CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   DFD X CFD (http://www.cfd-online.com/Forums/main/908-dfd-x-cfd.html)

Dimitrius Caloghero June 11, 1999 15:58

DFD X CFD
 
Anybody know anything about Digital Fluid Dynamics (DFD)?

John C. Chien June 11, 1999 16:56

Re: DFD X CFD
 
(1). Since everything is becoming digital, for example, digital watches, digital displays, digital electronic circuits, digital computers, digital video disks, digital compact disks, digital voice recorders, digital thermometers, digital signal processings, etc,....etc,..., a better name for CFD is going to be Digital Fluid Dynamics, that is DFD. (2). It is universal, because you can read it from left to right, or you can read it from right to left ( it will unify the east and the west with no mistake). (3). The other reasons are: when we move into the next century, we move forward, from "C" to "D". That will avoid the Y2K problem, I think. And people all over the world will recognize "D" as Digital right away. ( Some HR people have asked me a few times about "what is CFD?" in several job fairs. Normally, they lowered their voice first. Since then I have been hiding CFD deep in my resume. I got the feeling that CFD is the first job to cut in modern days. ) (4). I like Digital Fluid Dynamics, so I invented it. See, DFD is easy to write. It is also symmetric. (5). "Computational" is labor intensive, "Digital" is light and intelligent. (6). It will save time to use "Digital" vs "Computational". You can double the productivity right away. So, I think, it is a very good invention. (7). Digital Fluid Dynamics(DFD) saves time, saves money, improves productivity. You will be happier and live longer, if you use DFD everyday. (Computational Fluid Dynamics makes you feel like a slave of computer.) And I am sure that more people will visit this place if we start using "Digital Fluid Dynamics" (copyright reserved,1999,John C. Chien). (8). Don't try to look it up in the dictionary, it is not there yet. If you don't like it, you can always say that " It is John Chien's Digital Fluid Dynamics(DFD) ". Have a nice weekend.

Enrico Nobile June 13, 1999 07:40

Re: DFD X CFD
 
From the subject of your message, I'm guessing that you refer to the patented DIGITAL PHYSICS technology of Exa Corporation, 450 Bedford Street, Lexington, MA 02420 - 781/676-8500, FAX-781/676-8599, Email: info@exa.com, http://www.exa.com

They used this term for their enhanced particle-based method, used in their PowerFlow software, in order to distinguish it from the standard continuum (PDE-based) numerical approach, i.e. FEM, FV etc.

As far as I know, it seems they were the first to use the term "Digital Fluid Dynamics (DFD)" (see their Web page at http://www.exa.com/corporateprof.html).

As a final note, so far we did not have a chance to test/use PowerFlow (the cost for an Academic Licence here in Italy was very high), and therefore I cannot comment about accuracy, performances etc.

Regards,


John C. Chien June 13, 1999 14:31

Re: DFD X CFD
 
(1). I am not aware of the "Digital Physics" you just mentioned. I am not talking about the physics, I am talking about the fluid dynamics. (2).I am not familiar with the enhensed particle-based method and the related software PowerFlow. Would you be kind enough to tell more about it? (3). As I have said many many times here that everyone has his own definition of CFD. So, look like that there are at least two definitions of DFD. In my posting, I have defined DFD clearly, but I don't know how the other one is defined. (4). I am talking about CFD in general, not a particular particle method, or a specific software package. I think it is fine to talk about the price of a software. And it is important for a user to shop around. (5). I think it is all right to use DFD in anyway you like. Especially, if you are tired of CFD (Chinese Fast-food Delivery), you can always join DFD (Danish Food Department). (6). Thank you very much for your creative thinking.

John C. Chien June 13, 1999 19:51

Re: DFD X CFD
 
(1). I have just visited the home page of EXA corp. The company provides conventional "Computational Fluid Dynamics" service. The trade mark is related to "DIGITAL FLUID", which is particle-based analysis tools. (2) I did not find any technical information about this "DIGITAL FLUID" software package. Since in the description, the word "simulation" was used and it also emphasized the non-conventional CFD approach, it is probably hard to address the accuracy issue.(3).As I said before, large companies normally have tons of (several hundreds) design software on their systems. Therefore, the exact location of the software is not important. The important thing is whether a company can survive with tons of software sitting on their systems, especially without adequate manpower to use it. (4). It is a burden to a company to put a software on the system. Unless the company can show that the software is actually helping the sale and the quality of the products.(5). In old days, the government labs can not say that they want only one brand of computers in the office. As a result, it was very hard to network and it was very hard to maintain the software. Now the companies have learned the lesson, and most of the companies are trying to use one single brand of computer to eliminate the routine headache. (6). So, the goal of CFD is to provide better solution to fluid dynamics problems , rather than to create lines-after-lines of codes to occupy the computer system and to enslave the ordinary engineers.

Carlos Felipe Favaretto June 14, 1999 09:49

Re: DFD X CFD
 
Dear Mr. Chien,

For the sake of science, could you please keep the discussion in the technical level? All we have been asking you is how this new and fantastic DFD works. It would be a kind of kamikase action to use a code that is doing what-so-ever! I excepted more from you than just a bunch of silly jokes.


John C. Chien June 14, 1999 10:35

Re: DFD X CFD
 
(1). Please read my other posting. I did visit the home site mentioned earlier. But I could not find any technical information. (2). I think, you should send the e-mail directly to the vendor to get the technical information. (3). I would say, in many cases, the code users are being used to debug the code for the vendor. (4). All of the commercial software are sold on "as-is" basis. Read the fine print and users beware. (5). I have used commercial software, most of the time, I was getting the wrong answers. But I am still using the commercial codes. (6). There are many reasons for this. One is to keep the information up-to-date. And sometimes, you can still use part of the results if you are really good. (7). CFD is a place where you have to get involved. You learn by making mistakes. ( Remember that a really good CFD software will never appear on the market. It is always under tight control.)

Anders Jönson June 14, 1999 11:15

Re: DFD X CFD
 
Hi! I am surprised that you say that you could not find any technical information on Exas homepage! Check this adress!

http://www.exa.com/powerflow/papersdir.html

Lots of technical info there!

Anders

John C. Chien June 14, 1999 15:23

Re: DFD X CFD
 
(1). Thank you very much, I'll look into this address.

John C. Chien June 20, 1999 10:14

Re: DFD X CFD
 
(1). I have visited the powerflow site, but I could not find the technical information I was looking for. (2). I must say that the code is apparently doing somekind of simulation, because there were pictures presented. (3). Beyond this picture stage, I would like to see the results of benchmark testing. And since most of these benchmark test cases are relatively simple, it should be carried out by the vendor if he is seriously considering the code as a design tool. (4). As a simulation tool, and as an alternative method to simulate the Navier-Stokes equations, it should be all right as long as the results are repeatable and the solution is realistic, that is the trend of the solution must be correct to be useful to the users. (5). So, I strongly suggest that the vendor carry out the benchmark tests to verify the code first. ( Whether the results should be published or not is not an issue. Because sooner or later, the user will get to that point.) (5). Still thank you for the information about this particle simulation code.

Vincent Johnston June 30, 1999 11:54

Re: DFD X CFD
 
It's good to see that PowerFLOW is generating discussion on a public venue. For anyone who is seriously interested in our company or our code, feel free to contact me directly at vince@exa.com.

I also recommend a review of our website for further information regarding our company, products, and patented technology.



All times are GMT -4. The time now is 12:31.