Copy II PC Central Point Option Board vs KryoFlux

[Thraka]

What does the Option Board (deluxe) give me over the KryoFlux?

They both do the same thing in the end, archive a disk to a file and allow you to write it back. The option board isn't being made anymore, yet the KryoFlux is. The KryoFlux seems to be more compatible with more floppies/platforms and disk sizes.

 

[barythrin]

lol.. it's a holy war. You might search the forums or cctalk mailing list for some biased opinions. I think a lot of it has to do with folks that already have the technology, so if it works there's not much reason or interest in a newer product. The only bad part with the Kryoflux IMO is the lack of ability to convert the rip yourself. If that was released and I could also restore my images myself I'd probably buy one, but as far as I know it's still a rip it, upload it to them and they'll convert it and send it back to you deal. A recent thread though did seem like someone had some tools to convert one image to something compatible or that they could write back using a DOB.

There's also another recently published and created project dubbed the "DiskVacuum" that's being discussed here and there. It sounds like it's an open source interface/project with some common parts that may benefit the community.

The other deciding factor may be how technical do you want it to be? There are catweasel wizards out there, DOB users, Kryoflux I think SOUNDS fairly set it and forget it until you have to email/upload it and forget it until you get it back.

 

[Al Kossow]

I don't get it..

I don't get it..

lol.. it's a holy war.

What is the fascination with Deluxe Option Boards?

Is there some secret warez stash of images or magic software for these things?

It's been out of production for twenty years now and has no decent documentation and ancient software.

There was never an 'official' way to produce disk images with it.

 

[Al Kossow]

The only bad part with the Kryoflux IMO is the lack of ability to convert the rip yourself.

Did you ever price one if you're an archive or museum?
They sent an insulting message to one of the other CHM staff members when they questioned their pricing policy
after being shocked at what it was going to cost to buy one.

Unfortunately, Diskferret is dead.
I'm hoping some traction will build behind Jim Brain's Diskvacuum project.

 

[hargle]

What is the fascination with Deluxe Option Boards?

Quite simply, they just works as expected.

If you have a disk, you can image it, and you can write it back to another disk to check if your copy works.
You can then pass that image to others, and they too can write that image back to their disks with their OB. That disk has now been preserved.
I don't really know what else one needs to do with such a device. Preservation has practical limitations on its usefulness.

Is there some secret warez stash of images or magic software for these things?
It's been out of production for twenty years now and has no decent documentation and ancient software.
There was never an 'official' way to produce disk images with it.

I dunno what you mean by official, but between the 4 people I know who have option boards who are into game preservation and archiving, we are completely capable of producing disk images and sharing them. I guess we are building up a secret stash of images by the very nature of using the card, which we will gladly share with you!

Somehow we are able to get by with the available documentation and our ancient software. We are also copying ancient software with it, so it goes with the territory.

 

[dorkbert]

... we are completely capable of producing disk images and sharing them.

As we are fond of saying in the olden days... "backups we keep at our friends' place in case our house burns down..."

 

[Chuck(G)]

I don't see what's so special about any of these widgets, particularly since most people use them to archive copies.

You can build one yourself using a microprocessor--I've done it using something as primitive as a ATMega AVR and an SRAM chip. In fact, most MCUs have a "capture" timer feature that makes the job easier.

Writing copies is a bit more involved because you have to get rid of the quantization error in your sampling and also account for write precompensation. But there's code out there to help you with that.

That being said, I find that the old Catweasel MK III meets most of my needs, is simple to program and slips into a PCI slot.

It really isn't rocket science.

 

[barythrin]

I honestly have no idea regarding the DOB, it was just that or the CatWeasel that I was aware of in the 90s when I was looking for archiving tools. Admittedly I haven't used either but was under the impression that they weren't too difficult to use (other than seeing a lot of specific configuration settings on the catweasel). I'm confused about the warez comment though, I saw you said the same on the cctalk but wasn't quite getting the reference. Is it in reference to Transcopy for the option board not being free? I don't see the pricing on the kryoflux anymore but do recall a pretty insane price for all the software yourself ($500 or $2000?) I don't think I was aware of these kryoflux tools. Perhaps the device is becoming more practical if folks write converters or if the kryo team releases theirs.

