Topic: Hello I started Fortran with Whatfor

Hello everyone,

Took me 4 years to find this site, from $4000 to $600 compilers, much too much. After many years a an engineer and surveyor, I have many OLD fortran codes developed and scrounged over the years. Alas, my time is short but I am glad to be here, in more ways than one. Many are on old 5.25 floppies (since uploaded to HD). A veritable library and repertoire of engineering and surveying source code written for my own purposes. I have a continuous multi-span variable MOI girder program that I developed in the 70's.

Many of these I intend to donate.

Back to you later,

Mike,
Cullman, AL

nothing is so firmly believed as that which is least understood: Francis Jeffery

2 (edited by JohnWasilewski 2014-02-16 11:45:35)

Re: Hello I started Fortran with Whatfor

You're right,Mike,  this is an excellent FORTRAN developers' forum.
Have you installed Simply Fortran yet?  Strongly recommended.  Best I've used.
Also, Jeff is superbly supportive.  Goes far beyond just supporting the Approximatrix products.
He has also helped many of us quite a bit with overcoming esoteric problems with Fortran.

It sounds as if you and I (and others, I'm sure) share a common interest.
What sorts of engineering & surveying software in particular are you interested in?
I for one would be very interested to hear more.
What software have you developed?
What other engineering software source codes do you have by others?
DO you want any help in getting it all compiled and running using Simply Fortran?

My interest is in stuff like,
- skeletal frame & grillage analysis.
- FE analysis.
- Slope stability.
- Retaining structures.
- Bearing pressures, Boussinesq etc.
- Applications of Bearing capacity formulae.
- Hetenyi BEF analysis.
- Lateral load distribution on bridge decks.
- Equivalent loads on structures from prestressed cables, with losses.
- Groundwater seepage pressures.
- Steel sheet pile analysis.
- Wood Armer transformations.
- RC beams and slabs.

Do you have the source code for 'STRESS'?
(I have it, but only in a book, and it is peppered with bits of v old non-portable assembler).

Do you have a copy of Purdue University's 'STABL'?
I have, but only as a printout.

Would you like a copy of Taylor & Zienkiewicz, 'FEAPpv' (executable + source)?
I have this.

Would you like the source code for Ed Willson's SAP-IV
(with further development by Bruce Maison to 1994)?
I have a copy.

Have you read this:
Hinton & Owen book cover
It has the complete source code for a superb 8-node isoparametric quad-element plane stress and plane strain program.
I managed to implement it in the 1980s, with capacity up to about 80 elements, using F77 and an overlay linker on an Apple ][ with 64 k memory and 128k floppies.  I thought that a pretty remarkable amount of computing performance to get out of such a small machine.  On a large problem, it would run for 40 minutes before producing perhaps 60 pages of FE analysis output.  This was possible because the program authors used Bruce Irons' 'Frontal equation solver', which needs only a tiny amount of storage for the stiffness equations.

I'm currently using DISLIN to port a substantial frame analysis program I wrote years ago, to M$Windows.  This uses skyline storage of the stiffness equations.  It's nearly ready.  See screenshots posted in this forum under, 'Fortran GUI screenshots library'.  Are you interested in this?  Happy to disclose the source code if you'd like to contribute to its development.
---
John

3 (edited by JohnWasilewski 2014-02-16 19:41:15)

Re: Hello I started Fortran with Whatfor

Ought the topic caption posibly to refer to, 'WatFor' rather than 'WhatFor'?
('Waterloo Fortran' - see very interesting article at https://cs.uwaterloo.ca/40th/Chronology/1967.shtml )
--
J.

4 (edited by jmh8743 2014-02-17 01:13:13)

Re: Hello I started Fortran with Whatfor

Yep! WhaFOR, oh dumb me.  Right John, guess that was F II.

Hope I can give your post its due.
1) My first nonprismatic variable plate girder design w/HS20-16 loading ran on a 360, 7 tape loads, overlay after overlay. Many years ago, that was. 180-220-180 28' RDWY.

2) Anyway, MS from Joe Appleton at UAB, BUT got to work under Anton Wegmueller for a number of years. He was a terrific mentor and soon FE came easy. Mostly used sap and STRUDL in those years. Wrote my own posttensioned conc box, continuous for him as a special problem course. He was frustrated because he could not get a PE license. I was very young. Returned to Germany and succumbed to cancer shortly. The school had a purveyor to MatLAB or whatever it is called, mostly matrix operations.

3) Rented STRUDL for many years in private practice later. As a result ALDOT got many FE solutions  from me when the others were supplying moment distribution (relaxation) for assumed constant MOI. Not sure now where that, originally MIT developed, is available. Still have manuals, somewhere. Made deck analysis easy.

4) subsurface flow modeling with contaminant diffusion. Didn't pay.

5) Very little SS analysis=> used STRUDL for my mats.

6) Once the company succumbed to greed, kinda diverged to applications in surveying and have been here since 93. Just me. Did a few bridges but if you don't have a 20 ma team state organization sneer.

7) Used Visual Basic 5 & 6 due to not having a F compiler. Had one at the company, but it evaporated. I was CE and we had had 90 employees. Partner went to Vietnam. Poof.

8) My latest venture is digital photogrammetry to help my little company.

9) http://basicphotogrammetry.blogspot.com/  Here the camera position is unknown for both camera frames and to refine the error budget, overdetermined. Thus the basic collinarity equations are develop and least sq methology applied.

Taking some C++ now. Tough learning curve.

