公开(公告)号：US20210174569A1

公开(公告)日：2021-06-10

申请号：US17162550

申请日：2021-01-29

Applicant: NVIDIA Corporation

Inventor： Rouslan L. Dimitrov ,  Dale L. Kirkland ,  Emmett M. Kilgariff ,  Sachin Satish Idgunji ,  Siddharth Sharma

IPC: G06T15/00 ,  G06N3/08 ,  G06T17/10 ,  G06T15/80

Abstract: Graphics processing unit (GPU) performance and power efficiency is improved using machine learning to tune operating parameters based on performance monitor values and application information. Performance monitor values are processed using machine learning techniques to generate model parameters, which are used by a control unit within the GPU to provide real-time updates to the operating parameters. In one embodiment, a neural network processes the performance monitor values to generate operating parameters in real-time.

公开(公告)号：US10909738B2

公开(公告)日：2021-02-02

申请号：US15863780

申请日：2018-01-05

Applicant: NVIDIA Corporation

Inventor： Rouslan L. Dimitrov ,  Dale L. Kirkland ,  Emmett M. Kilgariff ,  Sachin Satish Idgunji ,  Siddharth Sharma

IPC: G06T15/00 ,  G06N3/08 ,  G06T17/10 ,  G06T15/80

Abstract: Graphics processing unit (GPU) performance and power efficiency is improved using machine learning to tune operating parameters based on performance monitor values and application information. Performance monitor values are processed using machine learning techniques to generate model parameters, which are used by a control unit within the GPU to provide real-time updates to the operating parameters. In one embodiment, a neural network processes the performance monitor values to generate operating parameters in real-time.

公开(公告)号：US20190213775A1

公开(公告)日：2019-07-11

申请号：US15863780

申请日：2018-01-05

Applicant: NVIDIA Corporation

Inventor： Rouslan L. Dimitrov ,  Dale L. Kirkland ,  Emmett M. Kilgariff ,  Sachin Satish Idgunji ,  Siddharth Sharma

IPC: G06T15/00 ,  G06T15/80 ,  G06T17/10 ,  G06N3/08

Abstract: Graphics processing unit (GPU) performance and power efficiency is improved using machine learning to tune operating parameters based on performance monitor values and application information. Performance monitor values are processed using machine learning techniques to generate model parameters, which are used by a control unit within the GPU to provide real-time updates to the operating parameters. In one embodiment, a neural network processes the performance monitor values to generate operating parameters in real-time.

公开(公告)号：US20190206018A1

公开(公告)日：2019-07-04

申请号：US15879377

申请日：2018-01-24

Applicant: NVIDIA Corporation

Inventor： Andrei Khodakovsky ,  Kirill A. Dmitriev ,  Rouslan L. Dimitrov ,  Tzyywei Hwang ,  Wishwesh Anil Gandhi ,  Lacky Vasant Shah

IPC: G06T1/20 ,  G06T1/60 ,  G06T15/00

CPC classification number: G06T1/20 ,  G06T1/60 ,  G06T15/005 ,  G06T2200/24 ,  G06T2210/52

Abstract: One or more copy commands are scheduled for locating one or more pages of data in a local memory of a graphics processing unit (GPU) for more efficient access to the pages of data during rendering. A first processing unit that is coupled to a first GPU receives a notification that an access request count has reached a specified threshold. The first processing unit schedules a copy command to copy the first page of data to a first memory circuit of the first GPU from a second memory circuit of the second GPU. The copy command is included within a GPU command stream.

公开(公告)号：US20140092113A1

公开(公告)日：2014-04-03

申请号：US14024550

申请日：2013-09-11

Applicant: NVIDIA Corporation

Inventor： Tom Petersen ,  David Wyatt ,  Paul van der Kouwe ,  Emmett M. Kilgariff ,  Laurence Harrison ,  Jensen Huang ,  Tony Tamasi ,  Gerrit A. Slavenburg ,  Thomas F. Fox ,  David Matthew Stears ,  Robert Jan Schutten ,  Ross Cunniff ,  Ajay Kamalvanshi ,  Robert Osborne ,  Rouslan L. Dimitrov

IPC: G09G5/00

CPC classification number: G09G5/001 ,  G09G5/12 ,  G09G5/395 ,  G09G2320/0247 ,  G09G2340/0435

Abstract: A system, method, and computer program product are provided for a dynamic display refresh. In use, a state of a display device is identified in which an entirety of an image frame is currently displayed by the display device. In response to the identification of the state, it is determined whether an entirety of a next image frame to be displayed has been rendered to memory. The next image frame is transmitted to the display device for display thereof, when it is determined that the entirety of the next image frame to be displayed has been rendered to the memory. Further, a refresh of the display device is delayed, when it is determined that the entirety of the next image frame to be displayed has not been rendered to the memory.

Abstract translation: 提供了系统，方法和计算机程序产品，用于动态显示刷新。 在使用中，识别显示装置当前正在显示整个图像帧的显示装置的状态。 响应于状态的识别，确定要显示的下一个图像帧的整体是否已经被呈现给存储器。 当确定要显示的下一个图像帧的整体已经被呈现给存储器时，下一个图像帧被发送到显示装置以进行显示。 此外，当确定要显示的下一图像帧的整体未被呈现给存储器时，显示装置的刷新被延迟。