Regarding Chuck's comment to the general public it's probably close to rocket science. I know it's simpler to those who have experience in AVR or other microcontroller programming but understanding the circuitry logic and the output you'd receive are still engineering and computer science that's above the average end user. I think the general community needs an easy, intuitive tool that any of our wives or kids could use to archive our stuff in the unfortunate event that something stupid (body found under their favorite mainframe) happens to us.

I also realize I did misspeak earlier about the Kryoflux not being able to write and my info is dated to maybe 2 years ago, but it couldn't write back what it recorded; it required you to send that raw image to them to convert to an IPF that it COULD write but only for some (primarily Commodore) formats. These tools above though look interesting along with that thread that Trixter and fs5500 were having also with some conversion tools..

 

[Unknown_K]

I have a DOB and it allows you to read original mac floppies as well. If you collect original vintage software and want to run it its nice to be able to copy a protected disk so you don't ruin the original. Most games and most major apps of the period have been cracked and released as warez, but there are some obscure apps and industrial programs that never were.

The major issues with all these boards is that if yours dies the images you made are worthless unless you can get another board. What would be nice is a board that makes an image that can be mounted and read with open source software. A virtual floppy like we use virtual CDROMs for installing software.

 

[Trixter]

Throwing another wrench into the works, I just found out about the Super Card Pro: http://www.cbmstuff.com/proddetail.php?prod=SCP

Option Board copies disks, but closed source, closed hardware, company out of business, and requires vintage gear. KryoFlux is open hardware and open source, but in the PC realm mostly only reads disks to images. The above project seems like it solves everybody's problem, but I haven't ordered one yet.

 

[per]

Throwing another wrench into the works, I just found out about the Super Card Pro: http://www.cbmstuff.com/proddetail.php?prod=SCP

Option Board copies disks, but closed source, closed hardware, company out of business, and requires vintage gear. KryoFlux is open hardware and open source, but in the PC realm mostly only reads disks to images. The above project seems like it solves everybody's problem, but I haven't ordered one yet.

I've been able to decode data out of Option Board Images. The magnetic-level hi-s and low-s are stored in raw format througout the image files at a 1:1 sampling clock rate (1 bit per tick), but the formatting is not always entirely clear. The memory buffers used to create these images aren't cleared before use, so there is a lot of junk left by other terminated programs that gets in the way of reverse engineering.

What worked was to parse the image file for MFM encoded track/sector/data chunks. These images are assembled track-wise, and the tracks are scanned sequentially. You then first look for track 0 head 0, and then you decode and organize the sectors that "makes sense" from there: Was only 8 sectors found before track/head change?; problably a 160K/320K disk - else 180K/360K. Did it go right to track 1?; Problably single sided disk (160K/180K) - else double sided (320K/360K). Was the new track number anything else than the-previous-track-number++?; Problably junk left in the buffer from an earlier image.

Otherwise, the Option Board gets five stars for convenience. It's installed just like a regular ISA card, and all you have to do to use it is basically to load up a program and insert a target disk in the systems disk drive. It doesn't get easier than that.

 

[Chuck(G)]

One issue with some of the later cards is that the software doesn't do an accurate job of copying. In other words, if, starting at index, 6,250 bytes are read before index is again encountered, you should be able to write a disk that's exactly 6,250 bytes between index pulses. But that rarely happens because of what I call "clone drift".

Suppose that you're sampling an MFM disk written with a clock rate of 1 MHz using a 10 MHz clock. So you have transitions that are 2, 3 and 4 original clocks apart. So, at 10MHz, you'll get samples that cluster around 20, 30 and 40 "magic board" clocks apart. But those counts will be truncated, rather than rounded, so you'll find that most cluster around 19, 29 and 39 clocks per sample. Unless your writing software knows to adjust these counts, bit cells will be slightly closer to each other every time you use a cloned disk to make another cloned copy. Not only that, but there are also issues of ISV (instantaneous speed variation) that can afflict drives to further spoil the counts.

