Appeal No. 1997-3273
Application No. 08/397,910
(Brief, page 10) that the prior art fails to show the claimed
subject matter. We agree.
Bronikowski is directed to a system for processing and
prioritizing alarms, not memory errors. Each alarm is sent to
one of three alarm sub-queues according to its severity. (See
column 9, lines 37-54, and column 10, lines 23-28).
Bronikowski does not store the addresses of memory errors.
Further, Bronikowski states (column 11, lines 9-13) that a
standard error code is returned when an alarm is written to an
alarm queue which is already full, rather than disabling the
clock responsive to the overflow.
The examiner applies Kimmel as showing (Answer, page 6)
that it "was notoriously well known in the art" to stop a
clock to prevent overflowing a buffer. However, merely that
it was well known to stop the clock to prevent overflow of a
buffer does not render it obvious to do so in a particular
system. For example, Kimmel is directed to an image
processing system with data buffers, rather than memory error
queues. Kimmel discloses (column 42, line 55-column 43, line
4) stopping the system clock before overflowing the data
buffer. However, nothing in Kimmel suggests stopping the
4
Page: Previous 1 2 3 4 5 6 7 Next
Last modified: November 3, 2007