Tesla C2090 and Arc A380 - video card comparison
Tesla C2090
Arc A380
Architecture
Fermi 2.0 (2010−2014)
Xe HPG (2020−2022)
GPU code name
GF110
DG2-128
Market segment
Workstation
Desktop
Release date
July 2011
June 2022
Interface
PCIe 2.0 x16
PCIe 4.0 x8
Supplementary power connectors
-
1x 8-pin
Boost clock speed
-
2050 MHz
Core clock speed
651 MHz
2000 MHz
Floating-point performance
1,331.2 gflops
-
Manufacturing process technology
40 nm
6 nm
Number of transistors
3,000 million
7,200 million
Pipelines / CUDA cores
512
1024
Power consumption (TDP)
250 Watt
75 Watt
Texture fill rate
41.66
131.2
Memory type
GDDR5
GDDR6
Maximum RAM amount
6 GB
6 GB
Memory bus width
384 Bit
96 Bit
Memory clock speed
3696 MHz
15500 MHz
Memory bandwidth
177.4 GB/s
186.0 GB/s
Display Connectors
No outputs
1x HDMI 2.1, 3x DisplayPort 2.0
HDMI
-
+
DirectX
12 (11_0)
12 Ultimate (12_2)
CUDA
2.0
-
OpenCL
1.1
3.0
OpenGL
4.6
4.6
Shader Model
5.1
6.6
Vulkan
N/A
1.3