Report an Error

NVIDIA GM200

Default
GM200
GM200-310-A1
GM200-310-A1
GM200-400-A1
GM200-400-A1
Die Shot
Fritzchens Fritz
Die Shot
Block Diagram
Block Diagram
SMM Diagram
SMM Diagram
NVIDIA's GM200 GPU uses the Maxwell 2.0 architecture and is made using a 28 nm production process at TSMC. With a die size of 601 mm² and a transistor count of 8,000 million it is a very big chip. GM200 supports DirectX 12.0 (Feature Level 12_1). It features 3072 shading units, 256 texture mapping units and 96 ROPs.

Graphics Processor

GPU Name
GM200
Architecture
Maxwell 2.0
Foundry
TSMC
Process Size
28 nm
Transistors
8,000 million
Density
13.3M / mm²
Die Size
601 mm²
Codename
NV120
Released
Mar 17th, 2015

Graphics Features

DirectX
12.0 (12_1)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.103
CUDA
5.2
Shader Model
6.4
PureVideo HD
VP6
VDPAU
Feature Set E

Render Config

Shading Units
3072
TMUs
256
ROPs
96
SMM Count
24
GPCs
6
SFUs
768
L1 Cache
48 KB per SMM
L2 Cache
3072 KB
Max. TDP
250 W

All Maxwell 2.0 GPUs

Graphics cards using the NVIDIA GM200 GPU

Name Chip Memory Shaders TMUs ROPs GPU Clock Boost Clock Memory Clock
NVIDIA GeForce GTX TITAN X GM200-400-A1 12 GB 3072 192 96 1000 MHz 1089 MHz 1753 MHz
NVIDIA Quadro M6000 12 GB 3072 192 96 988 MHz 1114 MHz 1653 MHz
NVIDIA GeForce GTX 980 Ti GM200-310-A1 6 GB 2816 176 96 1000 MHz 1076 MHz 1753 MHz
NVIDIA Tesla M40 GM200-895-A1 12 GB 3072 192 96 948 MHz 1114 MHz 1502 MHz
NVIDIA Quadro M6000 24 GB GM200-880-A1 24 GB 3072 256 96 988 MHz 1114 MHz 1653 MHz