Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!ucsd!ucbvax!CAPSRV.JHUAPL.EDU!koontz%capvax.decnet
From: koontz%capvax.decnet@CAPSRV.JHUAPL.EDU ("CAPVAX::KOONTZ")
Newsgroups: comp.sys.transputer
Subject: RE: I/O Calls on the Transputer
Message-ID: <9007301945.AA22808@tcgould.TN.CORNELL.EDU>
Date: 27 Jul 90 21:23:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 12

Netters,

    Woops!  A correction.  In my previous message on iocall.c benchmarking, the 
numbers for my B008 results are backwards.  It should be:

    Trial 1: DMA disabled (polling)  4.231 sec
    Trial 2: DMA enabled             3.798 sec

Yes, enabling DMA makes it go a little faster (as one would expect).
Sorry, but it's Friday...nuf said.

Ken Koontz  koontz@capsrv.jhuapl.edu