[EthanWalker]

Hi,
Happy new year to all.

I have seen many historical posts on this forum about why to choose one software over the other.

Many people suggested at the time that difficulties in meshing with Gambit and Fluent was the driving force for using Star CCM. however, since there has been many iterations of both programs since and many of the short comings have been overcome has the choice of one over the other simply come down to user preference such as Apple vs Samsung?

Or do the programs have more significant strengths and weaknesses?

Similarly, with Openfoam although an open source software do people feel that the reliability of results of an opensource software is significantly lower than its commercial counterparts.

I believe an updated comparison of the strengths and weaknesses would be invaluable to this community.

Looking forward to informative replies.

[LuckyTran]

Originally Posted by EthanWalker

I believe an updated comparison of the strengths and weaknesses would be invaluable to this community.

If you have already seen the comparisons, there's not much to update and I don't see any value in doing an update every year.

Originally Posted by EthanWalker

Similarly, with Openfoam although an open source software do people feel that the reliability of results of an opensource software is significantly lower than its commercial counterparts.

Computer code is not like human beings, it doesn't discriminate. All these codes do exactly what you tell them to do. Question is whether the user actually understands what they are doing? Each code reliable calculates what its code tells it to.

There really isn't much difference between Fluent/Star-CCM/OpenFoam in terms of what each code is trying to do. The biggest driver is personal preference. The major differences is between available solvers for specific types of problems. For example, Fluent had really nice combustion models for a long time. But then StarCCM implemented the same models (by stealing/hiring the software engineers from Ansys). OpenFOAM also did the same. But given the same models and same setup, each code does almost the exact same thing. But the user-interface for each of these is very different from one-another.

There are still some technical shortcomings in each software but I don't feel like this is in general enough to break a deal. For example, Star-CCM doesn't permit bounded central differencing as a discretization in a lot of places. But if you want to run a case with bounded central differencing and choose another program over it, that might as well be a personal preference.