Radeon Pro Vega 64 and Tesla M2090 - video card comparison
Radeon Pro Vega 64
Tesla M2090
Architecture
GCN 5.0 (2017−2020)
Fermi 2.0 (2010−2014)
GPU code name
Vega 10
GF110
Market segment
Workstation
Workstation
Release date
June 2017
July 2011
Interface
PCIe 3.0 x16
PCIe 2.0 x16
Supplementary power connectors
None
1x 6-pin + 1x 8-pin
Boost clock speed
1350 MHz
-
Core clock speed
1250 MHz
651 MHz
Floating-point performance
11,059 gflops
1,331.2 gflops
Manufacturing process technology
14 nm
40 nm
Number of transistors
12,500 million
3,000 million
Pipelines / CUDA cores
4096
512
Power consumption (TDP)
250 Watt
250 Watt
Texture fill rate
345.6
41.66
Memory type
HBM2
GDDR5
Maximum RAM amount
16 GB
6 GB
Memory bus width
2048 Bit
384 Bit
Memory clock speed
1572 MHz
3696 MHz
Memory bandwidth
402.4 GB/s
177.4 GB/s
Display Connectors
No outputs
No outputs
DirectX
12 (12_1)
12 (11_0)
CUDA
-
2.0
OpenCL
2.0
1.1
OpenGL
4.6
4.6
Shader Model
6.4
5.1
Vulkan
1.1.125
N/A