CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   WOW! (

clifford bradford March 24, 2001 23:42

I've not been to the Forum in a while and just reading for a few minutes i've been stunned by the number of people asking questions which are even below basic. Many of these posters don't even seem to have good grasp of Fluid mechanics much less CFD. I think it should be mandatory that everybody intersted in CFD read John Anderson's book or some other really introductory book. some of it though is just applying youe melon (ie brains) or just going to your nearest university library and reading a few books. I sorely miss the days when all they talked about on this forum was differencing schemes and LES (even those things bored me to death) now I can find is people wanting to know what code/grid generator etc to buy; what's the difference between finite difference and finite volume or what's the boundary condition on a solid surface. the intellectual level of the forum is falling fast and as a result many persons who used to post deep questions are not doing so and I'm not getting much out of it. Perhaps we need a newbie forum.

John C. Chien March 25, 2001 00:28

Re: WOW!
(1). It is very interesting. I think, you have changed a lot. (2). Have you written a 3-D Navier-Stokes code using finite-difference method with general coordinate transformation for turbulent separated flow? Have you obtained mesh independent solution? (3). Have you written a 3-D Navier-Stokes code using finite-volume method with unstructured mesh for multi-stage turbomachinery applications? Have you obtained the converged solution? (4). What's the boundary condition on the solid surface? None-slip condition? Is that all? Were you able to treat the turbulent boundary layer flow correctly with a simple non-slip condition? (5). " I'm not getting much out of it?" That is because you have been busy running somebody's code and are becoming a robot. It's not a joke. My experience is: running codes alone will not keep the company alive. (6). I think, there is no big change in the questions. Unless you have written 3-D FD or FV codes, you still don't know much about the definition of FE and FV. Unless you have obtained mesh independent solutions in 3-D, you have not accomplished anything yet.

George Bergantz March 25, 2001 02:22

Please- a middle way gentleman
John- can you perhaps make your point without being so severe? You seem to be rather cranky lately (your recent comment in the PHOENICS forum on being "smart enough" was a doozy!).

You indeed have a point, and I share your concern regarding Clifford's remarks perhaps being hasty.

Clifford- your interest and concern about the forum is good and it's odd you bring up the example of "finite difference vs. finite volume" because just the other day a student asked me this and I directed them to search this forum to see many and diverse opinions. I see a place in the forum for education, and most do a pretty good job of ignoring the really clueless questions. I'm proud of the way this forum welcomes newbies, and haven't felt any lack of focus or quality, although that may not be your experience. I have learned more than I care to admit frankly. For example the discussions on the nature of turbulence in CFD has been especially illuminating, even if old hat for others.

So jump in with some gritty questions and comments. You will find the usual suspects lurking around one hopes.

John C. Chien March 25, 2001 03:51

Re: WOW!
(1). Don't worry, he is an old friend.

Astrid March 25, 2001 07:20

Re: WOW!
Dear John,

Don't make yourself so unpopular. You start to sound like a greybeard in a dusty attic room fighting the CFD-world.

How can I take your comment serious when you are making jokes to an old friend? How should we (the other members on CFD-online) know he's an old friend who doesn't take your comment too seriously?

Regs, Astrid

Peter Petropoulos March 25, 2001 10:23

Re: WOW!
>the intellectual level of the forum is falling fast
:and as a result many persons who used to post deep
:questions are not doing so and I'm not getting much
:out of it.

Welcome to the "information superhighway"...Your post is dead on. And this has been happening across the forums and the newsgroups. I sent my first e-mail in 1982 and have been on-line since then through academic machines; I used to follow some of the sci.* newsgroups, and welcomed the comp-fluid-dyn one as a welcome addition to sci.math.num-analysis...but the noise level has been so high since about '94 that I no longer follow any of the newsgroups. My other interest (computational electromagnetics) had a newsgroup that got swamped in time dt by posts refuting Einstein.

A solution would be to return to moderated discussions. Now where are those AI programs promised to us by the comp scientists ?


