This message is posted with the intention of letting know the users of FIDAP 8.52 that there are some errors in the manual. In sections 7.1 of the FIPREP user's manual, we wncounter this error.

"USRBCF must return the specified flux value VAL for the IDF degree of freedom of node NODE at time TIME. The possible values for IDF are given by the following pointers: KDFU, KDFV, KDFW, KDFD, KDFT, KDFF, KDFK, KDFE, KDFS + I. "

But, I was informed by the FIDAP personnel that there is nothing called "KDFT" or "KDFS + I". There are only arguements namely "KDT" and "KDS + I".

Simlarly another error - or rather - lack of updation is in the "ID" array used. ITs equivalent is the "NUMEQA" in FIDAP 8.52

(1). This is highly welcome. (2). My son told me the other day that when he was doing the beta test of a new version of the operating system program, there were already several thousands bugs identified.(I was amazed. ) (3). In some codes, even the tutorials are using the old version GUI (which you don't have). (4). You have to be realistic, because the tire problem of supersonic concord has been identified before, other problem such as the wiring issues, suv tires, etc... all have some indications of the bad symptom before. (5). If no one provides the correct answer and warranty, then only the professional who has the right answer can safely use the codes. The codes used to supplement his knowledge and experience only. (6). So, I must urge the users of the commercial codes to gradually take this point of view. (7). For me, the commercial codes are sets of library routines, which have been tested for only limited number of sample cases.

Thx for your comments though the issues like concord appeal to me as superfluous I have no experience in the convestion problems and 'll be glad to receive info from knowledgeable users like you.

IF you use FIDAP, can I send you an e-mail ? thx

(1). I don't use FIDAP. But I have seen the FIDAP demo many years ago when the company was still an independent company. (2). I use some commercial codes, but I got paid to use the codes. (3). If the original developer of the code is still there, and the programmers are still there, then it is a good idea to contact them to get the first hand suggestions. When the code is no longer supported by the original developer, it is a dead body. You can study it, cut it into pieces, but still, it is not alive.

I think John's comments are applicable to all cfd codes, not just commercial ones - (and in fact all engineering tools . . .for that matter). All tools/cfd codes, have a range of applicability and have been validated for a small subset of real-world conditions. Its up to the engineer/user to ensure the tool is suitable for the task - that's our profession.

The trades-off between commercial / in house codes are that for i) inhouse codes you have a greater ability to check and modify the actual source code with in general high maintenance/development costs vs ii) commercial codes where you have less/no ability to check the source code and limited flexibility to modify it, with in general lower maintenance/development costs and perhaps higher productivity (depends on what you call productivity)

Greg