GeForce GTX 580 and Radeon HD 3850 - video card comparison
Passmark - benchmark score GeForce GTX 580 vs Radeon HD 3850
GeForce GTX 580
4702
Radeon HD 3850
404
GeForce GTX 580
Radeon HD 3850
Architecture
Fermi (2010−2014)
TeraScale (2005−2013)
GPU code name
GF110
RV670
Market segment
Desktop
Desktop
Release date
November 2010
November 2007
Bus support
PCI-E 2.0 x 16
-
Interface
PCIe 2.0 x16
PCIe 2.0 x16
Supplementary power connectors
One 6-pin and One 8-pin
1x 6-pin
SLI options
+
-
Core clock speed
772 MHz
668 MHz
CUDA cores
512
-
Floating-point performance
1,581.1 gflops
427.5 gflops
Manufacturing process technology
40 nm
55 nm
Number of transistors
3,000 million
666 million
Pipelines / CUDA cores
512
320
Power consumption (TDP)
244 Watt
75 Watt
Maximum GPU temperature
97 °C
-
Texture fill rate
49.4 billion/sec
10.69
Memory type
GDDR5
GDDR3
Maximum RAM amount
1536 MB
1 GB
Memory bus width
384 Bit
256 Bit
Memory clock speed
2004 MHz (4008 data rate)
1660 MHz
Memory bandwidth
192.4 GB/s
53.12 GB/s
Display Connectors
Mini HDMITwo Dual Link DVI
2x DVI, 1x S-Video
Multi monitor support
+
-
HDMI
+
-
Maximum VGA resolution
2048x1536
-
Audio input for HDMI
Internal
-
DirectX
12 (11_0)
10.1 (10_1)
CUDA
+
-
OpenCL
1.1
N/A
OpenGL
4.2
3.3
Shader Model
5.1
4.1
Vulkan
+
N/A