Hi guys,

I was just wondering, I am not using any commercial software, but I always see messages posted with questions such as:

- how do I implement boundary conditions with Software x version y (e.g. fluent, CFX or whatever)?

- How do I do simulate turbulent flow, etc...

So, are not all these companies providing Customer Services for that kind of questions? Aren't they helping people to handle their softwares and the black box? Or are these questions too technical for the software people (who might just be busyness people?)? Or is it just too expensive to have that service?

It is not that I am curious, but I tried hard to understand all that.

Since this site is supported by these companies, I understand that it might be for them one way of providing an alternative to the Customer Services, provided for free by a public forum indeed.

Dear Patrick,

I could not resist my temptation of replying to you. I think this is the best joke I heard in this year. I was wondering that the companies have so many CFD analysts to do the simulations but why is the Department of Energy is after all companies ! I am also wondering why a big company like General Motors got slammed by 4.9 billion dollar penalty in a car crash lawsuit! Don't they have 'Customer Services' who can tackle this kind of problem happening again in future? Thank you very much Partick to bring up this kind of questions.

(1). I am getting the feeling that it is a "junk mail" scheme. (2). Or something like "Internet Hit" statistics. (3). The goal is to sell more copies of the code, to create more statistics, not to solve any problems, GOOD GRIEF !!!

Dear John,

A bad workman is always contentious or belligerent with his tools. The problems are not with the commercial softwares, the problems are mostly with the persons who are using it. If the person claims that he or she is a good CFD analyst, no matter what, he or she will come up with may be not very good results but at least reasonable results. However, commercial softwares have some limitations. In some cases, codes are written by an individual or a group of individuals to serve special perpose which have great significant importance in science and engineering.

Patrick,

I've had formal training in CFX and Fluent, informal training with Flow-3D.

I've used all 3 codes, Flow-3D the most, then CFX, and then Fluent.

While I'll not attempt to rank the services, I'll just say that the response from all three ranged from good to excellent, and it varied over this range for any one provider from time to time.

It is possible to obtain these codes (older versions at least) without paying for the customer support features, especially in parts of the world where copyrights are not honored uniformly. I believe that 'educational' licenses without support are or were available. In these cases, the companies will not provide support and alternatives must be sought.

It is also possible that an individual does have access to support, but is frustrated by communications problems with the staff persons on duty. So he or she tries this site as an alternative.

Other possibilities ??

Thanks Jim,

for taking my question seriously.

I had the impression that the CFD companies are doing a serious job (just by looking at their webpages) and I was not sure where the problem was. But what you write makes sens. I guess this is happening with a lot of commercial softwares.

(1). I really like to get into this general discussions, but I can't right now. (2).I am checking out this 3-D commercial code, and we have network and server problems today. I can only check this forum when the code is running. (3). Sure, it is the user's problem. There are free software everywhere. It is not the problem with the software. (4). I think it is perfectly all right to ask any questions here. And it is also all right not to answer any questions. (5). Recently, I have been interested in various kind of rollerball pens. And it is interesting to know that there is also a pen forum site on the Internet. There was a question asked about how to use a fountain pen also. (6). The questions asked here really were not that important. The real problem is in the Internet War. The netzero web site news said yesterday that China has invaded Taiwan's Internet sites because they considered Taiwan (Free China) is part of communist China. So, both sides have engaged in this serious Internet War. This is something you need to worry about from now on.

I doubt if any of the CFD companies that support this site does so because this forum could give support for their codes. In fact, I even know that some vendors have instruced their employees not to provide support here - they want to remain in control of the user support since it provides important feedback, warrants the high prices they ask for their codes and also gives them better control of problematic customers - they avoid "bad press".

My experience from support from commercial vendors is also that it is usually very good. The major complaint I have about these vendors is that they often release beta or even alpha versions which are very buggy.

Another question is if it would be good to have a "support dialoge" between users here concerning commercial codes. It will of course take space and irritate some people, but it could also be valuable for many users of commercial codes since it can give an overview of common problems with different codes (and common solutions). I'm not sure what I think about this myself. Perhaps it's time to start a separate forum for discussions about commercial codes?