John C. Chien March 25, 2001 14:45

Re: WOW!
(1). CFD is not a popular subject. So, there is not need to pretend that posting messages here will become more un-polular or even more popular. (2). The reason why people are coming here is because they still have difficult CFD problems to solve. If a large and old company is still having difficult problems to solve, it is bad. The end result is : they will die. (3). To make changes in a large and old company is a very difficult task, not to mention that it must deal with CFD problem after re-structuring. (4). Even the Internet companies are disappearing, so why do you think that cfd forum can stay here with un-popular messages. (5). If we charge 20K dollars for each question posted, we are still following the cfd vendor's practice. It's not un-popular suggestion at all. (6). For free questions and answers in cfd forum, any un-popular messages are good messages. The more the better. Do you know why?

George Bergantz March 25, 2001 15:34

150 max number
This may not be relevant but I'll share it anyway:

"150 is the maximum number of people who can have a genuinely communicative relationship....that number comes up again and again in the anthropological literature...."

From "The Tipping Point: How Little Things Can Make a Big Difference" by Malcolm Gladwell.

But having said that I'm not clear on what the problem HERE really is. Is it having to wade through gratitious, non-specific postings? Flame wars? SPAM? Is it that the novelty soon wears off for the more informed participants, leaving only the blow-hards?

I too have wandered the canival of newsgroups. This is about as good as it gets IMHO.

Astrid March 25, 2001 16:05

Re: WOW!
- By saying that you make yourself unpopular, I mean that you could be more constructive.

- There is a shake-out in internet companies, the internet-economy bursts like a bubble. Some CFD-companies will disappear too, no question about that. I think the carefull companies will survive. Where are they...... who knows....?

- Un-popular meassages are good messages? Well, you mean there is work to do. So, speed up a little, John, and keep up the good work.


John C. Chien March 25, 2001 20:48

Re: WOW!
(1). Sorry, I can't even get on the Internet, because my free Internet time is already used up. (2). Say whatever you like, I can only ride the bicycle in the park and drink ice tea at K-mart. There is no way I will be able to read your message.(3). I have to use my other free Internet time for my e-mail. Bye.

Jonas Larsson March 27, 2001 02:33

Re: WOW!
I'm not sure I agree, the level of some questions is quite low, but this is not a recent development - it has been like this for a long time. You can still find several active "high-quality" threads.

Anyway, you address an important issue. I think that the only way to improve the cituation would be to open a moderated "research level" forum. A "newbie" forum is no solution - the "newbies" want help from "non-newbies", that is why they post. However, moderating a forum is a lot of work - the main forum today gets about 70 posts a week. I can't do this myself and still run the rest of the site. The software used for the forums can handle external moderators though, so if someone would like to be moderator for a "research forum" I'd be happy to set things up.

Sid March 27, 2001 03:52

Re: WOW!
Just ignore such "low-level" questions. Doesn't that work? The thing is, there are "experts" who kindly give detailed answers to such questions, which then often generates additional questions.

Now, look at John! He gives a nice answer to such a low-level question: the answer is in a CFD book. Then, no one else posts additional messages any more.

Bart Prast March 27, 2001 05:55

Re: WOW!
My suggestion is to open a forum for people with an minimum intelect of say 110? (sarcasm) I liked the fact that this forum was for everybody. A lot of people can ask every level of question. It is up to the reader I would say which type of question he would like to give an answer to. I do not like the snobish attitude some people have. Please have an open mind. Unless there is some kind of incrowd of CFD people that I'm not aware of.

John C. Chien March 27, 2001 06:18

Re: WOW!
(1). You can't ask people to donate answers which worth 20k. But I am sure that there are many people reading between lines to get some kind of feeling, even from very simple questions and answers. That's the reality. (2). I am still here, because it was my son's suggestion for me to visit the cfd-online. And I don't want him to know that cfd-online is no longer there after a couple of years. (3). There is no incentive to give out research information, or even a hint of what I am really thinking. (4). For me, cfd-forum is the place I exercise my brain, and provide some very simple guide to new comers. After that, if they need more, they have to earn it.

