GeForce RTX 2050 Max-Q

General
ManufacturerNVIDIA
SeriesGeForce 20 Mobile
GPU NameGA107
ArchitectureAmpere
GenerationGeForce 20 Mobile
Release DateDec 2021
Bus InterfacePCIe 3.0 x8
Process & Die
FoundrySamsung
Process Size8 nm
Transistors8,700 M
Transistor Density43,500 K/mm²
Die Size200 mm²
Chip PackageFCBGA-1358
Clock Speeds
Base Clock832 MHz
Boost Clock1,155 MHz
Memory
VRAM4 GB
Memory TypeGDDR6
Memory Clock1,475 MHz
Memory Bus64-bit
Bandwidth94.4 GB/s
Processing Units
Shading Units2,048
TMUs64
ROPs32
SMs16
Tensor Cores64
RT Cores32
Cache
L1 Cache64 KB
L2 Cache2 MB
Power & Thermals
TDP30 W
Power ConnectorsNone
Dimensions
Board Length229 mm
Board Width113 mm
Outputs
Display Connectors1x DVI, 1x HDMI 2.1, 2x DisplayPort 1.4a
API Support
DirectX12.2
OpenGL4.6
Vulkan1.3
OpenCL3.0
CUDA8.6
Shader Model6.8
Theoretical Performance
Pixel Rate36.96 GP/s
Texture Rate73.92 GT/s
FP16 (Half)9,462 GFLOPS
FP32 (Single)4,731 GFLOPS
FP64 (Double)147.8 GFLOPS
External
TechPowerUpView →