Hi Sadfar
No I have not checked A'pipe, but I know a few users and will ask them. I
used to do a bit of factory acceptance testing of PLC programs. It was far
quicker to make things crash if I could see the ladder logic first. The
logic gave me some idea of the program structure; making it easier to locate
potential weak points. With hidden or unintelligible source code, this is
not an option.
For my personal "acceptance" tests on proprietary software, I normally start
with a few impossible boundary conditions (e.g. negative pipe wall
roughness, excessive suction lifts) to check gross error trapping. Next I
try some "almost out of range" conditions (e.g. pipe diameter of 0.00001mm,
density of 0.00001 kg/m^3) with a view to spotting any distortion around
singularities/ end conditions. Finally I do a back to back test on a problem
for which I already have an independent checked solution. Sometimes I find
my checked solution is wrong. The OI interface problems normally appear all
by themselves, so I normally confine my checks to the maths engine.
Cheers
Steve McKenzie
-----Original Message-----
From: Safdar Abbas [mailto:safdar.abbas@trivor.com]
Sent: Tuesday, December 16, 2003 1:13 AM
To: PipingDesign@yahoogroups.com
Subject: RE: [PipingDesign] Re: Stress calculation
Hi Steve,
Have you checked AutoPIPE? I can help you because I'm working on plant design software testing. I would also like to see some industry specific test cases to test the software and see what 'absolutely erroneous crap' my software produces because its my job to find errors in piping design software.
Safdar
-----Original Message-----
From: Steve McKenzie [mailto:mechproj@xtra.co.nz]
Sent: Monday, December 15, 2003 4:10 PM
To: PipingDesign@yahoogroups.com
Subject: [PipingDesign] Re: Stress calculation
I wasnt thinking so much of designing an open source piece of software; more locating and adopting an existing "engine" and seeing if the group members could help in polishing it up. The "payment" would be free access to the latest version.
Problem #1: finding the software, if it exists. It neednt be a piping program; just one with a compatible I/O, database, processing format.
Problem #2: finding members who have the time to commit and the patience to wait for a usable solution.
Problem #3: Finding a coordinator to decide the architecture, split it up,farm it out and then pull it all together again .
Only 3 problems, but #1 could be a doozie.
Pipedreams are free!
Having seen the output from a well respected and expensive surge
analysis program, I am quite sure some clients would be absolutely
horrified to see the amount of absolutely erroneous crap such
programs can produce. The program jockey continues to patch the
boundary conditions until the program can handle the scenario. The
output appears to be deemed to be correct when the jockey can believe
the output ("just ignore those out-of-range spikes to the left of the
printout - I dont know why they are there").
Woohoo.
I was quite frankly appalled. Beautiful printouts, however. Apart
from the spikes, which apparently dont matter.
Cheers
Steve McKenzie
Yahoo! Groups Sponsor
ADVERTISEMENT <http://rd.yahoo.com/SIG=12cdmbrq0/M=267637.4116730.5333196.1261774/D=eg roupweb/S=1705083319:HM/EXP=1071573004/A=1853619/R=0/*http:/www.netflix. com/Default?mqso=60178356&partid=4116730> click here
<http://us.adserver.yahoo.com/l?M=267637.4116730.5333196.1261774/D=egrou pmail/S=:HM/A=1853619/rand=700663577>
Your use of Yahoo! Groups is subject to the Yahoo! <http://docs.yahoo.com/info/terms/> Terms of Service.
[Non-text portions of this message have been removed]
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ Received on Mon Dec 15 17:25:00 2003
This archive was generated by hypermail 2.1.8 : Mon Oct 27 2008 - 20:24:02 EDT