Sarah Palin does come across a lot like her namesake. Keep up the lip-service...

gorgeous bush almost ex-presidento

Take a look at OpenFoam it work with polyhedral meshes in the mid-90: Slide 3: http://powerlab.fsb.hr/ped/kturbo/Op...shHandling.pdf

Sarah you sound pissed!

poly existed way before the mid-90s.

now can you move aside and let me take charge of this thing.

do you feel wet???

Yessir

Georgeous Bush ex-Presidento

No doubt. ASC (aka ANSYS CFX) was doing this in 1985. See CFX TASCflow. CVFEM forms a polyhedral mesh.... implicitly in the solver... .

Funny how you did not get your facts straight. The solver was FLUENT, not CFX. The Italy team was the only one in the top three or four not using CFX.

OK, I read this article.

Meshing a single case with 1 billion cells and solving it end to end is a lot different than meshing a case with 1e9/40 cells and copying 40 times.

So, give me a break. What ANSYS and CD have done here is apples and oranges. Someone with FLUENT actually meshed the entire boat or keel or whatever, and solved it... not made copies of some mesh. It's totally different.

I also like the words about CCM++ "modern architecture". FLUENT has been this way for a long long time... longer than CCM++!!

oh yes, please bring it on.

we also broke a 1 billion mark record in November albiet of different sort. We spent 1 billion of campaigning alone.

Breaking into 1 billion is nothing for us. huh.

Ok Sarah might not work for an$y$ but I think it is a safe bet that Joe does!

Sarah these guys are pulling you. After what you have achieved i.e. one billion cells barrier, you have become a celebrity.

Joe have you actually used ccm+? If so you will realise how ridiculous it is to compare it to fluent, I have used both and will never move back to fluent.

P.s. where is ansys 12

"So, give me a break. What ANSYS and CD have done here is apples and oranges."

Apples: One billion polyhedral cells

Oranges: One billion tetrahedral cells a year late.

Not even the American Supreme Court or the FIA would give it to the Oranges.

"I also like the words about CCM++ "modern architecture". FLUENT has been this way for a long long time... longer than CCM++!!"

So Fluent has had a modern architecture for "a long long time". Didn't it occur to you that this fact alone means that the architecture isn't modern.

JV

Well if someone has used both of them and understands how engineering is evolving he knows that the difference between fluent and ccm is huge. ccm is so far behind that this is not even funny.

With all the names of modern architecture all you have got is shinning GUI done is java swing. and that is just GUI not solver.

In past Fluent has been providing you with
[*] pressure based solver[*] density based compressible solver (not present in ccm)

In terms of rubustness:[*] AMG solver[*] Geometric or full multigrid (not present in ccm+)

This was in older versions of Fluent. So with modern solver ccm+ has not even caught up with old archaic solver Fluent.

And how about AMG with RPM methods to solve very difficult systems. This is very modern. Where is this type of modern tech in ccm+.

If I will start to touch models that Fluent gives you for example how about LES versions that Fluent allows you to chose. I think ccm+ does not even provide half of them.

For a modern solver ccm+ is pretty archaic. (lot of catching for them to do).

I am not sure anyone is likely to argue that the range of physics in ccm+ are comparable to fluent, but then one code is about 5 times older than the other so that is to be expected. I admit when I first moved from fluent I was somewhat nervous for the same reason but then modern industrial cfd is so much more than just physics. And ccm+ is catching up very fast...

Fine if you want to be doing academic/R&D level type studies then I would suggest fluent may well be the way to go if there is a very specific "niche" model you need. But if you are doing the day in day out industrial cfd that modern engineering requires then ccm+ is a valid alternative (and in my case preferable). Fluent is just a solver after all, how do you prepare and mesh your geometries?

The other thing to think about is that one code has a future the other doesn't, ansys need a follow on code that will keep both cfx and fluent users happy. The fact that they are still struggling to integrate fluent into workbench, so much so they have missed their yearly release, stands testament to the problems they are having.

I think anyone who has used both and understands how engineering is evolving would know all this. Fluent user, have you used both in an industrial setting?

hi,

i agree with what you say.

my comment was because i am seeing that people come out and rubbish fluent when compared to ccm+, and that too just because gui of ccm+ is cooler.

It comes down to more than the GUI though. No one can argue against the that Fluent used to rule the world when it came to solver technology, but like all old software the rate of development has slowed to a crawl, especially since ANSYS try to hammer it into workbench like a square peg into a round hole.