Method for compressing data in a bit stream or bit pattern

Abstract

One embodiment of the present invention is directed to a method for a computer graphics system comprising compressing a plurality of groups of bits by shifting compressed groups of bits into bit positions that are to be removed during the compression, the logic being responsive to a mask, wherein contents of the mask define variable amounts that the plurality of bits are shifted during the compression. Another embodiment is directed to a method for compressing a plurality of groups of bits to eliminate groups of bits that are to be unaffected by a computation, performing the computation, and thereafter decompressing corresponding data after performing the computation to restore data corresponding to previously eliminated groups of bits.

1 . A method for compressing data in a graphics system comprising: 
 providing a mask identifying bits within a plurality of bits that are not to be impacted by a computation; and    generating a compressed bit stream, wherein the bits that are not to be impacted by the computation are not included in the compressed bit stream.    
   
   
       2 . The method of  claim 1 , wherein the providing a mask comprises providing a pixel mask corresponding to a tile of pixels to be displayed on a display.  
   
   
       3 . The method of  claim 1 , wherein further comprising obtaining depth information from a z-buffer and using the depth information for generating information containing within the provided mask.  
   
   
       4 . The method of  claim 1 , wherein the generating further comprises shifting individual bits by an amount equal to a number of bit positions, preceding the current bit position, that are unaffected by the computation.  
   
   
       5 . A method for compressing a plurality of bits comprising: 
 directing data to be compressed to a matrix of multiplexers, wherein the matrix of multiplexers are in a plurality of rows with multiplexers in a first row have inputs connected to signals defining bits to be compressed and multiplexers of successive rows have inputs connected to outputs of the multiplexers of the preceding row, wherein each successive row of multiplexers comprises fewer multiplexers than the previous row; and    controlling data select input signals for the plurality of multiplexers such that individual bits of the plurality of bits are shifted varying amounts, the shift amount being determined by a mask.    
   
   
       6 . The method of  claim 5 , wherein the controlling comprises shifting individual bits of the plurality of bits by varying amounts based on the contents of the mask, wherein each additional shift value effectively causes a shifted bit to overwrite a bit that is to be unaffected by a subsequent computation.  
   
   
       7 . The method of  claim 6 , wherein the controlling is configured to operate control logic to shift individual bits by an amount equal to a number of bit positions, preceding the current bit position, that are to be unaffected by the computation.  
   
   
       8 . The method of  claim 5 , wherein the mask is a pixel mask corresponding to a tile of pixels to be displayed on a display.  
   
   
       9 . A method for a computer graphics system comprising compressing a plurality of groups of bits based on a pixel mask, wherein contents of the pixel mask are derived by a determination of corresponding groups of bits that are to be unaffected by a computation, the contents of the pixel mask defining variable amounts that the plurality of bits are shifted during the compression.  
   
   
       10 . The method of  claim 9 , wherein each position of the pixel mask defines a shift amount for a group of bits.  
   
   
       11 . The method of  claim 10 , wherein the content of each position of the pixel mask is defined by a single bit, and the shift amount for a group of bits is defined by a summation of preceding pixel mask positions whose contents indicate corresponding pixels are not to be affected by a subsequent computation, wherein the positions of the pixel mask are arranged in an order and the preceding pixel mask positions are those that, as ordered, numerically precede a given position.  
   
   
       12 . A method comprising: 
 compressing a plurality of groups of bits to eliminate groups of bits that are to be unaffected by a computation;    performing the computation; and    decompressing corresponding data after performing the computation to restore data corresponding to previously eliminated groups of bits.    
   
   
       13 . The method of  claim 12 , wherein computation is a pixel shading operation.  
   
   
       14 . A method for a graphics system comprising: 
 providing a mask identifying positions within a plurality of positions of a bit stream that are to be removed during a compression operation; and    generating, based on the mask, a compressed bit stream, wherein the positions that to be removed are removed by variably shifting contents of successive positions by an appropriate amount so as to overwrite contents of positions that are to be removed.    
   
   
       15 . The method of  claim 14 , wherein the mask is a pixel mask corresponding to a tile of pixels to be displayed on a display.  
   
   
       16 . The method of  claim 14 , wherein the generating comprises shifting individual bits of the bitstream by varying amounts based on the contents of the mask, wherein each additional shifted bits effectively overwrites bits that are to be unaffected by the computation.  
   
   
       17 . The method of  claim 16 , further comprising controlling the generating to shift individual bits by an amount equal to a number of bit positions, preceding the current bit position, that are to be removed.  
   
   
       18 . A method for a computer graphics system comprising compressing a plurality of groups of bits by shifting compressed groups of bits into bit positions that are to be removed during the compression, the logic being responsive to a mask, wherein contents of the mask define variable amounts that the plurality of bits are shifted during the compression.