NVIDIA A40 PCIe vs NVIDIA RTX A4000H
We compared two Desktop platform GPUs: 48GB VRAM A40 PCIe and 16GB VRAM RTX A4000H to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA A40 PCIe 's Advantages
Boost Clock has increased by 12% (1740MHz vs 1560MHz)
More VRAM (48GB vs 16GB)
Larger VRAM bandwidth (695.8GB/s vs 448.0GB/s)
4608 additional rendering cores
NVIDIA RTX A4000H 's Advantages
Released 6 months late
Lower TDP (140W vs 300W)
Score
Benchmark
FP32 (float)
A40 PCIe
+95%
37.42 TFLOPS
RTX A4000H
19.17 TFLOPS
Graphics Card
Oct 2020
Release Date
Apr 2021
Tesla
Generation
Quadro Ampere
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1305 MHz
Base Clock
735 MHz
1740 MHz
Boost Clock
1560 MHz
1812 MHz
Memory Clock
1750 MHz
Memory
48GB
Memory Size
16GB
GDDR6
Memory Type
GDDR6
384bit
Memory Bus
256bit
695.8GB/s
Bandwidth
448.0GB/s
Render Config
-
-
-
84
SM Count
48
10752
Shading Units
6144
336
TMUs
192
112
ROPs
96
336
Tensor Cores
192
84
RT Cores
48
128 KB (per SM)
L1 Cache
128 KB (per SM)
6 MB
L2 Cache
4 MB
-
-
-
Theoretical Performance
194.9 GPixel/s
Pixel Rate
149.8 GPixel/s
584.6 GTexel/s
Texture Rate
299.5 GTexel/s
37.42 TFLOPS
FP16 (half)
19.17 TFLOPS
37.42 TFLOPS
FP32 (float)
19.17 TFLOPS
584.6 GFLOPS
FP64 (double)
299.5 GFLOPS
Board Design
300W
TDP
140W
700 W
Suggested PSU
300 W
3x DisplayPort 1.4a
Outputs
4x DisplayPort 1.4a
8-pin EPS
Power Connectors
1x 6-pin
Graphics Processor
GA102
GPU Name
GA104
-
GPU Variant
GA104-???-A1
Ampere
Architecture
Ampere
Samsung
Foundry
Samsung
8 nm
Process Size
8 nm
28.3 billion
Transistors
17.4 billion
628 mm²
Die Size
392 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
8.6
6.6
Shader Model
6.7
