GPU Comparison
GEFORCE
NVIDIA L40
CORE STATE
AD102
VRAM
48 GB
CLOCK SPEED
2490 MHz
TDP
300 W
BUS WIDTH
384 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2022
VS
GEFORCE
Quadro 4000M
CORE STATE
GF104
VRAM
2 GB
CLOCK SPEED
—
TDP
100 W
BUS WIDTH
256 bit
ARCHITECTURE
Fermi
PROCESS
40 nm
LAUNCH DATE
2011
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
L40
Quadro 4000M
Core Specs
Shading Units
18,176
336
-98.2%
Shaders
18,176
336
-98.2%
TMUs
568
56
-90.1%
ROPs
192
32
-83.3%
SM Count
142
7
-95.1%
Clocks
Base Clock
735 MHz
—
Boost Clock
2490 MHz
—
GPU Clock
—
475 MHz
Shader Clock
—
950 MHz
Memory Clock
2250 MHz
18 Gbps effective
625 MHz
2.5 Gbps effective
Memory
Memory Size
48 GB
2 GB
VRAM (MB)
49,152
2,048
-95.8%
Memory Type
GDDR6
GDDR5
Memory Bus
384 bit
256 bit
Bandwidth
864.0 GB/s
80.00 GB/s
Cache
L1 Cache
128 KB (per SM)
64 KB (per SM)
L2 Cache
96 MB
512 KB
Performance
Pixel Rate
478.1 GPixel/s
6.650 GPixel/s
Texture Rate
1,414.3 GTexel/s
26.60 GTexel/s
FP32 (TFLOPS)
90.52 TFLOPS
638.4 GFLOPS
FP64 (TFLOPS)
1,414.3 GFLOPS (1:64)
53.20 GFLOPS (1:12)
FP16 (TFLOPS)
90.52 TFLOPS (1:1)
—
AI/RT
RT Cores
142
—
Tensor Cores
568
—
Power
TDP
300 W
100 W
TDP (W)
300
100
-66.7%
Suggested PSU
700 W
—
Power Connectors
1x 16-pin
None
Architecture
Architecture
Ada Lovelace
Fermi
GPU Name
AD102
GF104
Generation
Server Ada
(Lxx)
Quadro Fermi-M
(x000M)
Process Size
5 nm
40 nm
Transistors
76,300 million
1,950 million
Die Size
609 mm²
332 mm²
Foundry
TSMC
TSMC
Density
125.3M / mm²
5.9M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.4
—
OpenCL
3.0
1.1
CUDA
8.9
2.1
Shader Model
6.8
5.1
Physical
Slot Width
Dual-slot
MXM Module
Length
267 mm
10.5 inches
—
Height
111 mm
4.4 inches
—
Outputs
4x DisplayPort 1.4a
Portable Device Dependent
Bus Interface
PCIe 4.0 x16
MXM-B (3.0)
Other
Production
End-of-life
End-of-life
Predecessor
Server Ampere
Quadro FX Mobile
Successor
Server Hopper
Quadro Kepler-M