RTX A40 and Tesla M2090 - video card comparison
RTX A40
Tesla M2090
Architecture
Ampere (2020−2022)
Fermi 2.0 (2010−2014)
GPU code name
GA102
GF110
Market segment
Desktop
Workstation
Release date
October 2020
July 2011
Interface
PCIe 4.0 x16
PCIe 2.0 x16
Supplementary power connectors
8-pin EPS
1x 6-pin + 1x 8-pin
Boost clock speed
1755 MHz
-
Core clock speed
-
651 MHz
Floating-point performance
-
1,331.2 gflops
Manufacturing process technology
8 nm
40 nm
Number of transistors
28,300 million
3,000 million
Pipelines / CUDA cores
10752
512
Power consumption (TDP)
300 Watt
250 Watt
Texture fill rate
589.7
41.66
Memory type
GDDR6
GDDR5
Maximum RAM amount
48 GB
6 GB
Memory bus width
384 Bit
384 Bit
Memory clock speed
14.5 GB/s
3696 MHz
Memory bandwidth
695.8 GB/s
177.4 GB/s
Display Connectors
3x DisplayPort
No outputs
DirectX
12 Ultimate (12_2)
12 (11_0)
CUDA
8.6
2.0
OpenCL
2.0
1.1
OpenGL
4.6
4.6
Shader Model
6.5
5.1
Vulkan
1.2
N/A