[FOUND] Trojan Lightpen Drivers?

Originally Posted by prowler

If you have a PC with two floppy disk drives, I would like you to try making another disk image with Disk2FDI Trial Version. This program is capable of capturing every valid disk block on a disk whether there are bad blocks on the same track or not.

I think I have a floppy cable knocking around which I could put a couple of 3.5" pc disk drive into. You do mean pc as in ibm pc right? I don't want to misunderstand and put the final coffin nails into the thing!

Originally Posted by prowler

AS a last resort, if the disk image cannot be reconstructed entirely from the disk images and files you can upload, I would like to try the disk in my Catweasel and KryoFlux setups. These are ideal for salvaging flaky disks, a they can be configured to concentrate on difficult tracks and sectors. I would, of course, return the disk to you afterwards.

No worries, I'm not too attached to it anyway, just wouldn't mind being able to use a successful copy of it will do me. When it comes to it, quick pm to me with your postal addy and we're away!

[prowler]

Hi Mike, bad news, I'm afraid.

Both the TrojanLP.adf and Lightbench.adf disk images contain complete rubbish from Disk Block 1375 on.

The reason why the startup-sequence file copies off the Lightbench.adf image and More fails at a later data block in the file chain is because those blocks are zeroed out in that disk image, so the checksum errors are eliminated.

Actually, the Lightbench.adf image does have some valid disk blocks in the range 1375-1759, but they are completely unrelated to the Trojan LightBench disk.

In the case of the TrojanLP.adf image, Disk Blocks 1376-1759 appear to start at offset 0x060 and end at offset 0x05F in the next block!

Block 1376 is at offset 0xAC000 in the TrojanLP.adf disk image. I made a copy of it and called it LPTrojan.adf. I cut 96 bytes from offset 0xAC000 in the file and appended them to the end of the image in Disk Block 1759. Those 96 bytes probably belong at the end of Disk Block 1375, so Disk Block 1759 still has a checksum error, but this has made Disk Blocks 1376-1758 valid! However, they still have absolutely nothing whatsoever to do with the Trojan LightBench Disk.

It's strange enough that everything from Disk Block 1375 in both images is complete garbage as it relates to the Trojan LightBench disk, but it's even more strange that everything in that region is completely different in either image and that in each case the valid blocks are in the correct place (or very nearly so in the TrojanLP.adf image), but on the wrong disk! How in hell could that have happened!

It will not be possible to recover the disk using the images uploaded so far, because of the data missing from the latter part of each, but please don't be discouraged from giving it another go.

Tomorrow, I'll have a look at the files you have recovered from the disk with DiskSalv...

Originally Posted by mike_cc

I think I have a floppy cable knocking around which I could put a couple of 3.5" pc disk drive into. You do mean pc as in ibm pc right? I don't want to misunderstand and put the final coffin nails into the thing!

Yes, a Pentium-class or better IBM-compatible PC will do the job.

Originally Posted by mike_cc

No worries, I'm not too attached to it anyway, just wouldn't mind being able to use a successful copy of it will do me. When it comes to it, quick pm to me with your postal addy and we're away!

Thanks. I'll send you a PM. After the Christmas/New Year period will be soon enough. I wouldn't like to see this one slip though our fingers.

[roy bates]

if its any consolation i have a disk as well,but its damaged if you want i can send you the disk prowler to see if you can get any info from it,and combine it with mike_cc disk to get a good one.

[prowler]

Originally Posted by roy bates

if its any consolation i have a disk as well,but its damaged if you want i can send you the disk prowler to see if you can get any info from it,and combine it with mike_cc disk to get a good one.

That's very generous of you, Roy! I'll send you a PM too.

BTW, did you know that there were in fact two disks for the Trojan Lightpen?

http://privat.bahnhof.se/wb409723/my...en%20-%202.jpg

It's quite possible that Mike and yourself may actually have both disks between you.

Originally Posted by prowler

That's very generous of you, Roy! I'll send you a PM too.

BTW, did you know that there were in fact two disks for the Trojan Lightpen?

http://privat.bahnhof.se/wb409723/my...en%20-%202.jpg

It's quite possible that Mike and yourself may actually have both disks between you.

Interesting and that solves a little mystery for me. A sneaky smegger had put a little circular sticker on the disk, although you could tell that behind it, it said 1.3.

