<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=utf-8">
<TITLE>Message</TITLE>
<META content="MSHTML 6.00.2800.1458" name=GENERATOR></HEAD>
<BODY>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2><SPAN
class=734431008-19102004>Hello Nexus
people, </SPAN></FONT></FONT></SPAN></DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2><SPAN
class=734431008-19102004></SPAN></FONT></FONT></SPAN> </DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2>Following the
request from the NIAC for instrument editors to try writing a definition using
the current base classes, </FONT></FONT></SPAN></DIV>
<DIV><FONT face=Arial size=2><SPAN class=343124907-19102004>I've constructed a
generic inverse geometry TOF machine template at:</SPAN></FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial color=#000000 size=2><A
href="http://www.neutron.anl.gov:8080/NeXus/88">http://www.neutron.anl.gov:8080/NeXus/88</A></FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial size=2>One of the
interesting things about constructing the definition is the need to decouple
notions of logical neutron path in describing the instrument, and instead to
consider the instrument symmetry. This is a very different approach to
constructing instrument definition files for Monte Carlo
simulations.</FONT></SPAN></DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial
size=2></FONT></SPAN> </DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2>The result is
that analysers, collimators and filters are considered in much the same way as
banks of detectors. This leads to an implied order and relationship between
'arms' of the instrument, which is currently not handled by the definition. For
example, it is implied that the collimator in position 1 of the collimator bank
array is associated with the analyser in position 1 and the detector in position
1. Currently, if these positions are mixed up, then it is only by reconstructing
the instrument via the (x,y,z) positions of each element, and visually examining
the result that any sense will be made of the description.<SPAN
class=734431008-19102004> </SPAN></FONT></FONT></SPAN></DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2><SPAN
class=734431008-19102004></SPAN></FONT></FONT></SPAN> </DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2><SPAN
class=734431008-19102004>I've also had to do some major hacking of the
NXdetector class to make sense of it, much like Tim Kelly's previous post to
this list. </SPAN></FONT></FONT></SPAN></DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial
size=2></FONT></SPAN> </DIV>
<DIV><SPAN class=343124907-19102004><SPAN class=734431008-19102004><FONT
face=Arial size=2>Regards,</FONT></SPAN></SPAN></DIV>
<DIV><SPAN class=343124907-19102004><FONT face=Arial><FONT size=2>Martyn<FONT
color=#0000ff><SPAN class=734431008-19102004><FONT color=#000000> Bull,
ISIS</FONT> </SPAN><SPAN
class=734431008-19102004> </SPAN></FONT></FONT></FONT></SPAN></DIV></BODY></HTML>