Forgot your password?
New user?
#686: Computed field integration doesn't handle inconsistent xi
View
(Anonymous)
issuedata
Classification:
Feature
Project:
cmgui
OS:
Version:
Importance:
Medium
Assigned to:
Created by: stevens
Created at: 2008-05-09
progress
Deadline:
2008-07-08 11:32
Percent done:
0
contact
Name:
Richard Christie
Description:
Where adjacent elements have opposing xi directions, integrated values from the last element are assumed to be carried forward to xi=(0,0,0). This gives the following 1-D integration of xi: 0-----xi->---1 1---<-xi-----0 0 1 2 1 What we want is: 0 1 1 2 This can be achieved by noting the true xi location for the initial_values, i.e. xi=1 for the second element in the above example, and ensuring the integration increases down xi. Question: Is there any reason why this new behaviour cannot be on at all times?
Transcript
#2:
2008-05-09 17:22 (blackett)
It was a really long time ago, I can't remember if there was a reason I thought I couldn't handle inconsistent xi, but I did go to a lot of effort to change meshes to have consistent xi so I must have thought it was important at the time.
#1:
2008-05-09 11:59 (rchristie)
categories: "()" ->
"['']"
OS: "Linux" ->
"['']"
version: "Development" ->
"['']"
title: "" ->
"Computed field integration doesn't handle inconsistent xi"
description: "" ->
"Where adjacent elements have opposing xi directions, integrated values from the last element are assumed to be carried forward to xi=(0,0,0).
This gives the following 1-D integration of xi:
0-----xi->---1 1---<-xi-----0
0 1 2 1
What we want is:
0 1 1 2
This can be achieved by noting the true xi location for the initial_values, i.e. xi=1 for the second element in the above example, and ensuring the integration increases down xi.
Question:
Is there any reason why this new behaviour cannot be on at all times?
"
PloneCollectorNG (C) 2003-2004 by ZOPYX - Software Development and Consulting Andreas Jung