Presumably the seller didn't want to put off potential buyers by restricting perceived use to wb1.3 systems. Bit underhanded but oh well. I've not had time to install a second 3.5 drive to my pc as I've been desperately trying to get a full whdload A600 kitted out for my nephew for Christmas. Not having a lot of luck with that either

Good luck with the recovered files, I hope you find something useful there. I'll try the 2xdiskdrive when I get a chance. Merry Christmas everyone

[Calgor]

I have this lightpen and the 2 original disks exactly like the ones in the previous linked picture (and a third custom disk from the previous owner - could be a modification of the wb2.0 disk as the volume label is the same).

Just dumped the disks and uploaded to the zone. They seem to work fine with a "dir" command in WinUAE so I believe they copied with no errors (transdisk I don't think verifies).

TrojanLightPenDrivers.lha

troj13.adf: Lightpen Driver and KWIKDRAW Workbench1.3 Disk
troj20.adf: Lightpen Driver and KWIKDRAW Workbench2.0 Disk
trojcust.adf: Label says "LIGHT PEN DRIVERS ETC. TO USE PEN YOU MUST BOOT FROM THIS DISC FIRST. NOT H/D."

Merry XMas!

[roy bates]

thanks alot calgor!!!,can some kind mod put this on the eab.server for future use?


ill try them out when i have time.

and can the title be edited to found please.

thanks everyone for your help.

[Calgor]

I should add that I haven't tested the disks with the light pen, but they should be fine - the drive noise while copying was extremely consistent apart from a very slight variation on the last few cylinders of the wb1.3 disk. Also the command I used to list the disks' contents after copying was actually "dir all" so I think that would cover more sectors of the disk than just a simple "dir".

Let us know how you go with using the disks with your light pen.

[roy bates]

will do!i cant test it at the mo though im repairing the 1084,the monitor works its just in bits,lol



EDIT:-prowler if you read this the damaged disk i have is the 1.3 disk,if the disk images that calgor has uploaded look good then it saves alot of time and effort on your part,again thanks everyone for this.

[roy bates]

ok the pen works great,ive yet to test the 1.3 disk but as far as i can see it works!


if the mouse pointer jumps around on screen make sure your NOT using a 100hz tv.

ive used it on a 1084s and a portable colur crt tv,it wont work on a lcd ,led or 100hz tv
thanks to everyone again,its most appriciated.


EDIT:-wow,this contradicts the info about the pen and driver!heres some interesting info.

ive tried the driver off the lightbench 2.0 disc it works on workbench 3.1 on a 1200,in the following screenmodes

lowres
lowres,interlaced
highres
highres,interlaced
superhighres
superhighres,interlaced

hvent treid it in ntsc yet, but it should work.

[prowler]

WOW! Nice one Calgor!

I thought I had made some useful progress on the Trojan Lightpen disk image by the time I logged in this evening, which convinced me that Mike's disk is probably recoverable.

The DiskSalved files in the TrojanLightbench.zip archive in The Zone - including the files that won't copy out of the TrojanLP.adf image - are more good than bad, which means that Mike's disk can almost certainly be imaged properly using a suitable disk drive coupled with the right controller.

The zip archive contains a Prefs/DPaintIV/C/Type file which does not show up in the files listed on the disk using the CLI. The file is actually in the disk image, but it had been deleted! The file can be reinstated by undeleting it.

I examined the salvaged files that won't copy out of the disk image with a binary file editor, and I discovered that the startup-sequence file is okay. The others have much more valid data in them than you would expect, given that they won't copy out of either of Mike's uploaded disk images properly. However - the startup-sequence file excepted - it is not possible to say for certain whether each file is good or bad merely by inspecting the code.

I extracted the zip archive within WinUAE and CMD is okay (toggles Cmd redirection of parallel.device), Cmd.info is okay, More is okay (lists ASCII files a page at a time), KwikDraw is okay (runs; quit by pressing the Escape key), lpen.iff is okay (requests volume 'DeluxePaintIII' to run DeluxePaintIII:DPaint, but the file opens okay in Paint Shop Pro and IrfanView for Windows), InstallPrinter appears to be okay (requests volume 'newcon') and GraphicDump appears to be okay as well, but does nothing whether parallel.device is redirected or not. Of course, some of these programs may not work properly anyway without a lightpen connected. The libraries I could not test, but valid files to compare them with can probably be found on Workbench 1.3.x disks.

Well, now that we have Calgor's uploads I will be interested to see whether I had actually made any useful progress or was wasting my time.

I will be looking at Calgor's disk images if I can find the time tomorrow.

BTW, I am still interested in trying to salvage your disks, Mike and Roy, if you're still willing to send them to me.

