GeForce GTX 680 vs Radeon RX 470
We compared two Desktop platform GPUs: 2GB VRAM GeForce GTX 680 and 4GB VRAM Radeon RX 470 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Radeon RX 470 's Advantages
Released 4 years and 5 months late
Boost Clock has increased by 14% (1206MHz vs 1058MHz)
More VRAM (4GB vs 2GB)
Larger VRAM bandwidth (211.2GB/s vs 192.3GB/s)
512 additional rendering cores
Lower TDP (120W vs 195W)
Score
Benchmark
FP32 (float)
GeForce GTX 680
3250
Radeon RX 470
+52%
4940
3DMark Time Spy
GeForce GTX 680
2004
Radeon RX 470
+83%
3679
Graphics Card
Mar 2012
Release Date
Aug 2016
GeForce 600
Generation
Arctic Islands
Desktop
Type
Desktop
PCIe 3.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1006 MHz
Base Clock
926 MHz
1058 MHz
Boost Clock
1206 MHz
1502 MHz
Memory Clock
1650 MHz
Memory
2GB
Memory Size
4GB
GDDR5
Memory Type
GDDR5
256bit
Memory Bus
256bit
192.3GB/s
Bandwidth
211.2GB/s
Render Config
-
Compute Units
32
1536
Shading Units
2048
128
TMUs
128
32
ROPs
32
16 KB (per SMX)
L1 Cache
16 KB (per CU)
512 KB
L2 Cache
2 MB
Theoretical Performance
33.86 GPixel/s
Pixel Rate
38.59 GPixel/s
135.4 GTexel/s
Texture Rate
154.4 GTexel/s
-
FP16 (half)
4.940 TFLOPS
3.250 TFLOPS
FP32 (float)
4.940 TFLOPS
135.4 GFLOPS
FP64 (double)
308.7 GFLOPS
Board Design
195W
TDP
120W
450 W
Suggested PSU
300 W
2x DVI
1x HDMI 1.4a
1x DisplayPort 1.2
Outputs
1x HDMI 2.0b
3x DisplayPort 1.4a
2x 6-pin
Power Connectors
1x 6-pin
Graphics Processor
GK104
GPU Name
Ellesmere
GK104-400-A2
GPU Variant
Polaris 10 PRO
(215-0876204)
Kepler
Architecture
GCN 4.0
TSMC
Foundry
GlobalFoundries
28 nm
Process Size
14 nm
3.54 billion
Transistors
5.7 billion
294 mm²
Die Size
232 mm²
Graphics Features
12 (11_0)
DirectX
12 (12_0)
4.6
OpenGL
4.6
3.0
OpenCL
2.1
1.1
Vulkan
1.2
3.0
CUDA
-
5.1
Shader Model
6.4