Video interface unit for mapping physical image data to logical tiles

Claim 2

In a video processing system having a video memory for storing video data having P rows and Q columns, a method for reading and writing image data to the video memory comprising the steps of: logically subdividing the video data of P rows and Q columns into a plurality of tiles having dimensions of p rows and q columns of image data; and mapping a given one of said tiles to a storage memory by placing the image data located on an even numbered row of said p rows and an even numbered column of said q columns on a first data bus and placing the image data located at an odd numbered row of said p rows and at an odd numbered columns of said q columns onto a second data bus.

Claim 3

The method of claim 1 further comprising the steps of: storing the image data placed on said first data bus in a first buffer; and storing the image data placed on said second data bus in a second buffer.

Claim 4

The method of claim 2 further comprising the steps of: interpolating between a first set of image data stored in said first and said second buffers and a second set of image data stored in said first and said second buffers; realigning, prior to said step of interpolating, said first and said second sets of sata to reconstitute the image data as a first array and a second array of image data respectively; and wherein said first and said second arrays have alternating odd and even rows of image data.

Claim 5

The method of claim 1 where said first and said second data bus together form a video data bus and further comprising the step of configuring the video data bus to be one of 32 or 16 bits wide.

Claim 6

The method of claim 4 wherein said video data bus is configured as a 32 bit data bus and wherein: the step of logically subdividing the video data into the plurality of tiles comprises the step of forming tiles having dimensions p=2 and q=2.

Claim 7

The method of claim 4 wherein said video data bus is configured as a 16 bit data bus and wherein: the step of logically subdividing the video data into the plurality of tiles comprises the step of forming tiles having dimensions p=2 and q=1.

Claim 8

The method of claim 1 wherein the step of mapping further comprises the step of addressing the given tile by a single row and a single column coordinate.

Claim 9

The method of claim 1 wherein the step of mapping further comprises writing individual ones of the image data in any one of a plurality of memory devices according to a value stored in a register.

Claim 10

The method of claim 1 wherein the step of mapping onto a first and a second data bus further comprises the step of operating said first and said second data bus on a first video clock asynchronous with a system clock.

Claim 11

The method of claim 1 further comprising the step of interpolating between a first set and a second set of image data.

Claim 12

The method of claim 1 further comprising the step of refreshing the video memory.

Claim 13

In a video processing system having a video memory for storing image data having P rows and Q columns, a device for reading and writing data to the video memory comprising: means, coupled to a video bus, for mapping a physical image date having P rows and Q columns logically subdivided into a plurality of tiles having dimensions of p rows and q columns, to a storage memory, by placing the image data located on an even numbered row of said p rows and an even numbered column of said q columns in a first buffer memory; and means, coupled to said video bus, for mapping said given one of said tiles to said storage memory by placing the image data located at an odd numbered row of said p rows and at an odd numbered column of said q columns into a second buffer memory.

Claim 14

In a video processing system having a video memory for storing image data having P rows and Q columns, a device for reading and writing data to the video memory comprising: means, coupled to a video bus, for mapping a physical image date having P rows and Q columns logically subdivided into a plurality of tiles having dimensions of p rows and q columns, to a storage memory, by placing the image data located on an even numbered row of said p rows and an even numbered column of said q columns in a first buffer memory; and means, coupled to said video bus, for mapping said given one of said tiles to said storage memory by placing the image data located at an odd numbered row of said p rows and at an odd numbered column of said q columns into a second buffer memory.

Claim 15

The device of claim 12 wherein said video bus comprises a 32 bit video bus and p=2 and q=2.

Claim 16

The device of claim 12 wherein said video bus comprises a 16 bit video bus and p=2 and q=l.

Claim 17

The device of claim 12 further comprising a means, coupled to the video memory, for refreshing the video memory.

Claim 18

The device of claim 12 further comprising a video bus arbiter for arbitrating access to the video bus.

Claim 19

The device of claim 12 further comprising a means for interpolating between a first set of image data stored in said first and second buffers.

