NVIDIA CMP 50HX vs NVIDIA L20
We compared two Desktop platform GPUs: 10GB VRAM CMP 50HX and 48GB VRAM L20 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA CMP 50HX 's Advantages
Lower TDP (250W vs 275W)
NVIDIA L20 's Advantages
Released 2 years and 5 months late
Boost Clock has increased by 63% (2520MHz vs 1545MHz)
More VRAM (48GB vs 10GB)
Larger VRAM bandwidth (864.0GB/s vs 560.0GB/s)
8192 additional rendering cores
Score
Benchmark
FP32 (float)
CMP 50HX
11.07 TFLOPS
L20
+436%
59.35 TFLOPS
Graphics Card
Jun 2021
Release Date
Nov 2023
Mining GPUs
Generation
Tesla Ada
Desktop
Type
Desktop
PCIe 1.0 x4
Bus Interface
PCIe 4.0 x16
Clock Speeds
1350 MHz
Base Clock
1440 MHz
1545 MHz
Boost Clock
2520 MHz
1750 MHz
Memory Clock
2250 MHz
Memory
10GB
Memory Size
48GB
GDDR6
Memory Type
GDDR6
320bit
Memory Bus
384bit
560.0GB/s
Bandwidth
864.0GB/s
Render Config
-
-
-
56
SM Count
92
3584
Shading Units
11776
192
TMUs
368
80
ROPs
128
448
Tensor Cores
368
56
RT Cores
92
64 KB (per SM)
L1 Cache
128 KB (per SM)
5 MB
L2 Cache
96 MB
-
-
-
Theoretical Performance
123.6 GPixel/s
Pixel Rate
322.6 GPixel/s
296.6 GTexel/s
Texture Rate
927.4 GTexel/s
22.15 TFLOPS
FP16 (half)
59.35 TFLOPS
11.07 TFLOPS
FP32 (float)
59.35 TFLOPS
346.1 GFLOPS
FP64 (double)
927.4 GFLOPS
Board Design
250W
TDP
275W
600 W
Suggested PSU
600 W
No outputs
Outputs
4x DisplayPort 1.4a
2x 8-pin
Power Connectors
1x 16-pin
Graphics Processor
TU102
GPU Name
AD102
TU102-100-A1
GPU Variant
AD102
Turing
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
12 nm
Process Size
5 nm
18.6 billion
Transistors
76.3 billion
754 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
7.5
CUDA
8.9
6.8
Shader Model
6.7
