Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
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.
How many out here like 4.2a and have something to say ?
Posted 28 oct. 2011, 16:12 UTC−4 Version 4.2a 9 Replies
Please login with a confirmed email address before reporting spam
Hi all
I'm just back from the Stuttgart conference, and I am testing out the new 4.2a, I like it :)
And what are your thought's out there ?
I have noted so far:
- improved documentation and help,
nice new features, such as
- - complex variable separation in sin-cos type or two real variables, (new option in the Solver Compile equation node)
- - copy mesh domain with automatic mirror and one to mutiple copy options
- - new combined physics (i'm starting to get lost with so many combinations ;)
- - impressive particle tracking
- - solid/structural tries to detects large deformations their settings are now in the study - step1 ... node
- - improved geometrical "measure data" (still hoping for mass, CoG, inertia tensor info too, soon ;) Try it, by selecting an entity and hit "measure"
- - the delete and rename (F2) window can be moved and reappear where last left, and not aleways in the middle of your screen
- - reports take by default the name of the model, so you do not need to rename it at each model "save-as"
- - a new forth Frame; the "geometrical frame" see help "frames"
- - in fact a lot of small things to make easier and make it qicker to set up an analyse a model
still discovering, just started ...
--
Have fun Comsoling
Ivar
I'm just back from the Stuttgart conference, and I am testing out the new 4.2a, I like it :)
And what are your thought's out there ?
I have noted so far:
- improved documentation and help,
nice new features, such as
- - complex variable separation in sin-cos type or two real variables, (new option in the Solver Compile equation node)
- - copy mesh domain with automatic mirror and one to mutiple copy options
- - new combined physics (i'm starting to get lost with so many combinations ;)
- - impressive particle tracking
- - solid/structural tries to detects large deformations their settings are now in the study - step1 ... node
- - improved geometrical "measure data" (still hoping for mass, CoG, inertia tensor info too, soon ;) Try it, by selecting an entity and hit "measure"
- - the delete and rename (F2) window can be moved and reappear where last left, and not aleways in the middle of your screen
- - reports take by default the name of the model, so you do not need to rename it at each model "save-as"
- - a new forth Frame; the "geometrical frame" see help "frames"
- - in fact a lot of small things to make easier and make it qicker to set up an analyse a model
still discovering, just started ...
--
Have fun Comsoling
Ivar
9 Replies Last Post 2 janv. 2012, 13:11 UTC−5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Major new capabilities at 4.2a are listed here: www.comsol.com/products/4.2a/
More details are available in the Release Notes manual, as always.
More details are available in the Release Notes manual, as always.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Jean-François
Certainly they are listed, but which one affect whom, and how ?
And how do we discover colateral effects ? That we users (or perhaps not even you at COMSOL) do not necessarily think of ?
That is the experience I'm interested in to hear about, to avoid repeating error other have debugged for me ;)
I can come with one thing: back in 3.5, I believe, I got nicely "burned" with antisymmetric BC conditions and the large deformed geometry in solid/structural, these are not compatible. By the way, now there is a red warning mark in the doc (under solid) when you read it through, so it's stated, but there are so many thing in the doc ;).
So by now, I always check if the "large deformation" is "on" if I use antisymmetry BC's.
The way large deformations is turned "on" in v4.2a has changed, good (now you can enforce small deformation, the opposite of 4.2 and let COMSOL decide for you, when large deformations are required, tat is the way I understand it).
If someone dig up a large trench across your usual road to the way to work, you are allowed to bump into it, ONCE, (while thinking on how to set up your next model, and not observing the usual road tracé), BUT NOT twice.
Certainly we all would prefer to not bump at all ;)
So, other comments ?
--
Have fun Comosling
Ivar
Certainly they are listed, but which one affect whom, and how ?
And how do we discover colateral effects ? That we users (or perhaps not even you at COMSOL) do not necessarily think of ?
That is the experience I'm interested in to hear about, to avoid repeating error other have debugged for me ;)
I can come with one thing: back in 3.5, I believe, I got nicely "burned" with antisymmetric BC conditions and the large deformed geometry in solid/structural, these are not compatible. By the way, now there is a red warning mark in the doc (under solid) when you read it through, so it's stated, but there are so many thing in the doc ;).
