Patents Assigned to NVIDIA Corp.
  • Patent number: 10566958
    Abstract: Injection locked oscillation circuits are applied along clock distribution circuit paths to increase clock signal bandwidth, reduce duty cycle error, reduce quadrature phase error, reduce clock signal jitter, and reduce clock signal power consumption.
    Type: Grant
    Filed: January 15, 2019
    Date of Patent: February 18, 2020
    Assignee: NVIDIA Corp.
    Inventors: Sanquan Song, Olakanmi Oluwole, John Poulton, Carl Thomas Gray
  • Patent number: 10558230
    Abstract: High-resolution switched digital regulators are disclosed having fast cross corner and variable temperature response, with constrained ripple. The strength of the power transistors utilized by the regulator are adjusted to control the current delivered to the load. The regulators utilize a slow control loop in parallel with a primary fast switching loop. The slow loop uses the switching signal of the primary loop to estimate the load current and set the power transistor size accordingly.
    Type: Grant
    Filed: September 19, 2018
    Date of Patent: February 11, 2020
    Assignee: NVIDIA Corp.
    Inventors: Sudhir Kudva, John Wilson
  • Publication number: 20190386677
    Abstract: A PAM-4 communication process divides a full burst of raw data into two half bursts, extracts a bit from each half burst and communicating the extracted bit on a DBI line, and encodes the remaining bits of the half burst to avoid maximum transitions between PAM-4 symbols on a data line.
    Type: Application
    Filed: August 30, 2019
    Publication date: December 19, 2019
    Applicant: NVIDIA Corp.
    Inventors: Sunil Sudhakaran, Russ Newcomb, Rohit Rathi
  • Patent number: 10491435
    Abstract: Methods of operating a serial data bus divide series of data bits into sequences of one or more bits and encode the sequences as N-level symbols, which are then transmitted at multiple discrete voltage levels. These methods may be utilized to communicate over serial data lines to improve bandwidth and reduce crosstalk and other sources of noise.
    Type: Grant
    Filed: March 28, 2019
    Date of Patent: November 26, 2019
    Assignee: NVIDIA Corp.
    Inventors: Donghyuk Lee, James Michael O'Connor, John Wilson
  • Patent number: 10491238
    Abstract: A PAM-4 communication process divides a full burst of raw data into two half bursts, extracts a bit from each half burst and communicating the extracted bit on a DBI line, and encodes the remaining bits of the half burst to avoid maximum transitions between PAM-4 symbols on a data line.
    Type: Grant
    Filed: September 6, 2018
    Date of Patent: November 26, 2019
    Assignee: NVIDIA Corp.
    Inventors: Sunil Sudhakaran, Russ Newcomb, Rohit Rathi
  • Patent number: 10489542
    Abstract: A neural network including an embedding layer to receive a gate function vector and an embedding width and alter a shape of the gate function vector by the embedding width, a concatenator to receive a gate feature input vector and concatenate the gate feature input vector with the gate function vector altered by the embedding width, a convolution layer to receive a window size, stride, and output feature size and generate an output convolution vector with a shape based on a length of the gate function vector, the window size of the convolution layer, and the output feature size of the convolution layer, and a fully connected layer to reduce the gate output convolution vector to a final path delay output.
    Type: Grant
    Filed: April 24, 2018
    Date of Patent: November 26, 2019
    Assignee: NVIDIA Corp.
    Inventors: Mark Ren, Brucek Khailany
  • Patent number: 10466968
    Abstract: A system including a series of partial product select encoders and partial product muxes, each of the partial product select encoders receiving a multiplier, receiving a carry input from a multiplier tree, and outputting a select signal to an associated partial product mux based on the multiplier and carry input, and each of the partial product muxes outputting a partial product based on the select signal and a multiplicand received.
    Type: Grant
    Filed: July 12, 2018
    Date of Patent: November 5, 2019
    Assignee: NVIDIA Corp.
    Inventor: Ilyas Elkin
  • Patent number: 10436840
    Abstract: A distributed test circuit includes partitions arranged in series to form a scan path, each partition including a scan multiplexer, a test data register, and a segment insertion bit component. The scan multiplexer of each partition provides inputs to the corresponding test data register of the each partition. Broadcast control logic generates a select signal to the scan multiplexer of each partition to place the test circuit in a broadcast mode when the select signal is asserted, and to switch the test circuit to a daisy mode when select signal is de-asserted. The segment insertion bit is operable to include or bypass each partition from the scan path.