The worst enemy of a commercial CFD vendor is a bad user. The problem is exacerbated when the bad user is too lazy to call the support desk yet capable of posting a support question to this forum. As already pointed out, the quality of support from all major commercial vendors is high . There really is no excuse for users in maintenance to post innane support questions here that certainly give the vendor a bad reputation. If the vendor has been selling cheap no-maintenance academic licenses that result in a whole range of support questions posted here then it serves the vendor right.

At the end of the day it is the business of business to make money. Commercial CFD is not divorced from this, it may have had academic roots but it is now thriving in a commercial environment. Respect this

Fred.

We have an 'educational' CFX license and even with this license we get good support from their helpdesk. As they prefer e-mail, it usually takes a while before we get an answer but their answers always solved our problems.

I think one of the reasons for the answers Patrick mentioned, is that this support is easily accessible as e-mail and the internet generally is. It might be that they feel silly towards the software developer's "specialists".

Gert-Jan

Sounds like you own Aavid stocks or something ;-) I see nothing wrong in users asking eachother about help, even if they could get the same or better help from the CFD code vendor. As a customer you are free to do whatever you want, even tell others that this and that code s*cks. I think most people who follow this forum will have the ability to sort out what is relevant and what is not anyway. The major concern I have is that the volume of these support related discussions can become very high and thus make other discussions difficult to follow. If this happens and if many people find that the support questions are irritating then we can start a separate forum for that - there seems to be a need for it soon.

I agree with you, Jona, although it doesn't sound like an Aavid stock owner. I don't have a gut to call our client enemy.

Errrrrrrrr? Client Enemy? Your parent company?

I think there is a real need for a forum to discus commercial CFD code - and how they work. It is my own experience that supportdesks in general are good and fast - I always get a useable answer whenever I have a problem. But it is getting increasingly expensive to use the supportdesks.

I often find it necessary to make "workarounds" to get my models to work - and I actually use a lot of time reprogramming the CFX code through the Fortran routines to implement new physical descriptions. However, this saves a lot of time compared to building the model from the bottom up. It would be nice to have a forum to give and receive tips on how to get things working in these codes without getting harsh words in return.

So…. If it is possible to split this forum up in two parts : One for theoretical, practical and experimental CFD and one for application of commercial codes then it would be a good thing. It would remove most of the questions which some finds irritating and other find challenging.

Regards

Michael

Good idea. Maybe though the commercial vendors who sponsor this site may want some low level of moderation, or maybe not.......

(1). A company uses commercial CFD codes, because it does not want to develop its own CFD capabilities, that is to use computer simulation to solve thermo/fluid problems. (2). This is purely economy. (3). The use of commercial CFD codes was considered an intermediate solution to this money saving approach. (4). But even with this approach, many companies are sub-contracting out the CFD analysis part of work. So the company does not even have to worry about the technical support and license fee at all. (5). As a result, CFD becomes a consulting business again, as it should be right from the begining. This is because when a code is detached from the developer or the experienced engineer, it can no longer function by itself. (6). What the user company need is the solution to his problem, not the experience to use a particular commercial CFD code. The need for solving a real problem is always there, but the need to use a particular commercial CFD code is not. (7). A company with its own investment in CFD capability can survive, because they have the control over the employee. A company uses a commercial CFD code can not survive in the long run, because this is a short term solution only. ( This does not mean that a company with its own CFD capability will definitely survive in the market place of tomorrow.) (8).Since there are always companies willing to cut cost in in-house CFD development , the need for temporary CFD consultants is always there. And this is the real market for CFD today. That is a CFD consultant may need some handy CFD tool box to make a living. (9). As for a large company using commercial CFD codes, they normally realize the risk involved, and have several commercial CFD codes on their systems. These are the companies which have technical problems in the first place. So, when a large company is using a commercial CFD code, you have to be careful and ask why? Actually, these companies should have their own CFD development center in the first place. (10). For the CFD code vendors, I would strongly suggest that they move into consulting business. A code just like a PC, it will become obsolete quickly. (11). As for the CFD-online, I don't think it will have any impact at all even if it changes its format everyday. (12). The point is, it is impossible to make a code like a human being. So, you are trying to attach it to the user. In this way, you think it would become something like a human being. But, the first reaction from the user is the body will reject the foreign body violently first. (13). You think everybody like to use only MS software? No, it's because they have the control over the market. (14). In reality, it is a global war, a large-scale CFD war. It would be a miracle, it a company can survive purely based on the use of commercial CFD code. This is the insider's point of view. So, to support or not to support is itself not an issue.

