NVIDIA A40 PCIe vs Intel Arc A780
We compared two Desktop platform GPUs: 48GB VRAM A40 PCIe and 16GB VRAM Arc A780 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA A40 PCIe 's Advantages
More VRAM (48GB vs 16GB)
Larger VRAM bandwidth (695.8GB/s vs 559.6GB/s)
6656 additional rendering cores
Intel Arc A780 's Advantages
Boost Clock has increased by 26% (2200MHz vs 1740MHz)
Lower TDP (200W vs 300W)
Score
Benchmark
FP32 (float)
A40 PCIe
+107%
37.42 TFLOPS
Arc A780
18.02 TFLOPS
Graphics Card
Oct 2020
Release Date
Unknown
Tesla
Generation
Alchemist(Arc 7)
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1305 MHz
Base Clock
2200 MHz
1740 MHz
Boost Clock
2200 MHz
1812 MHz
Memory Clock
1093 MHz
Memory
48GB
Memory Size
16GB
GDDR6
Memory Type
GDDR6X
384bit
Memory Bus
256bit
695.8GB/s
Bandwidth
559.6GB/s
Render Config
-
-
-
84
SM Count
-
10752
Shading Units
4096
336
TMUs
256
112
ROPs
128
336
Tensor Cores
512
84
RT Cores
32
128 KB (per SM)
L1 Cache
-
6 MB
L2 Cache
16 MB
-
-
-
Theoretical Performance
194.9 GPixel/s
Pixel Rate
281.6 GPixel/s
584.6 GTexel/s
Texture Rate
563.2 GTexel/s
37.42 TFLOPS
FP16 (half)
36.04 TFLOPS
37.42 TFLOPS
FP32 (float)
18.02 TFLOPS
584.6 GFLOPS
FP64 (double)
2.253 TFLOPS
Board Design
300W
TDP
200W
700 W
Suggested PSU
550 W
3x DisplayPort 1.4a
Outputs
1x HDMI 2.13x DisplayPort 2.0
8-pin EPS
Power Connectors
2x 8-pin
Graphics Processor
GA102
GPU Name
DG2-512
-
GPU Variant
ACM-G10
Ampere
Architecture
Generation 12.7
Samsung
Foundry
TSMC
8 nm
Process Size
6 nm
28.3 billion
Transistors
21.7 billion
628 mm²
Die Size
406 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
8.6
CUDA
-
6.6
Shader Model
6.6
