Appeal 2007-1573
Application 10/705,094
indicating that Narayanaswami does not expressly state that the processors
are different hardly requires that the processors are the same. At best, the
reference is ambiguous on this point.
We note, however, that Narayanaswami does indicate that allocation
techniques are dynamic such that a particular pixel or display region can be
processed for a first period of time and then rendered by a second processor
for another period of time (Narayanaswami, col. 2, ll. 7-18) (emphasis
added). But this statement, at best, merely acknowledges that the respective
processors will process their workloads in a given period of time. It does not
mean that the time taken by each processor is analyzed and used as a basis
for adjusting the respective portions of graphics image as claimed. Although
processing a given workload will certainly consume a certain amount of
time, and each processor’s workload in Narayanaswami is used as a basis for
allocation,7 there is simply nothing in the reference that indicates
specifically that the time taken to process the workload is used as a basis for
allocation. There are just too many variables in Narayanaswami’s system
that could affect processing time apart from the workload itself.8
Furthermore, the Examiner’s assertion that it is well known that
designing homogeneous type multiprocessors is “much easier” than
heterogeneous type multiprocessors is completely unsupported and, in any
event, is simply not pertinent to an anticipation rejection. Even assuming,
without deciding, that the Examiner is correct in this assertion, the statement
(conceding that Narayanaswami does not explicitly disclose that the
processors are identical or homogeneous).
7 See Narayanaswami, col. 5, ll. 11-13.
8 See, e.g., p. 4, supra, of this opinion (listing other factors affecting
processing time).
7
Page: Previous 1 2 3 4 5 6 7 8 9 Next
Last modified: September 9, 2013