Home GPU Comparison AMD Radeon RX 470 vs NVIDIA GeForce RTX 3090

AMD Radeon RX 470 vs NVIDIA GeForce RTX 3090

We compared two Desktop platform GPUs: 4GB VRAM Radeon RX 470 and 24GB VRAM GeForce RTX 3090 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

AMD Radeon RX 470 's Advantages
Lower TDP (120W vs 350W)
NVIDIA GeForce RTX 3090 's Advantages
Released 4 years and 1 months late
Boost Clock has increased by 41% (1695MHz vs 1206MHz)
More VRAM (24GB vs 4GB)
Larger VRAM bandwidth (936.2GB/s vs 211.2GB/s)
8448 additional rendering cores

Score

Benchmark

FP32 (float)
Radeon RX 470
4.94 TFLOPS
GeForce RTX 3090 +620%
35.58 TFLOPS
3DMark Time Spy
Radeon RX 470
3678
GeForce RTX 3090 +442%
19948
Shadow of the Tomb Raider 2160p
Radeon RX 470
12
GeForce RTX 3090 +675%
93
Shadow of the Tomb Raider 1440p
Radeon RX 470
25
GeForce RTX 3090 +516%
154
Shadow of the Tomb Raider 1080p
Radeon RX 470
40
GeForce RTX 3090 +400%
200
GTA 5 1440p
Radeon RX 470
34
GeForce RTX 3090 +435%
182
GTA 5 1080p
Radeon RX 470
98
GeForce RTX 3090 +113%
209
VS

Graphics Card

Aug 2016
Release Date
Sep 2020
Arctic Islands
Generation
GeForce 30
Desktop
Type
Desktop
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16

Clock Speeds

926 MHz
Base Clock
1395 MHz
1206 MHz
Boost Clock
1695 MHz
1650 MHz
Memory Clock
1219 MHz

Memory

4GB
Memory Size
24GB
GDDR5
Memory Type
GDDR6X
256bit
Memory Bus
384bit
211.2GB/s
Bandwidth
936.2GB/s

Render Config

32
Compute Units
-
-
SM Count
82
2048
Shading Units
10496
128
TMUs
328
32
ROPs
112
-
Tensor Cores
328
-
RT Cores
82
16 KB (per CU)
L1 Cache
128 KB (per SM)
2 MB
L2 Cache
6 MB
-
-
-

Theoretical Performance

38.59 GPixel/s
Pixel Rate
189.8 GPixel/s
154.4 GTexel/s
Texture Rate
556.0 GTexel/s
4.940 TFLOPS
FP16 (half)
35.58 TFLOPS
4.940 TFLOPS
FP32 (float)
35.58 TFLOPS
308.7 GFLOPS
FP64 (double)
556.0 GFLOPS

Board Design

120W
TDP
350W
300 W
Suggested PSU
750 W
1x HDMI 2.0b 3x DisplayPort 1.4a
Outputs
1x HDMI 2.1 3x DisplayPort 1.4a
1x 6-pin
Power Connectors
1x 12-pin

Graphics Processor

Ellesmere
GPU Name
GA102
Polaris 10 PRO (215-0876204)
GPU Variant
GA102-300-A1
GCN 4.0
Architecture
Ampere
GlobalFoundries
Foundry
Samsung
14 nm
Process Size
8 nm
5.7 billion
Transistors
28.3 billion
232 mm²
Die Size
628 mm²

Graphics Features

12 (12_0)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
2.1
OpenCL
3.0
1.2
Vulkan
1.3
-
CUDA
8.6
6.4
Shader Model
6.6

Related GPU Comparisons

Related News

© 2024 - TopCPU.net   Contact Us Privacy Policy