Thanks for the offers, but I have very little structural work now. I can sure sympathize with the FE run time and your effort there. Well done. Guess we shoulda known each other years ago. I am still rounding up my stashes of code.

Mike

Yes I've got Z & T, 2 volumes BUT was unsuccessful in my translation to VB 6. Worked on it for about a year part time. Looks like a beaver has been chewing on Vol 1. Consequently bought RISA 3D when it was cheap.

nothing is so firmly believed as that which is least understood: Francis Jeffery

Re: Hello I started Fortran with Whatfor

Greetings everyone.  I have been searching for about two years for anyone who could help me find the listing for the STRESS program and/or the STRUDL program.

I am very interested in this, it is entirely for personal development (non-commercial.)   I see that Mr. John Wasilewski has the STRESS code listing.  Is there any possibility of providing this to me?


Thanks,
John

6 (edited by JohnWasilewski Yesterday 09:23:15)

Re: Hello I started Fortran with Whatfor

John,

I thought I had a 'STRESS' listing file but I've had a quick search, and I've not yet found it. 
What I have certainly found (about three of four years ago) was a book, which I purchased, containing the complete listing.  I also can't find that at the moment, though I hope it is lurking somewhere in my house.  I'll keep trying to find both of the above and let you know the result in a day or three.

I can tell you, however, that the STRESS listing I found in a book was less useful than I had hoped.  This is because that particular listing was not coded wholly in Fortran.  I am fairly sure that this was a peculiarity of that listing, because in 1980, when I had been working in Hong Kong, my firm bought a Pr!me 250 mini-computer (as they were called in those days).  It came with a fll operating system, including a F77 compiler, but no applications software.  Pr!me was good enough to give me a set of listings for STRESS, SAP-IV, and other stuff, like COGO.  I also found in various publications the STABL listing I mentioned in an earlier post, some unstructured but otherwise good source code by Bowles, as well as other useful source codes, and I was able to create a reasonably good portfolio of structural engineering, geotechnical engineering and land surveying executables in Pr!mos's equivalent to the Win Doze 'Program Files' directory, called, 'CMDNC0'.

You don't need all this information of course, and I am just meandering like this because it is how I remember for sure that the STRESS source code I had then was pure F66.

Anyway, back to the listing I found in a book.  It has about ten or 15 pages of a very old form of ASSEMBLER embedded in the F66 code.  As far as I can recall reading, this was to make the hardware for which it was written, either do stuff that F66 couldn't do, or  else just run hotspots in the code faster - I forget which.   I had planned to type it all in and get it to work but I never did so.  This was because the tedium of deciphering the ASSEMBLE and encoding it in modern Fortran pushed this project too far down my to-do list. 

As I say, I'll write again in a day or three.
In the meantime, let me know by phone or email (or in this forum) if you'd like a copy of the source code I've written myself for my program, 'InStruct'.  Some details are posed here in this forum:

     http://forums.approximatrix.com/viewtopic.php?id=316

---
John

John@Wasilewski.co.uk
0044 7900 891 107

7 (edited by JohnWasilewski Today 12:14:01)

Re: Hello I started Fortran with Whatfor

As promised, here's my follow-up on my search for the STRESS source code.

Yes, I do have it, but you won't like it!
It is a printed listing.  It can be found in the following reference book, a copy of which I was able to buy a few years ago, after searching on-line for it as an 'old book':

The book is,

STRESS: A Reference Manual
A Problem-Oriented Computer Language
for Structural Engineering

the Massachusetts Institute of Technology

Steven J. Fenves
Robert D. Logcher
Samuel P.Mauch

The M.I.T. Press
Massachusetts Institute of Technology#Cambridge, Massachusetts

Copyright 1965
Library of Congress Catalog Card Number: 65-13830

I have just checked, and I note that, if you type in this search string, Google finds the book:
STRESS: A 'Reference Manual' 'Problem-Orinented Computer Language' 'Structural Engineering' MIT Fenves Logcher Mauch

The preface explains how work on STRESS becan in Autumn 1962, under Illinois university and MIT visiting faculty member Prof. S.J.Fenves.  The introduction tells us that STRESS was implemented on an IBM 7094.  Chapter 1 says the Fortran code is in FORTRAN II.  The book explains the entire program organisation, with many flow-charts and a lot of explanatory text.  Remarkably, the team developed a dynamic storage scheme, using 1-D arrays and pointers.
The 'bad' part is that there is a LOT of assembler, and there's a warning in the narrative that reprogramming will be needde on any system that differes from the 7094 in word format or length, in the internal represeantation of alphanumeric characters, or in access to secondary storage (tapes and disks).  The assembler code is used for the dynamic memory process, for the equation solution, and for many other things.  It's all embedded within the Fortran listing.  The program listing fills pages 217 to 382, at 62 lines per page, which comes to over 10,000 lines.

By today's standards, the code is truly awful.  In saying this, I mean no disrespect to the authors, who, with the technnology they had, have done a magnificent job.  It is just that this was groundbreaking coding of complex structural software, at a time when portable code was both difficult, with the machines and compilers they had, and not yet a priority.  To illustrate what I mean,  pretty well EVERY PAGE has 20 to 30 lines of numbered statements, and not far off the same number of GO TO statements.  Over one third of the code is potential spaghetti.

If I thought I could compile this with Simply Fortran in less than many hundreds of hours of study, I'd consider typing it all in. 
So far, I have not overcome my fear of such a large mountain to climb.
---
John

8 (edited by JohnWasilewski Today 12:15:13)

Re: Hello I started Fortran with Whatfor

PS, If you manage to find a usable version of the source code, please wil you share it with me!
---
John