John C. Chien March 27, 2001 06:50

Re: WOW!
(1). The answer to that is very simple. (2). Ask each code vendor to talk about their own code here, the code structure, the subroutines, the listings, the models, etc.. (3). If you are paying fees to use their products, and presenting questions here, then with the knowledge about the code, the users would be very happy to read this forum. If they are doing research, selling codes, and don't want to share with the users about the code, then it is really funny to ask others to donate high quality answers. (4). The high quality answers should be right in the codes you are using, right? Or , you would rather spend time to learn something else. The gold mine is right inside the codes you are using. So, ask them to post the code listing in the forum.

ken elms March 27, 2001 08:21

Re: WOW!
The plain and simple truth centres on how well the forum wishes to seriously address its relation to a commercial criteria ,an academic coverage of the hands on approach,or indeed its contribution to education.

As far as I`m CONCERNED to diversify even further into specialist topic areas will no doubt prove helpful to some but bore others.

Far too often the expectations from CFD -And I INCLUDE my own area of interest-Rotating pump impellers, has been too high.

I would like honesty of experience [all codes] to count ,but not wish to spoil genuine attempts by companies who do strive to get it right.

Cfd must be sold for what it is. Therefore,along the way there will be bias one way or the other. Can you image the investment into CFD such as football teams etc. The forum does not have that level of backing.Where are the supporting product manufacturers also to be found on the forum to increase awareness.

Will computers eventually put CFD into a domain of control in the design process where Joe Engineer fades into the distant past.

Keep the CFD ball moving and not keep kicking it in the same places.

clifford bradford April 1, 2001 21:17

Re: WOW!
Hello again. Once again I've been off the forum. this time it was because my "high speed internet connection" wouldn't connect. but enough of that.

I've liked the comments on my previous post. I think I posted it in the evening after a long day at work. I think that had something to do with the tone of the post. As some of you might I've been posting here for awhile and as one of the previous messages in the post said I feel this is one of the better technical discussion groups on the web.

Now that I think of it a newbie group is not a good idea. If it hadn't been for reading and barely understanding the more advanced posts when I was a newbie I'd have never even known what I didn't know, needed to know, or perhaps didn't want to know. However, for the novice CFDer in any area it is essential to get some background and be willing to go to your library, ask a professor or knowledgeable collegue and be discerning. I think this outlook is missing from some of the novice CFDers who are posting. As I pointed out before you need to know physically what the no slip condition is before you can ask about implementing it. It makes little sense to have a limited knowledge of Fluid Dynamics and expect to converse adequately with, say, John about CFD. He might come off sounding like a grumpy old man in his replies otherwise.

I will say though that I still think that the intellectual level of the group discussion is down (Ironic, coming from the "practical" guy). I think that perhaps we should be willing to simply start threads not as questions but just putting stuff out for discussion, such as "how successful has residual averaging been for you?". Also I think when we post questions (and replies) we should be to the point, answer or ask the question as clearly as you can and THEN jump on the soapbox. After all we all want to get on the soapbox :)

As usual I'd like to say that Jonas is my hero, although I'd like to say that if he's not graduated yet he needs to do so and get a real job! (laugh)

clifford bradford April 1, 2001 21:19

Re: WOW!
I think this is a great idea John. I haven't perused any of the other forums recently but I think that perhaps they should be more active in the discussion groups on the website.

John C. Chien April 3, 2001 13:22

(1). There is no such thing as a stupid question. This is especially true in computer related field. (2). CFD forum is a place like a large flea market, where questions and answers are displayed. (3). These questions and answers sometimes do not seem to be related or make sense at all. But, it is the reader's job to find and match the question and answer together. (4). This is the reason why you do "Internet Search" or "Internet Surfing". (5). Internet is only a "connection", it is not "MIT" or "Harvard". Posting a message here will not change your background at all. But that connection could be very important when someone or company needs your service in the future.(6). By asking a stupid question, it does not make a person stupid at all. (7). I can tell you that sometimes ago, a senior engineer working at a century old engineering company asked me a question,"what is CFD?" (the question could be: he is not using cfd, he does not know what cfd is, he think cfd is useless, he think cfd can only create more problems, and delay schedule, etc...)

All times are GMT -4. The time now is 13:23.