Michael S Frank

US Patent:

6335761, Jan 1, 2002

Filed:

Dec 15, 1998

Appl. No.:

09/212132

Inventors:

David I. J. Glen - Toronto, CA
Michael Frank - Newtown PA
David Strasser - Toronto, CA

Assignee:

ATI International S.R.L. - Christ Church

International Classification:

H04N 964

US Classification:

348557, 348453, 348599

Abstract:

A method and apparatus for converting the color base of an image input layer include processing that begins by interpreting a conversion flag. The processing then continues by converting the color base of the image input layer from a first color base to a second color base when the conversion flag indicates a color base conversion. Note that a color base corresponds to standardized colorometries of video signals, color space of video signals, and/or any other displaying characteristics of video signals that are standardized or may be subsequently standardized. Further note that an image input layer corresponds to a display or portion thereof (e. g. , a window or a picture in picture), wherein the display is capable of presenting images from multiple video and/or graphics data sources (e. g. , a television signal and a computer application).

US Patent:

6348886, Feb 19, 2002

Filed:

Jul 19, 2000

Appl. No.:

09/619209

Inventors:

Michael Frank - Newtown PA
Zaw Min Soe - Los Gatos CA

Assignee:

ATI Technologies - Thornhill

International Classification:

H03M 166

US Classification:

341144, 341120, 341118

Abstract:

A video DAC uses a charge pump to maintain a constant peak to peak amplitude within power supply variations over process and temperature, rather than an external reference signal or a band gap reference. The charge pump performs calibration at full scale output during horizontal sync levels or vertical sync pulses.

US Patent:

6462786, Oct 8, 2002

Filed:

Dec 15, 1998

Appl. No.:

09/212141

Inventors:

David I. J. Glen - Toronto, CA
Michael Frank - Newtown PA
Ed Callway - Toronto, CA

Assignee:

ATI International SRL - Christ Church

International Classification:

H04N 976

US Classification:

348599, 348453, 348555, 348557, 348584

Abstract:

A method and apparatus for blending a plurality of image input layers include processing that begins by converting each of a plurality of image input layers that have a color base that differs from a color base of a display into a image layer having the color base of the display, thereby producing converting image layers. The color base of the image input layers and of the output include the colorimetries of various standardized video signals, color space, and/or any other defining display characteristics of video signals that is currently standardized or maybe standardized. Further note that an image input layer corresponds to a window, a background and/or a display area of a display, where the display is capable of presenting more than one image, where the images originate from different video and/or graphics data sources (e. g. , a television signal, and a computer application display). The processing continues by blending each of the converted image layers with each of the plurality of image input layers that have the color base of the display to produce a blended output having the color base of the display.

US Patent:

6499072, Dec 24, 2002

Filed:

Sep 2, 1999

Appl. No.:

09/388975

Inventors:

Michael Frank - Newtown PA
John Raymond Wiseman - Newtown PA

Assignee:

ATI International Srl - Barbados

International Classification:

G06F 1300

US Classification:

710100, 710 29, 710 60, 709102

Abstract:

A data bus bandwidth allocation apparatus and method uses buffer entry feedback data from a buffer, such as an overflow buffer, that receives requested data over an unregulated bus. The data bus bandwidth allocation method and apparatus generates data issue delay data based on the buffer entry feedback data to adjust data commands to a data source, such as frame buffer memory used to feed a regulated bus. In one embodiment, the data issue rate regulator utilizes a programmable threshold corresponding to a threshold of data entries in an overflow. The overflow FIFO has feedback indicating, for example, the number of free entries or the number of full entries. The data issue rate regulator provides data rate regulation information to an adjustable delay sequencer. The adjustable delay sequencer selectively sequences data reads from the frame buffer memory so that data collisions do not occur over the memory read backbone.

