Original Workbench (and related) disks preservation - Checksums required

[TCD]

If I'm not mistaken that looks like ADF-Workshop by Mr Crashdisk himself

[Crashdisk]

@TCD
I confirm ^^

You'll find more information here:
https://eab.abime.net/showthread.php?t=71922

[rhester72]

That is absolutely brilliant! Wish I'd known about it earlier - thank you very much!

[Snoopy1234]

I've just bought an A3000 Workbench 2.04 set and it includes the elusive Install disk [367826-01]. I'll upload the adf when it arrives so checksum and details can be added to: http://amiga.workbench.free.fr/workbench.php

That is if it's unaltered - being an install disk it should be ok.

[Snoopy1234]

Someone has uploaded a 2.04 install disk to TOSEC Updates.

/turran/FTP/TOSEC Updates/2023-07/fix_Commodore Amiga - Operating Systems - Workbench/Workbench v2.04 rev 37.67 (1991)(Commodore)(A3000)(Disk 4 of 4)(Install).zip

If this is the missing A3000 Install Disk P/N: [367826-01] IDK. It's CRC-32 is 5810AEC1

The 2.04 set I ordered should arrive in a couple of days so I will image the install disk P/N: [367826-01] and upload it to my FTP folder for verification.

[Snoopy1234]

The 2.04 Workbench set arrived. I've uploaded the adf and disk scan of the Install disk PN: 367826-01 to my Uploads folder.
The CRC-32 is 319698B0
ADF-Workshop shows no alterations since duplication though shows some deleted files which could have been part of the original source or not.

This may be of use for The Amiga Workbench Project (FR) as the info is missing.

[TCD]

Just a reminder for both EAB and the file server: Because the OS files (ROMs and Workbench disks/files) are still being sold they are prohibited from upload to the zone or the file server. From the file server upload rules:

* DO NOT upload "Workbench Disks", "Kickstart ROMs/Disks" & software or music files still supported/for sale.

[Snoopy1234]

Originally Posted by TCD

Just a reminder for both EAB and the file server: Because the OS files (ROMs and Workbench disks/files) are still being sold they are prohibited from upload to the zone or the file server. From the file server upload rules:

Ah ok. Thanks for letting me know.

[derSammler]

Originally Posted by Snoopy1234

That is if it's unaltered - being an install disk it should be ok.

Sadly, finding an unaltered install disk is rather unlikely. When setting up the hard disk, it must write the drive definitions to the disk, otherwise you can not continue.

I have many 2.x install disks incl. the one that only came with the A600HD. None of them are in their original state. On the other hand, it does not really matter, as using it will modify it anyway.

[Snoopy1234]

Originally Posted by derSammler

Sadly, finding an unaltered install disk is rather unlikely. When setting up the hard disk, it must write the drive definitions to the disk, otherwise you can not continue.

I have many 2.x install disks incl. the one that only came with the A600HD. None of them are in their original state. On the other hand, it does not really matter, as using it will modify it anyway.

Interesting. I didn't know that. The TOSEC Update image and my adf both have the same write date and time according to ADF Workshop so I'm guessing that is a master datestamp they were both written from. Both images also report no alterations bar my adf which "had" extra files aux-handler, speak-handler and Say.info deleted. These files were not part of the 2.04 Install distro so I'm not sure where ADF Workshop got that from - there's no disk alterations recorded.

[Snoopy1234]

Commodore Workbench 2.04 Install (A3000)(367826-01) adf
CRC32 is: 319698B0

[8bitbubsy]

I have a dump of the original Workbench 1.3.3 Norway (p/n 317793-04) disk and its included Extras intl. (p/n 367618-01) disk. Both disks were write-protected and didn't have the write-protect sliders present, so they are very likely to be untampered. The WB disk had 3 bad sectors, but I reconstructed them from the Danish WB 1.3.3 disk (the content in the bad sectors was easy to locate in the DK disk, only had to modify the "next block" fields and recalculate block checksums after reconstructing). It's likely that the fixed dump is bit-intact since there are small file changes between these disks, but I added the [m] tag as I can't verify it. I have tested the fixed Workbench ADF and it works and operates just fine.

Is this of interest? This dump seems to be missing in TOSEC.

[Crashdisk]

Hi,
of course, this is of interest to TOSEC, but you'll understand that we're not here to share this kind of disk.
Maybe the disk obtained is already known...
Could you calculate the checksum (CRC32)?

[8bitbubsy]

It's not like I would share it to the public!

How do I calculate the checksum? Do I just run the whole ADF binary through a CRC32 routine?

[Crashdisk]

Originally Posted by 8bitbubsy

It's not like I would share it to the public!

How do I calculate the checksum? Do I just run the whole ADF binary through a CRC32 routine?

Simple methods of calculation include compressing the ADF file in Zip, Rar or 7Zip, and when you open the archive, the archiver's interface will display the CRC32.
It all depends on your working environment, but it's a classic...

[8bitbubsy]

Workbench v1.3.3 rev 34.34 (1990)(Commodore)(A500-A2000)(NO)(Disk 1 of 2)(Workbench)[m].adf $DB8D4B84
Workbench v1.3.3 rev 34.34 (1990)(Commodore)(A500-A2000)(NO)(Disk 2 of 2)(Extras).adf $E8EB34C4 (Extras international disk)

[Crashdisk]

[DB8D4B84] This file is unknown to us
[E8EB34C4] Workbench v1.3.3 rev 34.34 (1990)(Commodore)(Disk 2 of 2)(Extras)
For disk 1, I see 2 possibilities. Either your repair is not 100% compliant, or it really is the only known copy...

[dlfrsilver]

Originally Posted by Crashdisk

[DB8D4B84] This file is unknown to us
[E8EB34C4] Workbench v1.3.3 rev 34.34 (1990)(Commodore)(Disk 2 of 2)(Extras)
For disk 1, I see 2 possibilities. Either your repair is not 100% compliant, or it really is the only known copy...

You need IPFs of workbench disks to get perfect ADF with good CRC. Otherwise you will just stack up modified ADF.

[Crashdisk]

Originally Posted by dlfrsilver

You need IPFs of workbench disks to get perfect ADF with good CRC. Otherwise you will just stack up modified ADF.

Disk 1 has errors, and I don't see how converting it to IPF would result in a better ADF. What is the reparability rate?
If the disk has been altered by the system with the addition of an .info file or whatever, conversion to IPF won't correct that! Your analysis system will just tell you that the disk has been altered by a normal floppy drive, but that's all.
Am I wrong?


In any case, I have my doubts about the repair. Replacing the data, modifying the next block and recalculating the checksum are not enough to make a good repair, especially if the last sector of the file is affected...

[dlfrsilver]

Originally Posted by Crashdisk

Disk 1 has errors, and I don't see how converting it to IPF would result in a better ADF. What is the reparability rate?
If the disk has been altered by the system with the addition of an .info file or whatever, conversion to IPF won't correct that! Your analysis system will just tell you that the disk has been altered by a normal floppy drive, but that's all.
Am I wrong?


In any case, I have my doubts about the repair. Replacing the data, modifying the next block and recalculating the checksum are not enough to make a good repair, especially if the last sector of the file is affected...

I mean that you need IPF files from original disk, then convert IPF to ADF.

I have some dumps of my own workbench disks that are unmodified, from those you can get proper ADFs and so get valid CRCs.