Report an Error

NVIDIA GeForce 920M

Graphics Processor
GK208B
Cores
384
TMUs
32
ROPs
8
Memory Size
2 GB
Memory Type
DDR3
Bus Width
64 bit
The GeForce 920M is a mobile graphics chip by NVIDIA, launched in March 2015. Built on the 28 nm process, and based on the GK208B graphics processor, in its N16V-GM variant, the chip supports DirectX 12.0. Even though it supports DirectX 12, the feature level is only 11_0, which can be problematic with newer DirectX 12 titles. The GK208B graphics processor is a relatively small chip with a die area of only 87 mm² and 915 million transistors. It features 384 shading units, 32 texture mapping units and 8 ROPs. NVIDIA has paired 2,048 MB DDR3 memory with the GeForce 920M, which are connected using a 64-bit memory interface. The GPU is operating at a frequency of 954 MHz, memory is running at 900 MHz.
Its power draw is rated at 33 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Rather it is intended for use in laptop/notebooks and will use the output of the host mobile device. GeForce 920M is connected to the rest of the system using a PCI-Express 3.0 x8 interface.

Graphics Processor

GPU Name
GK208B
GPU Variant
N16V-GM
Architecture
Kepler 2.0
Foundry
TSMC
Process Size
28 nm
Transistors
915 million
Die Size
87 mm²

Mobile Graphics

Release Date
Mar 13th, 2015
Generation
GeForce 900M
Production
Active
Bus Interface
PCIe 3.0 x8

Clock Speeds

GPU Clock
954 MHz
Memory Clock
900 MHz
1800 MHz effective

Memory

Memory Size
2 GB
Memory Type
DDR3
Memory Bus
64 bit
Bandwidth
14.40 GB/s

Render Config

Shading Units
384
TMUs
32
ROPs
8
SMX Count
2
L1 Cache
16 KB (per SMX)
L2 Cache
128 KB

Theoretical Performance

Pixel Rate
7.632 GPixel/s
Texture Rate
30.53 GTexel/s
FP32 (float) performance
732.7 GFLOPS
FP64 (double) performance
244.2 GFLOPS (1:3)

Board Design

Slot Width
IGP
TDP
33 W
Outputs
No outputs

Graphics Features

DirectX
12.0 (11_0)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.126
CUDA
3.5
Shader Model
5.1

GK208B GPU Notes

L1 Cache is configurable from 16 KB up to 48 KB per SMX