So by now, I always check if the "large deformation" is "on" if I use antisymmetry BC's.
The way large deformations is turned "on" in v4.2a has changed, good (now you can enforce small deformation, the opposite of 4.2 and let COMSOL decide for you, when large deformations are required, tat is the way I understand it).
If someone dig up a large trench across your usual road to the way to work, you are allowed to bump into it, ONCE, (while thinking on how to set up your next model, and not observing the usual road tracé), BUT NOT twice.
Certainly we all would prefer to not bump at all ;)
So, other comments ?
--
Have fun Comosling
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I didn't try it yet, but I know that it will be possible to assign units to dependent variables in the math modes, which is very nice. It will not improve the results, but it will remove a lot of annoying orange lines...
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
and perhaps a few undetected unit conversion errors ?
I have always been in favour for strict unit verifications, and when I try to skip over, I often get caught in stupid errors ;)
--
Good luck
Ivar
and perhaps a few undetected unit conversion errors ?
I have always been in favour for strict unit verifications, and when I try to skip over, I often get caught in stupid errors ;)
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Ivar,
Well, it was always possible to end up with no units warning, but we had to twist the units of some parameters/variables in order to avoid the warning, and, as you say, it could become a source of error and quite an headache. Now this headache will be over!
François
Well, it was always possible to end up with no units warning, but we had to twist the units of some parameters/variables in order to avoid the warning, and, as you say, it could become a source of error and quite an headache. Now this headache will be over!
François
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I just wanted to say that I tried v4.2a at the Boston Conference and a little bit after the conference, but I really have not started to dig in until the last few days after I received the new license file and can use it routinely for all my work. I have noticed several little improvements, but two major improvements that help me in my work:
1) the automatic scaling has been improved whereby the scaling factors are adjusted through the first two iterations as opposed to just the first iteration. This has helped in the convergence dramatically. I was doing a lot of manual scaling, but I think I will now do much less.
2) for the direct solver, it appears, but have not confirmed, that the solver is using less memory. Has anyone else seen this ?
I am sure I will find many more great improvements in addition to the "macro" level improvements that are mentioned in the release notes. It is all the little things that impress me !
1) the automatic scaling has been improved whereby the scaling factors are adjusted through the first two iterations as opposed to just the first iteration. This has helped in the convergence dramatically. I was doing a lot of manual scaling, but I think I will now do much less.
2) for the direct solver, it appears, but have not confirmed, that the solver is using less memory. Has anyone else seen this ?
I am sure I will find many more great improvements in addition to the "macro" level improvements that are mentioned in the release notes. It is all the little things that impress me !
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
There used to be a function under View where i could see the model out of scale, it is missing now..
It used to be very useful, but now i have to enlarge a model many times to select an edge or side
It used to be very useful, but now i have to enlarge a model many times to select an edge or side
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
I think that you mean the "Preserve aspect ratio" check box, which can be useful to clear (at least temporarily) to make slender geometries fill the Graphics window. You find it in the Settings window for the Camera subnode under the View node for 3D geometries and in the Axis subnode under the View node for 2D geometries.
Best regards,
Magnus Ringh, COMSOL
I think that you mean the "Preserve aspect ratio" check box, which can be useful to clear (at least temporarily) to make slender geometries fill the Graphics window. You find it in the Settings window for the Camera subnode under the View node for 3D geometries and in the Axis subnode under the View node for 2D geometries.
Best regards,
Magnus Ringh, COMSOL
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
A general rule for V4: use the right click everywhere (even at the cost/risk of a "tennis elbow" ;)
and be aware that when you change some physics settings, or BCs new items (nodes, sub-nodes) appear in your "right-click lists". So do not expect you do know 100% what is under a node, new items might appear ;)
Do not forget that in Results you also have 2D and 3D Views, there too you might need to change the "preserve aspect ratio" for your result plots
--
Good luck
Ivar
A general rule for V4: use the right click everywhere (even at the cost/risk of a "tennis elbow" ;)
and be aware that when you change some physics settings, or BCs new items (nodes, sub-nodes) appear in your "right-click lists". So do not expect you do know 100% what is under a node, new items might appear ;)
Do not forget that in Results you also have 2D and 3D Views, there too you might need to change the "preserve aspect ratio" for your result plots
--
Good luck
Ivar