    Type: Grant
    Filed: March 26, 2018
    Date of Patent: October 8, 2019
    Assignee: NVIDIA Corp.
    Inventors: Jau Wu, Saurabh Gupta
  • Patent number: 10404505
    Abstract: A system comprising a PAM-4 transmitter coupled data lanes includes a least significant bit section and a most significant bit section for the symbols generated on each lane. A controller to determine a state of the PAM-4 transmitter and selectively inverts a polarity of the symbol bits on the lanes based on the state.
    Type: Grant
    Filed: October 26, 2018
    Date of Patent: September 3, 2019
    Assignee: NVIDIA Corp.
    Inventor: John Wilson
  • Patent number: 7593025
    Abstract: A transform engine is configured to rotate, and/or rotate and translate, one or more polygons in response to screen orientation. Thus, when texture, obtained from a pre-rotated image is applied to a rotated polygon used to render such an image, a rotated version of such an image is generated in response to screen orientation. Alternatively, a user may select a rotation to re-orient an image to a screen view position.
    Type: Grant
    Filed: December 7, 2007
    Date of Patent: September 22, 2009
    Assignee: NVIDIA Corp.
    Inventor: Abraham B. De Waal
  • Patent number: 7593021
    Abstract: An apparatus and method for converting color data from one color space to another color space. A driver determines that a set of shader program instructions perform a color conversion function and the set of shader program instructions are replaced with either a single shader program instruction or a flag is set within an existing shader program instruction to specify that output color data is represented in a nonlinear color format. The output color data is converted to the nonlinear color format prior to being stored in a frame buffer. Nonlinear color data read from the frame buffer is converted to a linear color format prior to shading, blending, or raster operations.
    Type: Grant
    Filed: September 13, 2004
    Date of Patent: September 22, 2009
    Assignee: NVIDIA Corp.
    Inventors: John D. Tynefield, Jr., Andrew J. Tao, Rui M. Bastos, Johnny S. Rhoades
  • Patent number: 7594229
    Abstract: A system including a method employed by a dedicated processor for allocating resources to other processors with a multi-processor computing environment. The dedicated processor is dedicated only to providing resource allocation to the other processors. Specifically, a script file is provided to the dedicated processor, the script containing information related to the resources required by the other processors. The script file is parsed by the dedicated processor to determine the resources required by the second processor. Thereafter, the dedicated processor dynamically allocates the resources and synchronizes resource allocation at the time needed by the other processors.
    Type: Grant
    Filed: October 9, 2001
    Date of Patent: September 22, 2009
    Assignee: NVIDIA Corp.
    Inventor: Ian Hirschsohn
  • Patent number: 7593018
    Abstract: A system and method for providing explicit weights for texture filtering permits filter weights to vary for each pixel within a primitive. A different filter kernel may be used for each pixel. The weights may be computed or read from a texture map. Because the weights are explicit, the fractional portions of the texture map coordinates that are typically used to determine a bilinearly filtered texel are not used.
    Type: Grant
    Filed: December 14, 2005
    Date of Patent: September 22, 2009
    Assignee: NVIDIA Corp.
    Inventors: Michael J. M. Toksvig, Walter E. Donovan
  • Patent number: 7548481
    Abstract: An aspect of the invention relates to a method of dynamically adjusting power consumption of a random access memory (RAM) coupled to a processor. Frequency of a memory clock signal coupled to the RAM is reduced. At least one supply voltage coupled to the RAM is reduced. At least one latency parameter of the RAM is configured in response to the reduced frequency and the reduced at least one supply voltage. The RAM may then be re-initialized. In this manner, voltage supplied to the RAM is reduced, thereby reducing power consumption in the RAM.
    Type: Grant
    Filed: December 8, 2006
    Date of Patent: June 16, 2009
    Assignee: NVIDIA Corp.
    Inventors: Thomas E. Dewey, Barry A. Wagner, Weijen Chao, Andrew R. Bell, David A. Bachman
  • Publication number: 20080012735
    Abstract: The present invention provides a system and method for introducing white noises into a digital audio signal so that there is progressive and cumulative degradation in audio quality after each successive reproduction of the audio sound signal in a fashion analogous to analog audio reproduction. The invention provides a white noise generator, and a digital entroping unit. In a preferred embodiment, the white noise generator is implemented by a hardware random number generator. The digital entroping unit controls the magnitude of white noise desired based on a random number generated by the random number generator, and adds the white noise to the input audio sound signal to produce a degraded audio sound signal. The magnitude of white noise can be controlled by using various masking and formatting of random number data.
