NVIDIA L20 vs NVIDIA RTX A4000H
We compared two Desktop platform GPUs: 48GB VRAM L20 and 16GB VRAM RTX A4000H to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L20 's Advantages
Released 2 years and 7 months late
Boost Clock has increased by 62% (2520MHz vs 1560MHz)
More VRAM (48GB vs 16GB)
Larger VRAM bandwidth (864.0GB/s vs 448.0GB/s)
5632 additional rendering cores
NVIDIA RTX A4000H 's Advantages
Lower TDP (140W vs 275W)
Score
Benchmark
FP32 (float)
L20
+209%
59.35 TFLOPS
RTX A4000H
19.17 TFLOPS
Graphics Card
Nov 2023
Release Date
Apr 2021
Tesla Ada
Generation
Quadro Ampere
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1440 MHz
Base Clock
735 MHz
2520 MHz
Boost Clock
1560 MHz
2250 MHz
Memory Clock
1750 MHz
Memory
48GB
Memory Size
16GB
GDDR6
Memory Type
GDDR6
384bit
Memory Bus
256bit
864.0GB/s
Bandwidth
448.0GB/s
Render Config
-
-
-
92
SM Count
48
11776
Shading Units
6144
368
TMUs
192
128
ROPs
96
368
Tensor Cores
192
92
RT Cores
48
128 KB (per SM)
L1 Cache
128 KB (per SM)
96 MB
L2 Cache
4 MB
-
-
-
Theoretical Performance
322.6 GPixel/s
Pixel Rate
149.8 GPixel/s
927.4 GTexel/s
Texture Rate
299.5 GTexel/s
59.35 TFLOPS
FP16 (half)
19.17 TFLOPS
59.35 TFLOPS
FP32 (float)
19.17 TFLOPS
927.4 GFLOPS
FP64 (double)
299.5 GFLOPS
Board Design
275W
TDP
140W
600 W
Suggested PSU
300 W
4x DisplayPort 1.4a
Outputs
4x DisplayPort 1.4a
1x 16-pin
Power Connectors
1x 6-pin
Graphics Processor
AD102
GPU Name
GA104
AD102
GPU Variant
GA104-???-A1
Ada Lovelace
Architecture
Ampere
TSMC
Foundry
Samsung
5 nm
Process Size
8 nm
76.3 billion
Transistors
17.4 billion
609 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
8.9
CUDA
8.6
6.7
Shader Model
6.7
