Ex parte TOMINAGA et al. - Page 10




                 Appeal No. 2001,0712                                                                                                                   
                 Application 08/148,887                                                                                                                 


                 combine prior art references.  In re Dembiczak, 175 F.3d 994,                                                                          
                 1000-01, 50 USPQ2d 1614, 1617-19 (Fed. Cir. 1999).                                                                                     
                                   On pages 11-13 of the brief, Appellants argue that                                                                   
                 there is no suggestion in the prior art to combine the                                                                                 
                 teachings of Hitchcock with the teachings of Chi.                                                                                      
                 Specifically, Appellants point out that Hitchcock teaches  a                                          5                                
                 method of automatically synthesizing data paths from a                                                                                 
                 behavioral description of a hardware design.  The EMUCS                                                                                
                 program implementing the method attempts to find a minimum                                                                             
                 cost implementation of a hardware design given a dataflow                                                                              
                 representation VT (Appellants' emphasis).  The program                                                                                 
                 proceeds by binding abstract data flow elements onto hardware                                                                          
                 elements in iterative fashion. Once all the hardware elements                                                                          
                 have been bound, the program determines which implementation                                                                           
                 would result in the optimum design given the parameter which                                                                           
                 is to be optimized.                                                                                                                    
                                                                  6                             7                                                      
                                   Appellants assert  that Chi teaches  a graph-based                                                                   


                          5Page 484, column 1.                                                                                                          
                          6Brief, page 12.                                                                                                              
                          7Page 286, column 1; page 270, columns 1-2 and table 2.                                                                       
                                                                          10                                                                            





Page:  Previous  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  Next 

Last modified: November 3, 2007