RTX A40 and Data Center GPU Flex 170 - video card comparison
RTX A40
Data Center GPU Flex 170
Architecture
Ampere (2020−2022)
Generation 12.7 (2022)
GPU code name
GA102
DG2-512
Market segment
Desktop
Desktop
Release date
October 2020
August 2022
Interface
PCIe 4.0 x16
PCIe 4.0 x16
Supplementary power connectors
8-pin EPS
1x 8-pin
Boost clock speed
1755 MHz
2050 MHz
Manufacturing process technology
8 nm
6 nm
Number of transistors
28,300 million
21,700 million
Pipelines / CUDA cores
10752
4096
Power consumption (TDP)
300 Watt
150 Watt
Texture fill rate
589.7
524.8
Memory type
GDDR6
GDDR6
Maximum RAM amount
48 GB
16 GB
Memory bus width
384 Bit
256 Bit
Memory clock speed
14.5 GB/s
16 GB/s
Memory bandwidth
695.8 GB/s
512.0 GB/s
Display Connectors
3x DisplayPort
1x HDMI 2.1, 3x DisplayPort 2.0
HDMI
-
+
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
CUDA
8.6
-
OpenCL
2.0
3.0
OpenGL
4.6
4.6
Shader Model
6.5
6.6
Vulkan
1.2
1.3