reducing Core Communication Latency and Improving Heat Dissipation

2

Changes in Intel Arrow Lake P/E Core Layout: Two Major Benefits

3

Time to Step Up, AMD and Intel! Nvidia Faces Antitrust Lawsuits in France, US, and EU

4

Intel Arrow Lake-S Features Dual Direct M.2 Slots: One PCIe 5.0 x4 and One PCIe 4.0 x4

5

AMD Ryzen 9000X3D Series to Feature Fully Unlocked Overclocking; DDR5-6400: The 'Sweet Spot' for Zen 5 Memory Speeds

6

Intel Z890 Motherboard Specs Leaked: Farewell to DDR4 Memory

7

AMD Ryzen 9000 Series Now Available for Pre-order in Europe: 9950X Priced at €660, Shipping by August 2

8

AMD's Latest Breakthrough Technology Could Change the Game in the Future

9

Researchers Utilize NVIDIA Omniverse Platform to Analyze and Improve 3D Stacked Chip Designs

10

TSMC to Receive Over 60 EUV Lithography Machines Over the Next Two Years, Investing More Than $12.3 Billion