Radeon RX 470 and GeForce GTX 470 - video card comparison
Passmark - benchmark score Radeon RX 470 vs GeForce GTX 470
Radeon RX 470
8339
GeForce GTX 470
3199
Radeon RX 470
GeForce GTX 470
Architecture
Polaris (2016−2019)
Fermi (2010−2014)
GPU code name
Polaris 10 Pro
GF100
Market segment
Desktop
Desktop
Release date
August 2016
April 2010
Bus support
-
16x PCI-E 2.0
Interface
PCIe 3.0 x16
PCIe 2.0 x16
Supplementary power connectors
1x 6-pin
Two 6-pins
SLI options
-
+
Boost clock speed
1206 MHz
-
Core clock speed
926 MHz
607 MHz
CUDA cores
-
448
Floating-point performance
4,940 gflops
1,088.6 gflops
Manufacturing process technology
14 nm
40 nm
Number of transistors
5,700 million
3,100 million
Pipelines / CUDA cores
2048
448
Power consumption (TDP)
120 Watt
225 Watt
Maximum GPU temperature
-
105 °C
Texture fill rate
154.4
34.0 billion/sec
Memory type
GDDR5
GDDR5
Maximum RAM amount
8 GB
1280 MB
Memory bus width
256 Bit
320 Bit
Memory clock speed
6600 MHz
1674 MHz (3348 data rate)
Memory bandwidth
211.2 GB/s
133.9 GB/s
Shared memory
-
-
Display Connectors
1x HDMI, 3x DisplayPort
Two Dual Link DVIMini HDMI
Multi monitor support
-
+
HDMI
+
+
Maximum VGA resolution
-
2048x1536
Audio input for HDMI
-
Internal
DirectX
12 (12_0)
12 (11_0)
CUDA
-
+
OpenCL
2.0
1.1
OpenGL
4.6
4.2
Shader Model
6.4
5.1
Vulkan
1.2.131
N/A
FreeSync
+
-