NVIDIA L40 vs NVIDIA Quadro P4000
We compared two Professional market GPUs: 48GB VRAM L40 and 8GB VRAM Quadro P4000 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L40 's Advantages
Released 5 years and 8 months late
Boost Clock has increased by 68% (2490MHz vs 1480MHz)
More VRAM (48GB vs 8GB)
Larger VRAM bandwidth (864.0GB/s vs 243.3GB/s)
16384 additional rendering cores
NVIDIA Quadro P4000 's Advantages
Lower TDP (105W vs 300W)
Score
Benchmark
FP32 (float)
L40
+1606%
90.52 TFLOPS
Quadro P4000
5.304 TFLOPS
Blender
L40
+610%
3901
Quadro P4000
549
Graphics Card
Oct 2022
Release Date
Feb 2017
Tesla Ada
Generation
Quadro
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
735 MHz
Base Clock
1202 MHz
2490 MHz
Boost Clock
1480 MHz
2250 MHz
Memory Clock
1901 MHz
Memory
48GB
Memory Size
8GB
GDDR6
Memory Type
GDDR5
384bit
Memory Bus
256bit
864.0GB/s
Bandwidth
243.3GB/s
Render Config
-
-
-
142
SM Count
14
18176
Shading Units
1792
568
TMUs
112
192
ROPs
64
568
Tensor Cores
-
142
RT Cores
-
128 KB (per SM)
L1 Cache
48 KB (per SM)
96 MB
L2 Cache
2 MB
-
-
-
Theoretical Performance
478.1 GPixel/s
Pixel Rate
94.72 GPixel/s
1414 GTexel/s
Texture Rate
165.8 GTexel/s
90.52 TFLOPS
FP16 (half)
82.88 GFLOPS
90.52 TFLOPS
FP32 (float)
5.304 TFLOPS
1414 GFLOPS
FP64 (double)
165.8 GFLOPS
Board Design
300W
TDP
105W
700 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
GP104
AD102-895-A1
GPU Variant
GP104-850-A1
Ada Lovelace
Architecture
Pascal
TSMC
Foundry
TSMC
4 nm
Process Size
16 nm
76.3 billion
Transistors
7.2 billion
608 mm²
Die Size
314 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 (12_1)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
8.9
CUDA
6.1
6.6
Shader Model
6.4
