Radeon Pro 580 and Tesla P6 - video card comparison
Radeon Pro 580
Tesla P6
Architecture
GCN 4.0 (2016−2020)
Pascal (2016−2021)
GPU code name
Polaris 20
GP104
Market segment
Mobile workstation
Workstation
Release date
June 2017
March 2017
Interface
PCIe 3.0 x16
MXM-B (3.0)
Supplementary power connectors
None
None
Boost clock speed
1200 MHz
1506 MHz
Core clock speed
1100 MHz
1012 MHz
Floating-point performance
6,175 gflops
6,169 gflops
Manufacturing process technology
14 nm
16 nm
Number of transistors
5,700 million
7,200 million
Pipelines / CUDA cores
2304
2048
Power consumption (TDP)
150 Watt
90 Watt
Texture fill rate
172.8
192.8
Memory type
GDDR5
GDDR5
Maximum RAM amount
8 GB
16 GB
Memory bus width
256 Bit
256 Bit
Memory clock speed
6780 MHz
6008 MHz
Memory bandwidth
217.0 GB/s
192.3 GB/s
Display Connectors
No outputs
No outputs
DirectX
12 (12_0)
12 (12_1)
CUDA
-
6.1
OpenCL
2.0
1.2
OpenGL
4.6
4.6
Shader Model
6.4
6.4
Vulkan
1.2.131
1.2.131