GPU Comparison

NVIDIA

GEFORCE

NVIDIA L20

CORE STATE AD102

VRAM 48 GB

CLOCK SPEED 2520 MHz

TDP 275 W

BUS WIDTH 384 bit

ARCHITECTURE Ada Lovelace

nm

PROCESS 5 nm

LAUNCH DATE 2023

VS

NVIDIA

GEFORCE

Quadro M4000

CORE STATE GM204

VRAM 8 GB

CLOCK SPEED —

TDP 120 W

BUS WIDTH 256 bit

ARCHITECTURE Maxwell 2.0

nm

PROCESS 28 nm

LAUNCH DATE 2015

PERFORMANCE BENCHMARKS

geekbench_opencl

266,428

20,128

3dmark_3dmark_steel_nomad_dx12

N/A

680

geekbench_vulkan

N/A

24,640

passmark_directx_10

N/A

33

passmark_directx_11

N/A

49

passmark_directx_12

N/A

26

passmark_directx_9

N/A

113

passmark_g2d

N/A

673

passmark_g3d

N/A

6,680

passmark_gpu_compute

N/A

2,660

DETAILED SPECIFICATIONS

SPECIFICATION

L20

Quadro M4000

Core Specs

Shading Units

11,776

1,664 -85.9%

Shaders

11,776

1,664 -85.9%

TMUs

368

104 -71.7%

ROPs

128

64 -50.0%

SM Count

92

—

Clocks

Base Clock

1440 MHz

—

Boost Clock

2520 MHz

—

GPU Clock

—

773 MHz

Memory Clock

2250 MHz 18 Gbps effective

1502 MHz 6 Gbps effective

Memory

Memory Size

48 GB

8 GB

VRAM (MB)

49,152

8,192 -83.3%

Memory Type

GDDR6

GDDR5

Memory Bus

384 bit

256 bit

Bandwidth

864.0 GB/s

192.3 GB/s

Cache

L1 Cache

128 KB (per SM)

48 KB (per SMM)

L2 Cache

96 MB

2 MB

Performance

Pixel Rate

322.6 GPixel/s

49.47 GPixel/s

Texture Rate

927.4 GTexel/s

80.39 GTexel/s

FP32 (TFLOPS)

59.35 TFLOPS

2.573 TFLOPS

FP64 (TFLOPS)

927.4 GFLOPS (1:64)

80.39 GFLOPS (1:32)

FP16 (TFLOPS)

59.35 TFLOPS (1:1)

—

AI/RT

RT Cores

92

—

Tensor Cores

368

—

Power

TDP

275 W

120 W

TDP (W)

275

120 -56.4%

Suggested PSU

600 W

300 W

Power Connectors

1x 16-pin

1x 6-pin

Architecture

Architecture

Ada Lovelace

Maxwell 2.0

GPU Name

AD102

GM204

Generation

Server Ada (Lxx)

Quadro Maxwell (Mx000)

Process Size

5 nm

28 nm

Transistors

76,300 million

5,200 million

Die Size

609 mm²

398 mm²

Foundry

TSMC

TSMC

Density

125.3M / mm²

13.1M / mm²

API Support

DirectX

12 Ultimate (12_2)

12 (12_1)

OpenGL

4.6

4.6

Vulkan

1.4

1.4

OpenCL

3.0

3.0

CUDA

8.9

5.2

Shader Model

6.8

6.8

Physical

Slot Width

Dual-slot

Single-slot

Length

267 mm 10.5 inches

241 mm 9.5 inches

Height

111 mm 4.4 inches

111 mm 4.4 inches

Outputs

4x DisplayPort 1.4a

4x DisplayPort 1.2

Bus Interface

PCIe 4.0 x16

PCIe 3.0 x16

Other

Production

Active

End-of-life

Predecessor

Server Ampere

Quadro Kepler

Successor

Server Hopper

Quadro Pascal

View L20 Details View Quadro M4000 Details