GeForce RTX 3090 vs Radeon Pro 580

GeForce RTX 3090
Memory: 24576Mb GDDR6XDESKTOP
Release date: 1 Sep 2020
GeForce RTX 3090
Radeon Pro 580
Memory: 8192Mb GDDR5PRO
Release date: 5 Jun 2017
Radeon Pro 580

Summary

Reasons to consider GeForce RTX 3090

This is a much newer product, it might have better long term support.
Higher theoretical gaming performance, based on specifications.
Supports PhysX
Supports G-Sync
Supports ShadowPlay (allows game streaming/recording with minimum performance penalty)
Supports Direct3D 12 Async Compute
Supports DirectX Raytracing (DXR)
Supports Deep Learning Super-Sampling (DLSS)

Reasons to consider Radeon Pro 580

200 watts lower power draw. This might be a strong point if your current power supply is not enough to handle the GeForce RTX 3090 .
Supports Direct3D 12 Async Compute
Supports FreeSync
Supports ReLive (allows game streaming/recording with minimum performance penalty)
Supports TrueAudio
Based on an outdated architecture (AMD GCN), there may be no performance optimizations for current games and applications

GeForce RTX 3090

HWBench recommends GeForce RTX 3090

Based on theoretical specifications.


Specifications

Core Configuration

GeForce RTX 3090Radeon Pro 580
GPU NameGA102 (GA102-300-A1)vsPolaris 20 (Polaris 20 XTX)
Fab Process8 nmvs14 nm
Die Size628 mm²vs232 mm²
Transistors28,000 millionvs5,700 million
Shaders10496vs2304
Compute Units82vs36
Core clock1395 MHzvs1257 MHz
ROPs112vs32
TMUs328vs144

Memory Configuration

GeForce RTX 3090Radeon Pro 580
Memory TypeGDDR6XvsGDDR5
Bus Width384 bitvs256 bit
Memory Speed1219 MHz
19.5 Gbps effective
vs2000 MHz
8000 MHz effective
Memory Size24576 Mbvs8192 Mb

Additional details

GeForce RTX 3090Radeon Pro 580
TDP350 wattsvs150 watts
Release Date1 Sep 2020vs5 Jun 2017

Raw Performance comparison

Pixel Rate

  • GeForce RTX 3090
    189.80 GP/s
  • Radeon Pro 580
    42.88 GP/s

GigaPixels - higher is better

Texel Rate

  • GeForce RTX 3090
    556.00 GT/s
  • Radeon Pro 580
    193.00 GT/s

GigaTexels - higher is better

Memory bandwidth

  • GeForce RTX 3090
    936.20 GB/s
  • Radeon Pro 580
    256.00 GB/s

GB/s - higher is better

Single precision performance

  • GeForce RTX 3090
    35580.00 GFLOPs
  • Radeon Pro 580
    6175.00 GFLOPs

GFLOPs - higher is better