Appeal No. 96-2379 Application 07/837,240 We agree with the examiner that it would have been obvious to one with ordinary skill in the art to first hold or buffer the resultant value in a register and then output the same. There is no reason why one with ordinary skill in the art would perceive that the resultant value must be provided directly from the adder and cannot first be placed in an output register. The basic skills of one with ordinary skill in the art would include such common sense and logical reasoning. Indeed, even the appellant's own specification describes and illustrates the use of output registers by prior art dithering techniques (Figure 3). The appellant argues that while the claimed invention recites only a single adder, Comins makes use of two adders in sequence. The argument is misplaced. The rejection on appeal is one for obviousness, not anticipation. As is shown in Comins' Figure 2, the carry bit from the output of the first adder 264 is inputted to the second adder 265 which also take the more significant bits of the calculated pixel value as input. From the perspective of one with ordinary skill in the art, the two- adder structure of Comins is equivalent to a single larger adder. Logic dictates that adding the lower bits of a number to a value in a first adder and then feeding the carry data to a second adder which also takes as input the higher bits of the same 7Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 NextLast modified: November 3, 2007