NVIDIA L40 vs NVIDIA Quadro K620M
We compared two Professional market GPUs: 48GB VRAM L40 and 2GB VRAM Quadro K620M to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L40 's Advantages
Released 7 years and 7 months late
Boost Clock has increased by 122% (2490MHz vs 1124MHz)
More VRAM (48GB vs 2GB)
Larger VRAM bandwidth (864.0GB/s vs 14.40GB/s)
17792 additional rendering cores
NVIDIA Quadro K620M 's Advantages
Lower TDP (30W vs 300W)
Score
Benchmark
FP32 (float)
L40
+10388%
90.52 TFLOPS
Quadro K620M
0.863 TFLOPS
Blender
L40
+4657%
3901
Quadro K620M
82
Graphics Card
Oct 2022
Release Date
Mar 2015
Tesla Ada
Generation
Quadro Mobile
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
MXM-A (3.0)
Clock Speeds
735 MHz
Base Clock
1029 MHz
2490 MHz
Boost Clock
1124 MHz
2250 MHz
Memory Clock
900 MHz
Memory
48GB
Memory Size
2GB
GDDR6
Memory Type
DDR3
384bit
Memory Bus
64bit
864.0GB/s
Bandwidth
14.40GB/s
Render Config
-
-
-
142
SM Count
-
18176
Shading Units
384
568
TMUs
16
192
ROPs
8
568
Tensor Cores
-
142
RT Cores
-
128 KB (per SM)
L1 Cache
64 KB (per SMM)
96 MB
L2 Cache
1024 KB
-
-
-
Theoretical Performance
478.1 GPixel/s
Pixel Rate
8.992 GPixel/s
1414 GTexel/s
Texture Rate
17.98 GTexel/s
90.52 TFLOPS
FP16 (half)
-
90.52 TFLOPS
FP32 (float)
863.2 GFLOPS
1414 GFLOPS
FP64 (double)
26.98 GFLOPS
Board Design
300W
TDP
30W
700 W
Suggested PSU
-
4x DisplayPort 1.4a
Outputs
No outputs
1x 16-pin
Power Connectors
None
Graphics Processor
AD102
GPU Name
GM108
AD102-895-A1
GPU Variant
N15M-Q3
Ada Lovelace
Architecture
Maxwell
TSMC
Foundry
TSMC
4 nm
Process Size
28 nm
76.3 billion
Transistors
Unknown
608 mm²
Die Size
Unknown
Graphics Features
12 Ultimate (12_2)
DirectX
12 (11_0)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
8.9
CUDA
5.0
6.6
Shader Model
5.1
