[Jame5]

What would everyone say is the fastest CFD solver in the world?

[vesp]

It is a question of accuracy vs. computing resources, not which one is the fastest. Being twice as fast but ten times less accurate than any solver is not difficult.

[MangoNrFive]

That depends on a lot of things, to name a few:
- The hardware you're running on
- Mesh-size
- Your type of simulation, boundary conditions etc.
- What you define as fast
For example coupled solvers generally take longer for a timestep compared to uncoupled solvers but reach convergence in less timesteps.

The commercial solvers are probably pretty good and on equal ground I'd suspect. I don't think there is a fastest solver because it always depends on your specific use case and what you want to achieve.

[Jame5]

Ok so what would you think the strengths and weaknesses of ANSYS Fluent are?

[MangoNrFive]

Not a fluent user myself, but as I heard fluent has a wider range of applications and more variety compared to CFX for example.

There are a lot of threads concerning this topic in this forum.

[LuckyTran]

Originally Posted by Jame5

Ok so what would you think the strengths and weaknesses of ANSYS Fluent are?

Okay... but what are you comparing to? If you judge a fish by its ability to climb a tree...

Heck even within Fluent there's a about a dozen different solver combinations that each have strengths and weaknesses over one-another. And to bring back an example already mentioned the coupled solver is nice and converges faster when you are talking about 1 cpu, but it doesn't parallelize well and eventually the slower converging but more parallel-friendly segregated approaches (SIMPLE/PISO) can catch up if you run it on a big enough machine. And this isn't Fluent's fault, it's the algorithm. Fluent is just another piece of code.