Intel Arc Pro A50 vs NVIDIA A40 PCIe
We compared two Desktop platform GPUs: 6GB VRAM Arc Pro A50 and 48GB VRAM A40 PCIe to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Intel Arc Pro A50 's Advantages
Released 1 years and 10 months late
Boost Clock has increased by 35% (2350MHz vs 1740MHz)
Lower TDP (75W vs 300W)
NVIDIA A40 PCIe 's Advantages
More VRAM (48GB vs 6GB)
Larger VRAM bandwidth (695.8GB/s vs 192.0GB/s)
9728 additional rendering cores
Score
Benchmark
FP32 (float)
Arc Pro A50
4.813 TFLOPS
A40 PCIe
+677%
37.42 TFLOPS
Graphics Card
Aug 2022
Release Date
Oct 2020
Alchemist
Generation
Tesla
Desktop
Type
Desktop
PCIe 4.0 x8
Bus Interface
PCIe 4.0 x16
Clock Speeds
2000 MHz
Base Clock
1305 MHz
2350 MHz
Boost Clock
1740 MHz
2000 MHz
Memory Clock
1812 MHz
Memory
6GB
Memory Size
48GB
GDDR6
Memory Type
GDDR6
96bit
Memory Bus
384bit
192.0GB/s
Bandwidth
695.8GB/s
Render Config
-
-
-
-
SM Count
84
1024
Shading Units
10752
64
TMUs
336
32
ROPs
112
-
Tensor Cores
336
8
RT Cores
84
-
L1 Cache
128 KB (per SM)
4 MB
L2 Cache
6 MB
-
-
-
Theoretical Performance
75.20 GPixel/s
Pixel Rate
194.9 GPixel/s
150.4 GTexel/s
Texture Rate
584.6 GTexel/s
9.626 TFLOPS
FP16 (half)
37.42 TFLOPS
4.813 TFLOPS
FP32 (float)
37.42 TFLOPS
1203 GFLOPS
FP64 (double)
584.6 GFLOPS
Board Design
75W
TDP
300W
250 W
Suggested PSU
700 W
4x mini-DisplayPort 2.0
Outputs
3x DisplayPort 1.4a
None
Power Connectors
8-pin EPS
Graphics Processor
DG2-128
GPU Name
GA102
ACM-G11
GPU Variant
-
Generation 12.7
Architecture
Ampere
TSMC
Foundry
Samsung
6 nm
Process Size
8 nm
7.2 billion
Transistors
28.3 billion
157 mm²
Die Size
628 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
-
CUDA
8.6
6.6
Shader Model
6.6