    Type: Application
    Filed: February 12, 2007
    Publication date: January 17, 2008
    Applicant: NVIDIA Corp.
    Inventor: JASON KIM
  • Publication number: 20040039954
    Abstract: A method for adapting power consumption of a processor based upon an application demand is provided. The method initiates with determining an application demand based upon a current processing operation. Then, a time interval associated with the application demand is determined. Next, unnecessary power consuming functions for the application demand are determined. Then, a clock frequency for the unnecessary power consuming functions is reduced for the time interval. In one embodiment, the power is terminated to the unnecessary power consuming functions. In another embodiment, the clock frequency of the processor is adjusted for at least a portion of the time interval. A program interface for adapting power consumption of a computer system, processor instructions for adapting power consumption of a computer system and a processor are included.
    Type: Application
    Filed: August 22, 2002
    Publication date: February 26, 2004
    Applicant: NVIDIA, CORP.
    Inventors: Jonathan B. White, James L. van Welzen
  • Publication number: 20030233568
    Abstract: Method and apparatus for enhanced security for communication over a network, and more particularly to control of security protocol negotiation to enable multiple clients to establish a virtual private network connection with a same remote address, is described. A mapping table accessible by a gateway computer is used to form associations between a local address for the client and a destination address for a peer and a Security Parameters Index associated with IPSec-protected traffic from the peer. When a packet is received at the gateway from a client it is checked to determine if it is an Internet Key Exchange (IKE) packet, whether an IKE session has already been recorded from this client in the mapping table for the destination address in the IKE packet, whether a Security Parameters Index has been observed in the clear from a remote computer associated with the destination address.
    Type: Application
    Filed: June 13, 2002
    Publication date: December 18, 2003
    Applicant: Nvidia Corp.
    Inventors: Thomas Albert Maufer, Sameer Nanda, Paul J. Sidenblad
  • Publication number: 20030233475
    Abstract: Method and apparatus for enhanced security for communication over a network, and more particularly to Network Address Translation (NAT) integration Internet Protocol Security (IPSec), is described. A client computer makes a second address request in order to prompt an address server to provide a public address. This address, recorded in a mapping table accessible by a gateway computer. This public address is used as a source address for packets from a client using IPSec. When the gateway computer identifies a packet's source address as one of it's public addresses, NAT is suspended for this packet, and the packet is routed without NAT. Incoming traffic is routed using the mapping table.
    Type: Application
    Filed: June 13, 2002
    Publication date: December 18, 2003
    Applicant: Nvidia Corp.
    Inventors: Thomas Albert Maufer, Sameer Nanda, Paul J. Sidenblad
  • Publication number: 20030233452
    Abstract: Method and apparatus for Internet Protocol Security (IPSec) and Network Address Translation (NAT) integration is described. A client obtains a public address from a gateway for IPSec communication. A mapping table is used to form associations between a local address for the client and a destination address for a peer, an Internet Security Association and Key Management Protocol (ISAKMP) Initiator Cookie and a Security Parameters Index associated with communication between the client and the peer. Incoming and outgoing routing may be done at the gateway using the mapping table.
    Type: Application
    Filed: June 13, 2002
    Publication date: December 18, 2003
    Applicant: Nvidia Corp.
    Inventors: Thomas Albert Maufer, Sameer Nanda, Paul J. Sidenblad
  • Publication number: 20030233576
    Abstract: Method and apparatus for integration of network address translation and source address security, including, but not limited to, determining whether a gateway computer is integrated for network address translation and source address security, is described. A client computer requests a first address from the gateway computer and then requests a second address from the gateway computer. The latter request is done with a different client identifier that is nearly equivalent, except for one bit, to the client identifier used for the prior address request. If the gateway computer is integrated for network address translation and source address security, in response to the latter request a public address will be provided from the gateway computer to the client computer.
    Type: Application
    Filed: June 13, 2002
    Publication date: December 18, 2003
    Applicant: Nvidia Corp.
    Inventors: Thomas Albert Maufer, Sameer Nanda, Paul J. Sidenblad