Lately, I’ve been on a FORTRAN rediscovery. The first programming class I took in college was FORTRAN (but I already knew BASIC). My first real programming job was in FORTRAN (processing data from automotive crash tests). My career went a different way for a bit, but I came back to an engineering area and did a little more FORTRAN on HP RTE-A systems.
Now I am doing a little FORTRAN-80 (like FORTRAN IV) on my TRS-80 systems and MS-FORTRAN (like FORTRAN 77) on my MS-DOS machines.
Punched cards did make for simple input.
I picked up a used non standard dos computer cheap, that came with a few floppies. One was a FORTRAN compler and libaries. Sadly I had no compete boot floppy that came with the sytem. No way to format new floppies.
Today we have the internet so finding old media is better than it was. I still can’t belive how slow the first IBM PC computers were.
How is data bigger than 64Kb handled and is floating point 32 or 64 bits?
Do you have complex numbers? Fortran was quite portable all things considered.
Ben.
It does take a bit to compile and link (mostly link) a program. It’s certainly not press a button and run your test.
In MS-DOS, things like arrays cannot be more than 64K. I’m not sure about the TRS-80, but that one only has 128K, so having a data block > 64K really isn’t workable. FORTRAN does allow you to create scratch files. So you could get around that by using disk “memory”.
I’d have to check the manual, but I think you can do INTEGER8 and REAL8.
I haven’t played with them yet, but the MS-FORTRAN, at least, has support for COMPLEX numbers.
Oddly, Microsoft FORTRAN 3.13 for MS-DOS didn’t have a COMPLEX type. Its floating point was either 32-bit (REAL*4) or 64-bit (REAL*8 or DOUBLE PRECISION). It can’t handle objects individually larger than 64Kb.
According to the Microsoft FORTRAN 3.20 manual that I have:
“The COMPLEX 8 type has been added, as well as specific and generic type conversion intrinsics that support it as defined by the full FORTRAN 77 language. COMPLEX 16 has also been added as a Microsoft FORTRAN extension.”
Also, surprisingly, it says “Arrays and common blocks longer than 64K are now supported.”
The table showing the memory requirements of all types shows up to LOGICAL 4, INTEGER 4, REAL 8, DOUBLE PRECISION (note shows that this is the same as REAL 8) and COMPLEX 16.
I find the bookshelves also very interesting. Here are 2 of mine, comparing to the other’s, my books are of a much less variety.
Everyone has the Leventhal Z80 book but me. 
Well… I don’t have a copy either, but then I don’t have any Z80 books at all…
-Gordon
I couldn’t see my Leventhal then found it on an unexpected shelf… but it’s the 68000 one!
Edit: bonus shelfie of recently-accessed and unhomed acquisitions
Ron, where did you do your crash test programming? I had a summer student job (also with FORTRAN) at Cornell Aeronautical Labs (now Calspan) back in the '60s, and I know they had a crash test division.
At the GM Proving Grounds in Milford, MI. While they didn’t let me actually crash any cars, I did get to sit in on a fair share of crash tests.
Hoping to see more shelfs where are the remaining OG’S
Ah I should take a picture of my mini-shelf of computer books. For a short while I tried collecting the really old ones but they are/were hard to find. I think the oldest I have is from 1946
What an excellent shelfie! Not one, but two copies of Hollingdale and Tootill - one of the very important texts in my personal interest and development. I devoured it! I found some online copies at reputable places:
High Speed Computing Devices 1950 by Engineering Research Associates
Mathematics and Computers 1957 by Stibitz and Larrivee
Basics of Digital Computers 1958 by Murphy
Digital Computer Primer 1959 by McCormick
Electronic Computers: Principles and Applications 1960 by Ivall
Electronic Computers 1970 by Hollingdale and Tootill