(1). I could not understand why "ANONYMOUS" was used on the network before. But now, I think, it is an excellant idea. (2). I am using free Yahoo e-mail, I am also using free netzero web server, so I can get on this free CFD-online. (3). It is important to learn how to free ourselves from the name, emotion and ... when surfing on Internet. (4). Things will work out smoothly when we use "ANONYMOUS USER", "ANONYMOUS VENDOR", "ANONYMOUS CODE", "ANONYMOUS METHOD",,,... (5). INTERNET was born FREE, and should remain FREE, regardless of whether one likes it or not.

I think it is a good idea to split this discussion forum: Commercial code user groups, Test cases and validation, Solvers, Turbulence models, ...

(1). The way this version of cfd-online is structured gives an impression that it is highly commercial. (2). From the job listing, one can see that experience in commercial codes are desirable. So, these are commercial codes users or vendors looking for engineers. (3). On every page, there is always a commercial code vendor's trade mark somewhere. (4). And I am sure that the webmaster also would like to have a good relationship with the commercial code vendors and sponsors. (5). It is then very difficult for the CFD readers not to think that it is a forum for the commercial codes users. (6). Since the technical questions normally requires some understanding of the user's problem, it will take time. And the best format is to have a technical support e-mail number listed here for each code vendor, so that a reader can forward his question to the vendor's support department. Whether it will be answered or not, it is totally depend on the user-vendor relationship because that is a business world. When you buy a product, there is always a number one can call when there is a need to do so. When I have problems with commercial code our people always says I have to call the vendor for answer because it is part of the purchase agreement. I normally demand the technical support through the phone right away until the question is answered. We don't have the patient for the e-mail. (7). It would be very hard to understand that a vendor would sell a high power tool ( or guns) to a student without proper training, support, or service. Can one give away free codes ( or guns) to student? Does the student user really understand the meaning of "AS IS" condition? (8). What you are really trying to say is you would like to gain back the non-commercial part from this web site which has too much commercial flavors already. This is the real world. When you step into it , you see the consequence. (9). My suggestion is it is a good idea to create one , a separate web site without any commercial flavor at all. (10). What I am saying is the reason why people come here to ask questions about commercial codes is because they think this web site is for the commercial codes. (otherwise, there will be a sign saying: private club, membership only.) For myself, I will support any CFD web site, commercial or non-commercial.

There is one point you missed about companies deciding to rely on commercial software over in-house codes. When someone is hired to write and maintain an in-house code, it takes a while to write a program from scratch. But if that person leaves for a better oppurtunity or retirement, etc. it can take a long time for a new person to become familiar enough with the code to maintain, debug and modify it effectively. This also depends on the complexity of the code, how well the program was documented and how well the program was structured and written. I inherited a program (not CFD, but still relevant) that was almost finished, but so poorly structured and documented that I put it aside and rewrote my own. I have since left that company and they had to get someone else to maintain and expand it. Companies also have to worry about employees who may purposely not document their code for the sake of job security. It's sad but people have done things like this before. And an employee could always take a copy of the code he wrote along with him to his next job. The new employer has a written code and a person who understands it, the old one will need a couple of months to find and hire a new programmer and several more months (at least) for that new person to become familiar enough with the code to start getting good results.

It think your prediction about companies using more CFD consultants is right on the money. It allows companies to only pay for these services when they need them and get quick results. It's also advantageous to the CFD analysts. They don't have to spend endless hours in pointless meetings and deal with the internal politics of large corporations.