Problems writing back to floppy disk

All questions regarding the dumping of media go here.
robbo007
Posts: 87
Joined: Wed May 23, 2012 2:56 pm
Location: Spain

Problems writing back to floppy disk

Post by robbo007 »

Hi guys,
I'm having problems writing back to floppy disk dumps Ive made with KF. The original disks I have are 360kb MSDOS and they are copy protected with Softguard SuperLok v3.00 KIT. When trying to run the program from floppy as its meant to I get a Key disk error which is the copy protection failing.

Here is the dump. Can anyone assist with this type of Copy Protection? Do I need to do anything different when dumping besides using 40 tracks? Or something different when writing back to floppy?

Im using preservation raw to write back to floppy.

http://www.mediafire.com/file/49bs5eou0 ... 9.zip/file

brightcaster
Posts: 292
Joined: Fri Nov 08, 2013 10:48 pm

Re: Problems writing back to floppy disk

Post by brightcaster »

Your dump looks a bit strange as the tracks are counting up to track 82 with all the odd track numbers are missing giving a total of 41 tracks. So I guess you dumped and wrote with a 80 track drive which should work as long as you read again with a 80 track drive on you test machine.

Did you use a new or degaused DD disk? Track 5 on side 0 looks very tricky with different sector sizes and on track 17 there is some writing in sector and index gap which might be difficult with drives not running at the correct speed...

David
Last edited by brightcaster on Wed May 20, 2020 9:49 pm, edited 1 time in total.

brightcaster
Posts: 292
Joined: Fri Nov 08, 2013 10:48 pm

Re: Problems writing back to floppy disk

Post by brightcaster »

I just found the thread on the other forum at https://forum.winworldpc.com/discussion ... p-kryoflux where you discussed the matter with the masters and experts of the art. I think I should be quiet now, as I have nothing essential to add to that ;-)

David

SomeGuy
Posts: 272
Joined: Wed Feb 18, 2015 8:18 pm

Re: Problems writing back to floppy disk

Post by SomeGuy »

I've been busy, but since I saw the post here I finally took a quick look at them.

The "structure" of the protection on these disks looks the same as MusicPrinter Plus 1.1.

I tried the commands listed on the winworld forum to try converting to a PSI file for PCE. Sideman D50 and Sequencer MKIII ran, but Sequencer Plus MKI and Sequencer MKII complain about missing keydisk.

Writing one of these disks and reading it back looks like an identical copy, it didn't run.

So I still don't know what the real issue is.

I don't have any real way to tell for sure where exactly the protection thinks it is failing. I'll just reiterate a couple of hunched. I see some odd stuff at the end of track 17 that seems to run all the way to the end of the track. If it is looking at that, it could be trying to read across the "splice" area at the index, and getting a different result. It could also be some timing sensitive operation that I can't see when looking at MFM data. Perhaps someone here has a better idea of how to check for that. I've been told that it is not weakbits, and I don't see anything that differs per rotation inside any of the decodable sectors.
brightcaster wrote:
Wed May 20, 2020 9:19 pm
Your dump looks a bit strange as the tracks are counting up to track 82 with all the odd track numbers are missing giving a total of 41 tracks. So I guess you dumped and wrote with a 80 track drive which should work as long as you read again with a 80 track drive on you test machine.
In the discussion about MusicPrinter Plus 1.1, I had better luck writing an image that was dumped from a 1.2mb drive. Although nothing between those and the 360k dumps looked different.

SomeGuy
Posts: 272
Joined: Wed Feb 18, 2015 8:18 pm

Re: Problems writing back to floppy disk

Post by SomeGuy »

Poked around in PCE and confirmed that is at least part of what is going on. It does a diagnostic read on the first sector of track 17 head 0, which reads the entire track plus part of a second revolution. Then it looks for some pattern just before the first sector on the second revolution, and if it does not like what it sees, it fails the protection. (Was mainly testing with the Sideman D50 disk).

So that means any break at the index is likely to break the protection. The normal operation of duplicating with a kryoflux will leave a slight break there, although depending on the drive writing, and the drive reading a copy might sometimes be good enough to work.

PCE doesn't care about any of that since the source flux streams do not have a break there, although I'm still not quite clear why Sequencer Plus MKI and Sequencer MKII will won't run in PCE.

robbo007
Posts: 87
Joined: Wed May 23, 2012 2:56 pm
Location: Spain

Re: Problems writing back to floppy disk

Post by robbo007 »

Interesting. So is there a way to force KF to not leave that slight break?

I have the TC images converted by NRS which work on DosBox-TC. Not sure if they would help on this investigation? But all three versions and the Sideman run fine. Unfortunately I have a Deluxe option board and cant write these to floppy to test as the softguard CP was removed from TC from version 4.x onwards. I wanted to write to floppy then dump the copy again to see if it made any difference.

I'll include everything here:
https://www.mediafire.com/file/s0rigo7q ... tc.7z/file
https://www.mediafire.com/file/amxq2or1 ... C.ZIP/file

ZrX
Posts: 618
Joined: Tue Dec 06, 2011 9:09 pm

Re: Problems writing back to floppy disk

Post by ZrX »

The break is generated by the drive when the write current is removed from the heads. Some generate larger glitch, some shorter.

KryoFlux, or DTC, when writing from raw files just writes the track from index to index as it has no further knowledge if a break should be made to appear somewhere else. Such information would only be available when writing from IPFs for which all the protections and other quirks have been analysed and recorded.

robbo007
Posts: 87
Joined: Wed May 23, 2012 2:56 pm
Location: Spain

Re: Problems writing back to floppy disk

Post by robbo007 »

ZrX wrote:
Sun May 24, 2020 10:29 pm
The break is generated by the drive when the write current is removed from the heads. Some generate larger glitch, some shorter.

KryoFlux, or DTC, when writing from raw files just writes the track from index to index as it has no further knowledge if a break should be made to appear somewhere else. Such information would only be available when writing from IPFs for which all the protections and other quirks have been analysed and recorded.
Ok thanks. Not sure if the SPS will create a IPFS for these. One thing I did find was the original MKIII would not load under Dos 5.x only under 3.21.

ZrX
Posts: 618
Joined: Tue Dec 06, 2011 9:09 pm

Re: Problems writing back to floppy disk

Post by ZrX »

There's Keir's open source Disk-Analyser which can convert from raw to IPF but I've never tried it so don't know what formats does it support.

User avatar
mr.vince
Posts: 2135
Joined: Tue Oct 05, 2010 5:48 pm

Re: Problems writing back to floppy disk

Post by mr.vince »

You could also re-format the raw streams to juggle the data in the track as needed to move the write splice to the write splice. Since we generate one file for each track, it should be easy to re-format and exchange.

Post Reply