Dears, I'm a Fluent user,but now I'm working in a new company who would like to use Star-CD (it seems to be less expencive).

What are the differences between the two codes (pre, post, analysis capability) for industrial applications, according to you?

What are the merits and the defects?

Your opinion is very important. Thanks a lot

Giovanni

Hi: I only know from the sellers of the STAR_CD that initially this code is developed for CAR design(car shape and internal combustion engine design). In order to increase the marketing fraction, some other modules are added gradually later. So generally FLUENT is more "strong" than STAR-cd, but Star-cd have some better modules for CAR design.

HarryCAR design

See STAR user forum.

Chinor

As Harry said, STAR-CD has a unique feature of the simulation of the compression and expansion stroke in internal combustion engines by the so called cell de/activation. That what makes STAR-CD suitable for Internal combustion engines more than Fluent.

I have benchmarked one against the other, and I can tell you that they are very much the same. The solver that is. They are as good or as bad as each other in that sense. Fluent is a bit more user friendly as far as mesh generation is concerned, but that is no big deal. You can always get a separate mesh generation tool and just change your solver.It will still be cheaper in the long run. Star has some mesh movement/removal capability which is very unique (if you have any use for it). In general STAR is cheaper and they are willing to work with the customer more. As far as support is concerned, they both suck, fluent a little more than STAR.

The memory requirements of Fluent are about twice that of STAR.

I do not agree with the comment that user support at Fluent sucks. They are pretty responsive.

Chinor

(1). I think, you should go by the cost first. (2). If it can not solve your problem then, you can always try the other one. (3). There is no sense of using a more expansive product to produce the same results. (4). Since in many cases, the decision to use a particular product is outside the power of the engineer who is going to use it, it is kind of meaningless to say which one is better in a general sense. (5). The suggestion is: negotiate with the vendor and try to find a cheaper one. You can also negotiate with both to get both codes on your system for a trial period so that you can determine which (or both) is suitable for your applications. (6). The code is useless, unless it actually can help you to solve your problems. That's the bottom line. (7). CFD codes are not antique for collection. It has no value if it can not help you to solve your problems. (8). So, what problems are you trying to solve?

Item (3) should read" (3). There is no sense of using a more expensive product to produce the same results".

And what about their facilities whit UDFs and user Modules?

STAR is in Fortran and the user has access to every little variable that there is. You can make the code to do anything and every thing that your heart desires by writing a subroutine. Fluent, however, is not as versatile, i.e., you don't have access to every variable out there. Some parts are open to the user to write udf and some parts are not. And then there is convergence issues. When you compile a UDF for fluent, code convergence goes out the door. With STAR the solver doesnot behave any differently.

S,

your comments are very interesting. Are you saying that for STAR-CD all the internal variables and data structures are WELL documented and accesible??? If that's the case I would congratulate the vendors and hope others will follow that lead.

And in what did your udfs do that affected Fluent??

I'm using Fluent, and really the biggest problem I've had is getting good documentation on the internal data structures of the code. You can access all internal data of fluent through udfs, the only thing is Fluent Inc have not documented howyou do this , and because of the complexity of an unstructured code its not easy to work it out.

I found that Fluent 4.5, which is structured and uses Fortran was much easier to work out the internal data structures.

On stability, I've added udfs for chemical reactions and that works fine despite that how difficult they are to get convergence. In this case reducing relaxation was necessary.

Greg

Hi,

Those familiar with both codes, please comment on the following issue: which classes of problems are better solved by fluent and which - by starcd? Is there any kind of problems which can be solved by one code and ultimately not solvable for the other (in this respect what about transonic and supersonic flows?). Mind ,I'm not talking about pre and postprocessing here.

The bottom line: in case of equal availability of both codes - cost issues not involved here - for a user equally unfamiliar with both codes, which choice is "optimal" depending on the kind of a problem? Thanks in advance.

Yulia.

No You cannot access all the internal data structure in Fluent. Each time the user needs a particular variable which is not accessed, he/she has to go through Fluent Inc. support and you are under their discretion as when they will make it available. It has happened to us many times.

Yes with STAR ALL the VARIABLES are available. AND so easily available that you need very little documentation and you can figure it out easily.

No it does not have to do with the fact that fluent is "unstructured", so is "STAR".

Yes stability is a big deal with fluent. "CODE ROBUSTNESS" is not well thought through in fluent.

Please don't take these comments personal. I know that Cansis is the distributor for fluent in Aus. STAR is realy a more versatile code.

As the codes stand today, their capabilities are very similar except:

Radiation is not very strong with STAR.

Fluent does not have moving mesh/ mesh removal capability that means you cannot do moving boundaries, such as valves opening and closing or heart valves flapping or gears rotating and so on with fluent. STAR has this capability.

Both codes can handle transonic/supersonic flows. However, all the boundary conditions necessary for such flows are not necessarily coded in fluent. AND if these types of flows are in conjunction with moving pistons and moving valves then the comment in the previous paragraph applies.

Just out of interest because I've never worked with STAR. According to all the comments STAR is cheaper, versatile, uses less memory and can do everything that FLUENT does. Why does FLUENT then hold the largest percentage of the world market and is recognized as the leading commercial CFD code?

There is an aspect to any sales that you are missing. AND that is marketting.

FLUENT INC. has by far the best marketting machine than any other code out there including "solid mechanics" codes. In other words this has nothing to do with code capability and code versatility.

The subject of discussion was "technical code capability" and that is what was discussed.

It's not just marketing, Fluent is more user friendly and is much easier to learn. I am friends with people who have sold both STAR and Fluent, and STAR is really tough to sell to a novice user. I used to work for the STAR vendor and had many an opportunity to teach new users how to use the code. The expression "deer in the headlights" is directly applicable.

I think if you want ultimate flexibility with a bitch of a learning curve, then STAR is absolutely the way to go. If you want to get off the ground quickly and not worry about becoming a "power user", then Fluent is the way to go.

Regards,

Chinor

I'm guessing 'Dave' works for Fluent. Anyone who's actually used Fluent wouldn't call it the best code.

Correction Jon - I do not work for FLUENT and I did not say that it is the best code. I am just interested to know why FLUENT sells more licenses if the other codes have the same features at a fraction of the price. As far as I am concerned I think the best codes depending upon your application are the one's developed 'in house'. I can think of a lot of companies who develops their own combustion codes which is more powerful than what STAR or FLUENT can offer. Please read this through 'twice' Jon before replying.

(1). I have just noticed that another aerospace company has changed hands second time in a year. (2) And they were looking for an engineer to do CFD using commercial CFD codes. (3). As I have said before, CFD is the core technology of aerospace company. It covers a wide range of companies. In order to dress up the company for sale, they need codes, pictures, activities in CFD. And commercial CFD codes satisfy this requirement. (4). No engineer can afford buying an expensive license, except the management people in the company. And they normally have a very good story and strategic plan to show you. And if you want to keep your job, it is a good idea to say good thing about the company's decision. (5). So, it is a political and economic game which is linked to the image and sales of companies.