A2 and L40 CNX - video card comparison
A2
L40 CNX
Architecture
Ampere (2020−2022)
Ada Lovelace
GPU code name
GA107
AD102
Market segment
Desktop
Desktop
Release date
November 2021
October 2022
Interface
PCIe 4.0 x8
PCIe 4.0 x16
Supplementary power connectors
None
1x 16-pin
Boost clock speed
1770 MHz
2475 MHz
Manufacturing process technology
8 nm
4 nm
Number of transistors
-
76,300 million
Pipelines / CUDA cores
1280
18176
Power consumption (TDP)
60 Watt
300 Watt
Texture fill rate
70.80
1,406
Memory type
GDDR6
GDDR6
Maximum RAM amount
16 GB
24 GB
Memory bus width
128 Bit
384 Bit
Memory clock speed
12.5 GB/s
18 GB/s
Memory bandwidth
200.1 GB/s
864.0 GB/s
Display Connectors
No outputs
1x HDMI 2.1, 3x DisplayPort 1.4a
HDMI
-
+
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
CUDA
8.6
8.9
OpenCL
3.0
3.0
OpenGL
4.6
4.6
Shader Model
6.6
6.7
Vulkan
1.2
1.3