Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
what is difference between comsol version 3.4 and comsol version 3.5a?
Posted 27 gen 2010, 21:47 GMT-5 1 Reply
Please login with a confirmed email address before reporting spam
I have a simulation file of comsol old version software completion(comsol version3.4),now opening with the comsol new version(comsol version 3.5a). I didn't modify any parameter,after re- slicing a mesh but appear a different simulation result,what is difference between comsol version 3.4 and comsol version 3.5a? Would you give me some suggestions?
Thanks a lot in advance for your help.
Best Regards
Yu-Mei
Thanks a lot in advance for your help.
Best Regards
Yu-Mei
Attachments:
1 Reply Last Post 28 gen 2010, 01:37 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
You have an important question there. I havnt observed that before, but such things happens.
I would suggest that you carefully check the BCs (boundary conditions), as the renumbering/indexing of your geometry might have been affected, at least that is where I would start.
There could also be some improvements on the normalisation of the solvers, if some of your matrices were ill conditionned then I would suspect that the newer results should be more precise.
I would also propose to dump the 3.4 as a "reset model" and "save as" ".m" file with a V3.4, then run it and generate a report. Then repeat the same procedure in v3.5a.
Finally, run a text comparator (à la Windiff or UltraCompare for me) on the two versions to try to identify any differences, as well on the .m file as on the html report
Unfortunately I have only a V3.5a running now.
Keep us informed, as this is important,
I assume "support" could also give you some clues, the "whats new" doc might bring you also some ideas.
Good luck
Ivar
You have an important question there. I havnt observed that before, but such things happens.
I would suggest that you carefully check the BCs (boundary conditions), as the renumbering/indexing of your geometry might have been affected, at least that is where I would start.
There could also be some improvements on the normalisation of the solvers, if some of your matrices were ill conditionned then I would suspect that the newer results should be more precise.
I would also propose to dump the 3.4 as a "reset model" and "save as" ".m" file with a V3.4, then run it and generate a report. Then repeat the same procedure in v3.5a.
Finally, run a text comparator (à la Windiff or UltraCompare for me) on the two versions to try to identify any differences, as well on the .m file as on the html report
Unfortunately I have only a V3.5a running now.
Keep us informed, as this is important,
I assume "support" could also give you some clues, the "whats new" doc might bring you also some ideas.
Good luck
Ivar