Welcome, Guest. Please login or register.

Author Topic: Catweasel success stories?  (Read 6342 times)

Description:

0 Members and 1 Guest are viewing this topic.

Offline Schoenfeld

  • Full Member
  • ***
  • Join Date: Mar 2004
  • Posts: 175
    • Show all replies
    • http://icomp.de
Re: Catweasel success stories?
« on: December 03, 2006, 03:02:05 PM »
InTheSand:
The start-cancel-start again-trick is not necessary any more with the drivers that have been published more than three months ago. Please update your drivers if you still have to do that. Also read the readme files of the updated archives, because you might run into typical Windows-problems where the PnP manager pretends to have updated the drivers, but in reality left the old files in place.

Everyone:
The Catweasel can already read more bad/weak/damaged disks than a real Amiga can. We currently have two more things for data recovery in the works:

Automatic correction of streams if bits are found "off-center" - this will be a completely automatic process that will run in the background whenever checksums are not correct. Re-interpretation of the data will take place before a physical re-try is started. The routine already gave very good results on 25-year old 8" disks, and we're trying to tweak it in a way that no manual values have to be entered. The algorithm is very promising, as one of our 8" disks appeared unformatted to the original machine. The Catweasel can read the data off the disk, which has CRC16 checksums (almost impossible to go wrong with that).

The second thing is a method of judging if a signal comes from a mis-aligned drive. We're still working out the mathematics behind that, so all you need to have is enough courage to loosen the adjustment screws of your disk drive. This will let you adjust a drive for a specific disk (or set of disks) that has (have) been written with a mis-aligned drive. Again, this is very promising, as we found that almost all drives that you can buy new today are mis-aligned.

The imagetool does not have batch-backup features yet. We've focussed on the real bugs in the past, but we're now at a stage where we can think of new features like the above.

Jens Schönfeld
 

Offline Schoenfeld

  • Full Member
  • ***
  • Join Date: Mar 2004
  • Posts: 175
    • Show all replies
    • http://icomp.de
Re: Catweasel success stories?
« Reply #1 on: December 04, 2006, 03:26:04 PM »
ThEcRoW:
I'm referring to the version published on august 24th, 2006. The file with version 2.0.0.0 is still there, but I have replaced the link with the file name for version 2.1.0.0. Yes, you might have missed an update :-)

Jens Schönfeld
 

Offline Schoenfeld

  • Full Member
  • ***
  • Join Date: Mar 2004
  • Posts: 175
    • Show all replies
    • http://icomp.de
Re: Catweasel success stories?
« Reply #2 on: December 04, 2006, 03:42:00 PM »
LoadWB:

The 1541 and 1571 drives have limited access to the raw data that comes through the PLL. It's enough to make a good guess about the head alignment, but that's all you have: An "educated guess", but not the true information. Please double-check if you're really doing head alignment, or just motor speed adjustment.

The Catweasel moves more signal processing into the software, giving more raw information to the computer, so the computer can calculate a "quality of signal" that you can use to judge if you're moving in the right direction with your adjustment.

The Amiga on the other hand has no access to the raw information at all. Writing some software that allows you to adjust your drives is impossible. I do remember a program that lets you adjust the motor speed of the old NEC1036 drives (on some very early FredFish disk!).

Jens Schönfeld
 

Offline Schoenfeld

  • Full Member
  • ***
  • Join Date: Mar 2004
  • Posts: 175
    • Show all replies
    • http://icomp.de
Re: Catweasel success stories?
« Reply #3 on: December 21, 2006, 11:09:59 PM »
Just to let everyone know: There has been another update to the Catweasel drivers. Some more error messages, finally a retry button in the imagetool, and two types of "PLL" to choose from if you have disks that refuse to read properly. Not the full forensic functions that we're working on, but already a good point to start from.

Troy Davis made a lot of suggestions to improve the Imagetool, thanks for that. Most of those features need to wait until some more major re-builds of the driver have been accomplished. At this time, the decoders are located in the Kernel driver, so the Imagetool can only pass on information like "checksum error" to the user. After moving the decoders into the imagetool, it "knows" a lot more, and can run all the forensic algorithms that we have developed.

On another note, would anyone be interested in high-speed Amiga disk reading? I have an idea of speeding up drives and making batch conversion much faster. However, if nobody is willing to pay for such modified drives, I might as well skip on that development.

Jens Schönfeld