[Offer] Electric Desk 1.04 (SCP, IMG) [5.25]

callmejack

Here is version 1.04 of the software suite "Electric Desk" (Version 1.11 is already present on winworldpc)

Dumps in SCP and IMG (obtained from SCP).

Backup disk has bad sectors, but fortunately System disk is intact. Floppies are unmodified.

https://mega.nz/file/XCJ0hTYI#usAQokfO5V-tDxgU-xumVNsCot6SDMveuU7wF7E8gDQ

SomeGuy

Nice find! It is always good to find additional versions.

I'll probably just omit the backup disk then. Out of curiosity, what happened to it? Was it shedding?

Hmmm, this version is copy protected.

ibmpc5150

Wow, thanks very much for offer Electric Desk.
There seems to be no problem except System backup backup.
It works well with copy protection on Track 9

callmejack

@SomeGuy it’s strange, I don’t know because I see no visible damage on the surface.

SomeGuy

That is curious. Looking at the SCP dump again, that entire side seems very noisy.

Typically when I see something like that my first thought is a fouled drive head. I'll stop the dump if I catch it before it is done (Kryoflux shows the graph on each track as it reads them), inspect the disk for any damage, and run a cleaning disk in the drive.

But I would guess you already did that.

If a couple of flux dumps don't show any change, then it very well may be how it was mastered. Perhaps the drive it was mastered in ages ago was out of alignment or dirty.

Because of the equipment I have set up, I might try reading such a disk on one of my machines with a real FDC and either a 360k or 1.2mb drive using Disk2Img or just copying files. In this case, that would not get the entire thing as it is copy protected, but if successful that could be merged back in to a flux image.

I have encountered a few times when a 360k disk read in my usual 360k drive shows all kinds of odd errors, but then then when simply read with a different drive it is magically much better.

And in those cases, there may not be any physical damage. In a couple of cases where originality was not important, I experimented, reformatting and media testing disk and they tested 100% flawless.

Of course, in this case you don't need to worry about it since the other disk is identical.

But I like to experiment and understand such issues because one day a problematic disk may be the only copy.

I just got through archiving a box full of 5.25" floppy disks. About half of them had visible gunk on the surface, several were in really bad condition, and there were at least a couple of those infamous Wabash brand disks. I managed to coax the data off of all of them, but next time it might not be as easy.

callmejack

Because it was a backup disks I didn’t bother spending much time tweaking with it, but you’re right, it’s an opportunity to learn something that could be useful for a future disk that is similarly damaged. I’ll give it a try.

callmejack

I used a qtip with alcohol to clean side 1. I didn't notice any dirt on the qtips. Then I tried disk2img (different floppy drive) but after 2 hours it was only 25% so I interrupted (see elect2.img). I also made a Transcopy dump (electric.tc) but I don't know how to convert it to .IMG. Is there a way using PCE tools ?

https://mega.nz/file/7LoEjJaD#Yx-RGNqPRWM7BmYLvjKBvDcPZMwSKzwH0nsm2RUNm4Y

ibmpc5150

Your back-up disk still has problem on side 1 (From 19-39)
But I think original disk is ok and no difference except track 40/0 (FM sector)

SomeGuy

I probably would not have left it running for two entire hours - usually if there is not significant progress after 10 or 15 minutes, that suggests other action is needed. But if the disk did not rip itself up, then that suggests the physical media is ok.

What type of drive did you read it in? a 360k or 1.2mb?

Transcopy images are a bit different from KF or SCP images - although it is a flux reader, it samples at a lower resolution and the resulting file is basically 1 to 1 to the MFM encoded data.

You can indeed use the PCE tools to extract an image. The difference is you skip the PFI step. For example:

pri ELECTRIC.TC -p decode mfm disk1.psi -f -v
psi disk1.psi disk1.img
psi disk1.psi -L > disk1.txt

The results here are not too different, starting at track 11 we see:

11 0 0 11 0 1 512 MFM POS=1273
11 0 1 11 0 2 512 MFM POS=6505
11 0 2 11 0 3 512 MFM POS=11737
11 0 3 11 0 4 512 MFM POS=16969
11 0 4 11 0 5 512 MFM POS=22201
11 0 5 11 0 6 512 MFM POS=27433
11 0 6 11 0 7 512 MFM POS=32665
11 0 7 11 0 8 512 MFM POS=37897
11 0 8 11 0 9 512 MFM POS=43129

