NVIDIA L20 vs NVIDIA GeForce RTX 4090
We compared two Desktop platform GPUs: 48GB VRAM L20 and 24GB VRAM GeForce RTX 4090 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L20 's Advantages
Released 1 years and 2 months late
More VRAM (48GB vs 24GB)
Lower TDP (275W vs 450W)
NVIDIA GeForce RTX 4090 's Advantages
Larger VRAM bandwidth (1.01TB/s vs 864.0GB/s)
4608 additional rendering cores
Score
Benchmark
FP32 (float)
L20
59.35 TFLOPS
GeForce RTX 4090
+39%
82.58 TFLOPS
Graphics Card
Nov 2023
Release Date
Sep 2022
Tesla Ada
Generation
GeForce 40
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1440 MHz
Base Clock
2235 MHz
2520 MHz
Boost Clock
2520 MHz
2250 MHz
Memory Clock
1313 MHz
Memory
48GB
Memory Size
24GB
GDDR6
Memory Type
GDDR6X
384bit
Memory Bus
384bit
864.0GB/s
Bandwidth
1.01TB/s
Render Config
-
-
-
92
SM Count
128
11776
Shading Units
16384
368
TMUs
512
128
ROPs
176
368
Tensor Cores
512
92
RT Cores
128
128 KB (per SM)
L1 Cache
128 KB (per SM)
96 MB
L2 Cache
72 MB
-
-
-
Theoretical Performance
322.6 GPixel/s
Pixel Rate
443.5 GPixel/s
927.4 GTexel/s
Texture Rate
1290 GTexel/s
59.35 TFLOPS
FP16 (half)
82.58 TFLOPS
59.35 TFLOPS
FP32 (float)
82.58 TFLOPS
927.4 GFLOPS
FP64 (double)
1290 GFLOPS
Board Design
275W
TDP
450W
600 W
Suggested PSU
850 W
4x DisplayPort 1.4a
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
1x 16-pin
Power Connectors
1x 16-pin
Graphics Processor
AD102
GPU Name
AD102
AD102
GPU Variant
AD102-300-A1
Ada Lovelace
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
76.3 billion
Transistors
76.3 billion
609 mm²
Die Size
609 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
8.9
6.7
Shader Model
6.8
