NVIDIA L40 vs NVIDIA GeForce RTX 4090

Comparative analysis of NVIDIA L40 and NVIDIA GeForce RTX 4090 videocards for all known characteristics in the following categories: Essentials, Technical info, Video outputs and ports, Compatibility, dimensions and requirements, API support, Memory. Benchmark videocards performance analysis: Geekbench - OpenCL, PassMark - G2D Mark, PassMark - G3D Mark, 3DMark Fire Strike - Graphics Score, CompuBench 1.5 Desktop - Face Detection (mPixels/s), CompuBench 1.5 Desktop - T-Rex (Frames/s), CompuBench 1.5 Desktop - Video Composition (Frames/s), CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s), CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s).

 

Differences

Reasons to consider the NVIDIA L40

  • Around 10% higher texture fill rate: 1,414 GTexel/s vs 1,290 GTexel/s
  • Around 11% higher pipelines: 18176 vs 16384
  • Around 50% lower typical power consumption: 300 Watt vs 450 Watt
  • 2x more maximum memory size: 48 GB vs 24 GB
  • Around 71% higher memory clock speed: 2250 MHz, 18 Gbps effective vs 1313 MHz, 21 Gbps effective
  • Around 4% better performance in Geekbench - OpenCL: 331026 vs 317791
Specifications (specs)
Texture fill rate 1,414 GTexel/s vs 1,290 GTexel/s
Pipelines 18176 vs 16384
Thermal Design Power (TDP) 300 Watt vs 450 Watt
Maximum memory size 48 GB vs 24 GB
Memory clock speed 2250 MHz, 18 Gbps effective vs 1313 MHz, 21 Gbps effective
Benchmarks
Geekbench - OpenCL 331026 vs 317791

Reasons to consider the NVIDIA GeForce RTX 4090

  • 3x more core clock speed: 2235 MHz vs 735 MHz
  • Around 1% higher boost clock speed: 2520 MHz vs 2490 MHz
  • A newer manufacturing process allows for a more powerful, yet cooler running videocard: 4 nm vs 5 nm
Core clock speed 2235 MHz vs 735 MHz
Boost clock speed 2520 MHz vs 2490 MHz
Manufacturing process technology 4 nm vs 5 nm

Compare benchmarks

GPU 1: NVIDIA L40
GPU 2: NVIDIA GeForce RTX 4090

Geekbench - OpenCL
GPU 1
GPU 2
331026
317791
Name NVIDIA L40 NVIDIA GeForce RTX 4090
Geekbench - OpenCL 331026 317791
PassMark - G2D Mark 1297
PassMark - G3D Mark 38287
3DMark Fire Strike - Graphics Score 9223
CompuBench 1.5 Desktop - Face Detection (mPixels/s) 582.642
CompuBench 1.5 Desktop - T-Rex (Frames/s) 98.472
CompuBench 1.5 Desktop - Video Composition (Frames/s) 178.756
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s) 2968.159
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s) 0

Compare specifications (specs)

NVIDIA L40 NVIDIA GeForce RTX 4090

Essentials

Architecture Ada Lovelace Ada Lovelace
Code name AD102 AD102
Launch date 13 Oct 2022 20 Sep 2022
Place in performance rating 2 13

Technical info

Boost clock speed 2490 MHz 2520 MHz
Core clock speed 735 MHz 2235 MHz
Manufacturing process technology 5 nm 4 nm
Pipelines 18176 16384
Pixel fill rate 478.1 GPixel/s 443.5 GPixel/s
Texture fill rate 1,414 GTexel/s 1,290 GTexel/s
Thermal Design Power (TDP) 300 Watt 450 Watt
Transistor count 76300 million 76300 million
Peak Double Precision (FP64) Performance 1,290 GFLOPS (1:64)
Peak Half Precision (FP16) Performance 82.58 TFLOPS (1:1)
Peak Single Precision (FP32) Performance 82.58 TFLOPS

Video outputs and ports

Display Connectors 4x DisplayPort 1.4a 1x HDMI 2.1, 3x DisplayPort 1.4a

Compatibility, dimensions and requirements

Form factor Dual-slot Triple-slot
Interface PCIe 4.0 x16 PCIe 4.0 x16
Length 267 mm, 10.5 inches 304 mm, 12 inches
Recommended system power (PSU) 700 Watt 850 Watt
Supplementary power connectors 1x 16-pin 1x 16-pin
Width 111 mm, 4.4 inches 137 mm, 5.4 inches
Height 61 mm, 2.4 inches

API support

DirectX 12 Ultimate (12_2) 12 Ultimate (12_2)
OpenCL 3.0 3.0
OpenGL 4.6 4.6
Shader Model 6.7 6.7
Vulkan

Memory

Maximum RAM amount 48 GB 24 GB
Memory bandwidth 864.0 GB/s 1,008 GB/s
Memory bus width 384 bit 384 bit
Memory clock speed 2250 MHz, 18 Gbps effective 1313 MHz, 21 Gbps effective
Memory type GDDR6 GDDR6X