11 1 0 11 1 1 512 MFM POS=1275
11 1 1 11 1 2 512 MFM POS=6507
11 1 2 11 1 3 512 MFM CRC-DATA POS=11739
11 1 3 11 1 4 512 MFM CRC-DATA POS=16971
11 1 4 11 1 5 512 MFM CRC-DATA POS=22203
11 1 5 11 1 6 512 MFM POS=27435
11 1 6 11 1 7 512 MFM POS=32667
11 1 7 11 1 8 512 MFM POS=37899
11 1 8 11 1 9 512 MFM POS=43131

12 0 0 12 0 1 512 MFM POS=1272
12 0 1 12 0 2 512 MFM POS=6504
12 0 2 12 0 3 512 MFM POS=11736
12 0 3 12 0 4 512 MFM POS=16968
12 0 4 12 0 5 512 MFM POS=22200
12 0 5 12 0 6 512 MFM POS=27432
12 0 6 12 0 7 512 MFM POS=32664
12 0 7 12 0 8 512 MFM POS=37896
12 0 8 12 0 9 512 MFM POS=43128

12 1 0 12 1 1 512 MFM POS=1274
12 1 1 12 1 2 512 MFM POS=6506
12 1 2 12 1 3 512 MFM CRC-DATA POS=11738
12 1 3 12 1 4 512 MFM POS=16970
12 1 4 12 1 5 512 MFM CRC-DATA POS=22202
12 1 5 12 1 6 512 MFM POS=27434
12 1 6 12 1 7 512 MFM CRC-DATA POS=32666
12 1 7 12 1 8 512 MFM POS=37898
12 1 8 12 1 9 512 MFM POS=43130

And then as expected it just goes down hill from here.

callmejack

@SomeGuy thanks very much for the TC convert commands. Would that work also for Macintosh disks using TCM ?

Concerning the dumps, they were both made using a 360K drive

SomeGuy

The last time I checked, the PCE tools did not support images written by TCM.

Normally when I encounter read errors like this on good physical media, I try a 1.2mb drive. Although generally a genuine 360k drive gets best results, there are certain kinds of issues that 1.2mb drive may have better results reading. (For example, if the disk was mastered on a 1.2mb/96TPI drive but the vendor neglected to fully degauss existing test signals or pre-formatting.)

Did you happen to use the exact same 360k floppy drive for these tests or different ones? What model of floppy drive(s) do you have? I happened to notice on a couple of your other 36ok dumps, the second head seemed slightly weaker/noisier than I would have expected even though the images decoded OK. It would be good to compare. One can never have too many floppy drives

callmejack

I used different drive, one for SCP and another one (on another machine) for disk2img and TC dumps. I tried with a HD drive with SCP and it worked !

https://mega.nz/file/bboGxKQC#n7lLFbshER23JMmLy5uIkdS1BWAQ1sxTJn1EGG8h2vs

Does it mean that my 360K drives are slightly misaligned ?
Thanks very much anyway for the tip, I would never thought about using a HD drive as they are supposed to be less reliable reading 360K floppies.

SomeGuy

I doubt it is alignment - it is very uncommon for a PC floppy drive to go out of alignment unless someone took it apart and moved the track zero sensor.

With that incredibly much much difference in readability, the disk probably was mastered in a 96TPI drive and not properly degaussed first. The magnetic track written on a 96TPI is thinner than that of a 48tpi drive, so if there is any magnetic flux next to the track it interferes when a 48tpi drive tries to read it. Like I said, I have run in to this a few times before. Yes, with most other problems, a genuine 360k drives does better.

This was a huge problem when IBM released the IBM AT computer and it's 1.2mb floppy drive. If you wanted to reliably share data with an XT machine with a 360k drive you had to add a genuine 360k drive to your IBM AT.

I do have a second theory based on the other dumps - perhaps the spring and arm holding the second head in place might losing their "springiness" or whatever the proper term is. I have two Mitsubishi M4851 360k drives here that have developed such a problem on their second (top) head.

So one more little experiment I might try here: Try this disk with disk2img again on one of your 360k drives - as it reaches the bad area, try gently pushing down on the top drive head. (If the disk stops spinning or it screeches, then you pressed down too hard).

If this makes little difference, then theory A (360k disk mastered in 1.2mb drive) is still most likely. If disk2img magically starts reading through more of the sectors, then theory B (problems with second heads) is possible.