Claim 20

The device of claim 12 further comprising a circular buffer coupled to said first buffer and said second buffer for storing a first set of data from said first buffer together with a second set of data from said second buffer as an array of image data having alternating odd and even rows.

Claim 21

The device of claim 12, wherein the video memory includes a serial access video memory and a DRAM, the device further comprising: a means, coupled to said video bus, for controlling transfers between the serial access video memory and the DRAM.

Claim 22

The device of claim 12, wherein the video memory includes a serial access video memory and a DRAM, the device further comprising: means, coupled to the video bus, for controlling transfers between the serial access video memory and the DRAM.

Claim 23

The device of claim 12 wherein said device operates on a video clock asynchronous from a system clock.

Claim 24

The device according to claim 12 further comprising: a latch having an input coupled to an output of the video memory and an output coupled to said storage memory; and a latching signal pin on the device coupled to a CAS signal, wherein the CAS signal is output from a CAS pin of the device to the video memory and wherein the latching signal pin forms a latching signal input to said latch.

Claim 25

In a video processing system having a video memory for storing image data having P rows and Q columns, a device for reading and writing data to the video memory comprising: means, coupled to a video bus, for mapping a physical image data having P rows and Q columns logically subdivided into a plurality of tiles having dimensions of p rows and q columns, to a storage memory, a given one of said tiles, by placing the image data located on an even numbered row of said p rows and an even numbered column of said q columns in a first buffer memory and for mapping said given one of said tiles to said storage memory by placing the image data located at an odd numbered row of said p rows and at an odd numbered columns of said q columns into a second buffer memory; a circular buffer, coupled to said first and to said second buffer memory for storing a first set of data from said first buffer together with a second set of data from said second buffer as an array of image data having alternating odd and even rows; an interpolator, coupled to said circular buffer for interpolating the array of image data; an arbiter, coupled to the video bus, for controlling access to the video bus; and a state machine, coupled to the means for mapping, to said interpolator and to said arbiter for sequencing operation of said means for mapping, said interpolator, and said arbiter.

Claim 26

In a video processing system having a video memory for storing image data having P rows and Q columns, a device for reading and writing data to the video memory comprising: means, coupled to a video bus, for mapping a physical image data having P rows and Q columns logically subdivided into a plurality of tiles having dimensions of p rows and q columns, to a storage memory, a given one of said tiles, by placing the image data located on an even numbered row of said p rows and an even numbered column of said q columns in a first buffer memory and for mapping said given one of said tiles to said storage memory by placing the image data located at an odd numbered row of said p rows and at an odd numbered columns of said q columns into a second buffer memory; a circular buffer, coupled to said first and to said second buffer memory for storing a first set of data from said first buffer together with a second set of data from said second buffer as an array of image data having alternating odd and even rows; an interpolator, coupled to said circular buffer for interpolating the array of image data; an arbiter, coupled to the video bus, for controlling access to the video bus; and a state machine, coupled to the means for mapping, to said interpolator and to said arbiter for sequencing operation of said means for mapping, said interpolator, and said arbiter.

Claim 27

The device of claim 23 further comprising a refresh controller, coupled to the arbiter for refreshing the video memory.

Claim 28

A latch device for a DRAM memory interface comprising: a latching signal pin on the DRAM memory interface coupled to a CAS signal output from the DRAM memory interface CAS pin to the DRAM memory; a latch having: an input coupled to an output of the DRAM memory; an output coupled to said interface; a latching signal input coupled to said latching signal pin; and wherein when said latching signal pin is disabled, said latch is transparent.

Claim 29

A latch device for a DRAM memory interface comprising: a latching signal pin on the DRAM memory interface coupled to a CAS signal output from the DRAM memory interface CAS pin to the DRAM memory; a latch having: an input coupled to an output of the DRAM memory; an output coupled to said interface; a latching signal input coupled to said latching signal pin; and wherein when said latching signal pin is disabled, said latch is transparent.