[roy bates]

im still willing to send you the disk i have if you want prowler.its 1.3 lightpen disk.

the disk's that calgor uploaded work perfectly though as ive said,but if you want send me a pm and il send it to you after christmas.


be aware though that the two disk are actualy for two os's 1.3 and 2.0 and above 1.3 lightbench doesent work on workbench 2 and above.

[prowler]

Thanks, Roy.

I would still like Mike's as well - especially as they're both the 1.3 versions. If he's still willing, I'll send you both a PM with my address details.

It's always astonished me that there are so many TOSEC variations for some disks. I'm hoping to find that your two disks turn out to be identical with Calgor's 1.3 disk image.

PS. Don't forget to include your address details with the disk so that I can return it to you afterwards.

[roy bates]

no probs prowler,you can keep the disk if you want its no use me.

but ill send it with my address in a note all the same.

EDIT:-i just had a thought,should i have this labled up in some way as this is magnetic media,if it gets past through a xray it will be erased.

[prowler]

This evening, I checked Calgor's disk images and none has an error - all files copy out okay!

Furthermore, I discovered that the Troj13.adf disk image contains files which exactly match those which span bad sectors on Mike's TrojanLP.adf image. And, even better, I have compared those thirteen files with those Mike DiskSalved from his original Trojan Lightpen disk and they are all identical!

This is the best indication we could have that Mike's disk is recoverable! And it will be worth it, because it appears to be a later sub-version of the software on Calgor's disk.

[Supamax]

Originally Posted by Calgor

Just dumped the disks and uploaded to the zone. They seem to work fine with a "dir" command in WinUAE so I believe they copied with no errors (transdisk I don't think verifies).

Thanks a lot Calgor!

A suggestion: when dumping a disk, ALWAYS do it TWICE (at least).
I always do, then I compare the two dumps with a binary editor (UltraCompare Pro). If they are identical, then I'm sure the dump is Ok.

[Calgor]

Originally Posted by prowler

This evening, I checked Calgor's disk images and none has an error - all files copy out okay!

Furthermore, I discovered that the Troj13.adf disk image contains files which exactly match those which span bad sectors on Mike's TrojanLP.adf image. And, even better, I have compared those thirteen files with those Mike DiskSalved from his original Trojan Lightpen disk and they are all identical!

This is the best indication we could have that Mike's disk is recoverable! And it will be worth it, because it appears to be a later sub-version of the software on Calgor's disk.

How does one determine if it is updated? I suppose if the version of the official programs is later or the files of those official programs are different. Many people mod their original disks! Well you can check out the custom disk I provided for diffs if any with the 2.0 disk.

Originally Posted by Supamax

Thanks a lot Calgor!

A suggestion: when dumping a disk, ALWAYS do it TWICE (at least).
I always do, then I compare the two dumps with a binary editor (UltraCompare Pro). If they are identical, then I'm sure the dump is Ok.

Very good tip although I have encountered consistent data corruption before where the checksum of the original file needed to be compared with the checksum of the written file (hardware or driver problem so all written files had the same error). Would prefer the program to do all the work instead of me (suppose I could write a simple shell script though )

I suspect once I find the right program, I will use it to make 2 dumps as you suggest.

[prowler]

Originally Posted by Calgor

How does one determine if it is updated? I suppose if the version of the official programs is later or the files of those official programs are different. Many people mod their original disks! Well you can check out the custom disk I provided for diffs if any with the 2.0 disk.

When I compared what has been salvaged so far from Mike's disk to the contents of your 1.3 disk, many files are the same, but some are not, and those have later date stamps on Mike's disk. Also, there is a subdirectory 'prefs/DPaintIV/C' on Mike's disk, which is not present on your disk, containing some additional command line utilities. One of these - More - has, however, been deleted - probably because this is already included in the utilities directory.

Wow - a lot of progress has been made! I just got back from the missus' parents where we've been for Christmas and read this. So cool that Calgor has the disks! Ok I'm considering that the best way to salvage the 1.3 version that I have, is to just post the disk to prowler, so I'll do that. In the meantime, I'll check the zone for calgor's disk images as I'm very excited about getting to use the pen!

[Supamax]

Originally Posted by Calgor

I have encountered consistent data corruption before where the checksum of the original file needed to be compared with the checksum of the written file (hardware or driver problem so all written files had the same error). Would prefer the program to do all the work instead of me (suppose I could write a simple shell script though )

I don't know if I understood correctly what you mean ... however, the compare I was referring to is a binary compare, so I can see which bytes (if there are) are different between the two dumps.
If they are different, I make a third dump (or more) so I can find out which is the correct dump file.