Phase Identifier

[JohnW2]

... but can't help butting in with recollections of writing punched cards (1960s for IBM). Awful; one mistake and two days later the whole lot came back without any explanation. This could happen several weeks running for the same analysis.

That's extremely familiar. Although somewhat tangential to the courses I was doing, it was (in late 60s) deemed to be a good idea to 'teach us some computing', so we had a 'practical class' for a couple of hours once per week. We would slave away trying to write FORTRAN code on 'coding forms', which were then handed to people who punched our code onto cards, and then shipped them a mile or so down the road to 'the university computer'. Come next week's class, our coding forms would come back, complete with the corresponding compiler output (rarely got beyond that stage!) - which almost invariably said something along the lines of "Syntax Error in Line 7". By that process, it took most of us a whole term to write and debug a few dozen lines of code which would actually run and do roughly what we had intended!

Kind Regards, John

 

[ban-all-sheds]

I once had the joy of using a PL/1 compiler which did automatic syntax correction.

I remember one time it made 7 passes through one line of code, fixing an "error" each time until it was syntactically correct.

By then it was semantic b****cks, but hey, it compiled cleanly.


This must stop now.

 

[SimonH2]

Could be that it's not the drive/disks but what the OS puts on them. MS are infamous for "doing their own thing" and there's absolutely no guarantee whatsoever that what their current OSs write on a floppy is "compatible" with something expecting what an older OS put on it.

Might be worth firing up a Linux machine (or virtual machine). You'd have more control over what it does on the disk, and if it is a software issue it might help show it one way or the other.

 

[bernardgreen]

Could be that it's not the drive/disks but what the OS puts on them. MS are infamous for "doing their own thing"

So true ! An aquaintance who provides data recovery service uses DOS as the operating system for recovering data from damaged drives. He has found that very often a file that MS will not or cannot read can be recovered intact using DOS and simple disk read function.

 

[ban-all-sheds]

Windoze, particularly NT , used to be very aggressive about deciding that disks it did not recognise should have an MBR with an NT signature written to them. This was rather unfortunate if they were actually full of other stuff. Great care was needed with dual-boot systems, LUN masking on SAN disks etc to prevent Windoze seeing stuff it shouldn't.

 

[JohnW2]

So true ! ... He has found that very often a file that MS will not or cannot read can be recovered intact using DOS and simple disk read function.

I'm a little confused by the distinction you appear to be making between "MS" and "DOS". What do you mean by the latter - something other than "MS-DOS"? Are you simply talking about media than cannot be read directly (maybe not even recognised) by Windows?

The operating system run by a computer on which one attempts to recover data from old-format/non-standard format/whatever media is essentially not all that important, but the utilities one uses for the recovery may well have to bypass/ignore all of the OS's file handling routines and use direct BIOS calls to examine the media 'sector-by-sector'. In the past, I've spent many (hundreds) of 'happy' hours playing that game!

Kind Regards, John

 

[scousespark]

I worked in IT (mid 90s) for a firm which used Honwywell DSP minis for printing. The print site was in Rochdale and our development site in Liverpool.

We had no way of deploying code, other than manully transporting it on disks (and the same applied to testing). The disk packs were huge and weighed a ton. We always took 2, as there had been an instance when they went out and the disk was corrupt.

The lead analyst didn't drive and was near retirement and I got all the development work. I made a fortune in expenses driving between the site (sometimes 3 times a week) and we were paid expenses in cash.

In my first performance review I asked why I got all the print changes. Apparently, the first time I went out I just grabbed the 2 cases and lugged them up 6 flights of stairs. I took this job after resigning my commision with the Royal marines, so it wasn't a real effort. The analyst was made up and so my fate was therefore sealed. I couldn't do it now BTW.

 

[ban-all-sheds]

We had no way of deploying code, other than manully transporting it on disks.

An early version of what became known as Hondanet, which for a few years had a competitively high bandwidth for file transfers. But its latency was poor.

 

[Spark123]

Even before this?

 

[ban-all-sheds]

No, but a guy on a bike with tapes in his panniers could shift data over reasonable distances faster than that.

 

[Monkeh]

We had no way of deploying code, other than manully transporting it on disks.

An early version of what became known as Hondanet, which for a few years had a competitively high bandwidth for file transfers. But its latency was poor.

Volvonet if you had a lot to move.

 

[bernardgreen]

I'm a little confused by the distinction you appear to be making between "MS" and "DOS". What do you mean by the latter - something other than "MS-DOS"?

DOS before Window, both MS DOS and the Original IBM DOS.

In these very basic operating systems it is easy to directly access the hardware registers of the various peripherals and, as you say, bypass the file handling functions of the OS tp read sector by sector and moving the read head a small distance off the centre of track to read any remnants of a deleted but not erased track.

 

[Spark123]

I thought PC DOS (IBM) was just a rebranded version of MS DOS?
I can remember using MS DOS along with GEM desktop using the 5¼" floppies many moons ago, think it was an amstrad 1640

 

[JohnW2]

I thought PC DOS (IBM) was just a rebranded version of MS DOS?

So did I - hence my question/point.

Kind Regards, John

 

[bernardgreen]

From memory,

There was a disc operating system for the Intel 8080 processor which was not called DOS but CPM ( cannot recall what CPM stood for ). The arrival of the new 8086 processor ( as used in IBM's new Personal Computors ) required a new operating system. The CPM software was the basis of 86-DOS but modified to suit the new architectre of the 8086. MicroSoft bought 86-DOS, renamed it MS-DOS and developed it as a system that would run on both true IBM PCs and also the clones that were coming onto the market.