US Patent:

7103226, Sep 5, 2006

Filed:

Mar 23, 1998

Appl. No.:

09/046179

Inventors:

Michael Frank - Newtown PA, US
Gary Root - Harleysville PA, US
Richard J. Selvaggi - Doylestown PA, US

Assignee:

ATI Technologies, Inc. - Markham

International Classification:

G06K 9/40

US Classification:

382254, 382162, 382166, 345153, 345154, 3484071

Abstract:

A video processor integrates the processing of video and graphics data by restructuring the discrete blocks of luminance and chrominance component data associated with video image data (Y,U,V) into composite pixel values (YUYV), and performing the subsequent processing, in particular motion compensation, upon this composite data form. The discrete blocks of video image data are queue and processed in parallel to form a composite pixel value associated with each pixel within the image areas represented by the discrete blocks. By forming composite pixel values for reference data and error term data, common processing elements can be optimized and used for both video and graphics image processing. For example, the trilinear interpolator commonly used for 3-D graphics filtering and texturing can be used to form predicted macroblocks for motion compensation, and the specular adder used for graphics lighting effects can be used to process the motion compensation error terms.

US Patent:

62854046, Sep 4, 2001

Filed:

Aug 3, 1998

Appl. No.:

9/126971

Inventors:

Michael Frank - Newtown PA

Assignee:

ATI Technologies Inc. - Thornhill

International Classification:

H04N 701

US Classification:

348469

Abstract:

A systolic video encoding system processes image data from a frame buffer at a core clock rate that is independent of the sample rate of the image data. The video encoder of this invention uses the core clock rate of the host image processing system to process image from the frame buffer at this core clock rate. The image data is pumped out of the frame buffer, processed by each of the processes of the video encoder when the data reaches each of the processes, and the encoded samples are stored in a raster sample buffer for subsequent processing. The image data is continually pumped out of the frame buffer at the core clock rate until the raster sample buffer is full. As the samples are extracted from the raster sample buffer, subsequent image data is pumped into the video encoding system, producing a systolic processing effect. By allowing operation at a core clock rate that is independent of, and substantially higher than, the image sample rate, elements of the system can be used in a time-shared manner, thereby reducing the cost and size of the video encoding system.

US Patent:

60974001, Aug 1, 2000

Filed:

Jun 1, 1998

Appl. No.:

9/087712

Inventors:

Steven C. Dilliplane - Yardley PA
Michael Frank - Newtown PA

Assignee:

ATI Technologies, Inc. - Thornhill

International Classification:

G06F 1500

US Classification:

345443

Abstract:

A method and apparatus for anti-aliasing a rendered scene begins processing by rendering a plurality of objects of an image to a buffer to produce a rendered image. For example, the image corresponds to a particular frame of data to be displayed on a computer monitor, television, etc. Each of the objects corresponds to portions of the image and is comprised of object-elements (e. g. , triangles). While the plurality of objects is being rendered, edge information of the objects that are in a most foreground position is maintained. The edge information includes a pixel offset value, which indicates the offset of an edge from a pixel center, and gradient information of the object edge. After the image has been rendered and the edge information has been collected for a particular frame of data, the rendered image is filtered based on the edge information. In particular, a two-dimensional spatial filter that derives its filtering parameters from the edge information is applied to the rendered image.

US Patent:

62560700, Jul 3, 2001

Filed:

Oct 13, 1998

Appl. No.:

9/170519

Inventors:

Michael Frank - Newtown PA

Assignee:

Ati International Srl - Christ Church

International Classification:

H04N 708
H04N 974

US Classification:

348473

Abstract:

A video encoding system combines the processing of image data and closed caption data in the formation of a raster television signal. By combining the processing of image data and closed caption data, a single discrete time oscillator (DTO) can be used to generate oscillation signals that are required for each process. In a preferred embodiment, the DTO generates the 500 kHz signal that is used as the "run in" clock for closed caption data, and also generates the subcarrier signal that is used for the modulation of the chrominance components of a composite video signal.