From cary Thu Feb 14 17:07:27 2008
Return-Path: <cary>
Received: (from cary@localhost)
	by afone.as.arizona.edu (8.11.7p3+Sun/8.11.6) id m1F07QT29357;
	Thu, 14 Feb 2008 17:07:26 -0700 (MST)
Date: Thu, 14 Feb 2008 17:07:26 -0700 (MST)
From: Cary Kittrell <cary>
Message-Id: <200802150007.m1F07QT29357@afone.as.arizona.edu>
To: rallen
Cc: swarner
Subject: what I was suggesting was:
Content-Length: 605
Status: O


I think we could test your idea with only trivial
code modifications by:

    1) editing the four missing LVDT's out of `nailMap.c' --
       don't "mask", but remove them entirely -- and then
       
       
    2) adjusting the value of `NumNails' to 12.
    
    
    
This way, instead of trying to ignore "masked" LVDTs 
at all points in the code -- something we're now not
sure that we do in all places -- the program would now read only
the good LVDTs, and do calculations based entirely on
those readings.

If you are correct, then I would expect much better
performance immediately.


-- cary