CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Future of GAMBIT (https://www.cfd-online.com/Forums/fluent/46971-future-gambit.html)

Kasper January 9, 2008 05:30

Future of GAMBIT
 
What is in your opinion the future of GAMBIT?

Will it still be developed and maintained after the merge of Fluent and Ansys?

What about in 5, 10 and 15 years?

The reason is that we have recieved an offer for a one time license fee.

BR Kasper

Maik January 9, 2008 12:35

Re: Future of GAMBIT
 
Currently Gambit seems to be in the year 1999. IMHO there will be more and better plugins or add-ons for the common CAD-packages. It is too time-consuming to generate geometries in Gambit - so this function is needless.

Rajit January 10, 2008 00:53

Re: Future of GAMBIT
 
Gambit is going to get obsolete for sure in few more years to come. It is simply a issue of paying the gambit providers. I was told by my provder that ansys design modeler and ansys CFX-mesh will become the preprocessor and more likely the fluent solver might be retained.

it is a bit of gamble. No one knows anything for sure unless and untill we hear an official news from ANSYS.

Thanks Regards Rajit

Karl January 10, 2008 02:45

Re: Future of GAMBIT
 
The last information that i get was that further development of GAMBIT has been already discontinued. Some features as for example the sizing functions will be implemented in the ANSYS workbench. Only TGrid with his Surface Wrapping technology will be further developed.

Kasper January 10, 2008 03:37

Re: Future of GAMBIT
 
Thank you for your contribution to the topic.

Yes I agree that Gambit sometimes is time consuming. But if your compare Gambit to Star-CCM+ which is the only option for mesh generation that I have currently, you will see that you have many more options in Gambit to control your mesh (size functions, pave, cooper...).

In my opinion Gambit is very powerfull for those that know the craft.

Best Regards

Kasper Skriver


Kasper January 10, 2008 03:46

Re: Future of GAMBIT
 
Thank you,

Where did you get this information?

Best Regards Kasper Skriver

Razvan January 11, 2008 03:30

Re: Future of GAMBIT
 
Up to this moment, I've been using ICEM CFD, ANSA, TGRID, CFX-Pre and GAMBIT. All of them have their strenghts and weaknesses.

When it comes to structured meshing, ICEM HEXA is undoubtely the best, the only improvement I can think of is the smoothing algorithms (which I think are best in GRIDGEN).

But when in need of unstructured meshing, the problem is a little more complicated:

- ANSA has excellent capabilities, but it is kind of "too difficult" to deal with sometimes (some operations are heavily slowed down and un-necessarily complicated by the FEA basis of the software), and it lacks mesh sizing control, especially for volume meshing (it has some degree of control, but way too general);

- ICEM TETRA is nice, but it is quite difficult to use,and the octree-based volume meshing algorithm is producing low quality meshes, not good enough for CFD;

- CFX-Pre is very simple, but also simplistic, and it is not a real preprocessor for the serious CFD user;

- TGRID is user-friendly enough, the semi-automatic procedure generates good volume meshes, and it has very powerful mesh-editing capabilities, but has no mesh sizing control, which is its biggest flaw (apart from the fact that it is only a half-preprocessor, as it needs a surface mesh to start with);

- GAMBIT is relatively easy to handle, produces excellent 2D and 3D unstructured meshes with the help of the size-functions (and also by the optimization of the defaults), but the geometry-handling leaves much to be desired and although seriously improved in the last couple of releases, is still not appropriate.

So, in my opinion, none of these softwares is perfect. But the best possible combination would be:

ANSA (geometry handling functionalities and 2d meshing)

+ TGRID (3D meshing and mesh editing capabilities)

+ GAMBIT (2d & 3D meshing and size-functions)

+ ICEM (GUI and some geometry handling functions).

I can only hope that at least the TGRID+GAMBIT+ICEM combination would be possible in the case of complete GAMBIT discontinuance.

All the best,

Razvan

Nathan January 18, 2008 19:57

Re: Future of GAMBIT
 
CFX pre is not a meshing tool and should not be compared to gambit. CFX pre is the GUI pre-processor the writes the definition file that the CFX solver uses. Perhaps Razvan meant CFX-mesh and/or Ansys Wrokbench Meshing.

Carlos January 19, 2008 16:50

Re: Future of GAMBIT
 
Try this: http://www.ansys.com/special/icem-cfd-qanda/

Razvan January 21, 2008 05:29

Re: Future of GAMBIT
 
Well, I didn't now most of the things I read there, so thank you, Carlos!

I guess the things are somehow going in the right direction, although I think I will always hate DesignModeler, and never use it. I will probably still use Gambit, TGrid and ICEM combined, until something or someone will be able to convince me otherwise.

All the best to everyone,

Razvan


All times are GMT -4. The time now is 17:46.