GeForce RTX 3070 vs Arc Pro A40
We compared two Desktop platform GPUs: 8GB VRAM GeForce RTX 3070 and 6GB VRAM Arc Pro A40 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
GeForce RTX 3070 's Advantages
Boost Clock has increased by 1% (1725MHz vs 1700MHz)
More VRAM (8GB vs 6GB)
Larger VRAM bandwidth (448.0GB/s vs 192.0GB/s)
4864 additional rendering cores
Arc Pro A40 's Advantages
Released 1 years and 11 months late
Lower TDP (50W vs 220W)
Score
Benchmark
FP32 (float)
GeForce RTX 3070
+483%
20310
Arc Pro A40
3482
3DMark Time Spy
GeForce RTX 3070
+309%
13641
Arc Pro A40
3328
Graphics Card
Sep 2020
Release Date
Aug 2022
GeForce 30
Generation
Alchemist
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x8
Clock Speeds
1500 MHz
Base Clock
1500 MHz
1725 MHz
Boost Clock
1700 MHz
1750 MHz
Memory Clock
2000 MHz
Memory
8GB
Memory Size
6GB
GDDR6
Memory Type
GDDR6
256bit
Memory Bus
96bit
448.0GB/s
Bandwidth
192.0GB/s
Render Config
46
SM Count
-
5888
Shading Units
1024
184
TMUs
64
96
ROPs
32
184
Tensor Cores
-
46
RT Cores
8
128 KB (per SM)
L1 Cache
-
4 MB
L2 Cache
4 MB
Theoretical Performance
165.6 GPixel/s
Pixel Rate
54.40 GPixel/s
317.4 GTexel/s
Texture Rate
108.8 GTexel/s
20.31 TFLOPS
FP16 (half)
6.963 TFLOPS
20.31 TFLOPS
FP32 (float)
3.482 TFLOPS
317.4 GFLOPS
FP64 (double)
870.4 GFLOPS
Board Design
220W
TDP
50W
550 W
Suggested PSU
250 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
4x mini-DisplayPort 2.0
1x 12-pin
Power Connectors
None
Graphics Processor
GA104
GPU Name
DG2-128
GA104-300-A1
GPU Variant
ACM-G11
Ampere
Architecture
Generation 12.7
Samsung
Foundry
TSMC
8 nm
Process Size
6 nm
17.4 billion
Transistors
7.2 billion
392 mm²
Die Size
157 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