Now, move to the center of the disk. You're now in an area where transitions close to each other will tend to "spread" the interval apart when writing. Your writing software needs to pre-compensate for this effect.

It's easy to read a disk, but much harder to create an accurate copy.

 

[barythrin]

Wow. That SuperCard Pro really is interesting. Such a young product (released pretty much end of Q4 2013) and I guess just Jim as the developer but it's doing some great stuff already as far as a support for most MFM/FM recordings (same as KryoFlux). Several users have tested it with the more difficult copy protection schemes and errors have already been fixed in the software. Definitely looks promising. Anyone here try it?

 

[Thraka]

I'm really close to buying one. Which is why I was really wondering about just selling off my option board on ebay. These seem like a better option. I'm not really interested in collecting old ISA cards, so I don't need one for a "collection"

 

[per]

Suppose that you're sampling an MFM disk written with a clock rate of 1 MHz using a 10 MHz clock. So you have transitions that are 2, 3 and 4 original clocks apart. So, at 10MHz, you'll get samples that cluster around 20, 30 and 40 "magic board" clocks apart. But those counts will be truncated, rather than rounded, so you'll find that most cluster around 19, 29 and 39 clocks per sample. Unless your writing software knows to adjust these counts, bit cells will be slightly closer to each other every time you use a cloned disk to make another cloned copy. Not only that, but there are also issues of ISV (instantaneous speed variation) that can afflict drives to further spoil the counts.

This brings up the question; At which layer is it nessecary to preserve the discs? I consider a typical disk to have the following layers:

1. Physical Medium - Micrometers per transition, Micrometers between each transition. Other physical stuff like Index hole, Track division, dimensions, etc... Even laser-hole stuff.
   -> bit representation; timing and layout stuff, bitrate, spinrate, etc...
2. Raw Data - Raw bits sampled on the medium by a clock.
   -> Encoded - FM/MFM/GCR
3. LL Data - Data encoded as Raw Data.
   -> LL Format - Header/Data packets; sector organization.
4. HL Data - Data stored in the data packets of the LL Format.
   -> HL Format - OS File System.
5. File data - Organized by the file system.

A standard "Copy A:* B:*" in DOS will only preserve the OS File System layer.
A typical raw image of a disk will preserve the HL Data layer.
The DOB tries to preserve the Raw Data layer, while keeping track of some stuff about the Physical Medium layer.

 

[Shadow Lord]

I've done it using something as primitive as a ATMega AVR and an SRAM chip.
... It really isn't rocket science.

Chuck(G) != Your typical computer user (even one who uses vintage computers)

Just saying...

 

[SomeGuy]

This brings up the question; At which layer is it nessecary to preserve the discs?

That is a question that bugs me a bit too. You have to ask yourself WHAT you are really archiving, and why.

Ideally, the only thing that would need to be archived are files. For things like driver disks, shareware, or low-volume software, anything lower is not really relevant. The exact location of all of the data can vary wildly as the disks may have been produced by hand or at varying times.

But when you get in to system disks, or "booters" then it is good to have a raw sector image of the disks because you may be dealing with bits of data that do not exists as normal files, such as boot sectors. For mass produced disks, this is also a good thing to have for "authenticity", as well as to appease buggy installers that expect volume labels and such. Add to that, most modern emulators can easily handle raw disk images but not file archives. If disk isn't copy protected, then are the lower level magnetic pulses relevant?

Since we don't live in an ideal world, then there are the dirty copy protected disks. The low level magnetic bits are only needed to make the files or images run without further modification. If the archives are intended to preserve a usable experience, rather than a series of ones and zeros, then just apply an unprotect and let the copy protected bits rot! But again, this is not an ideal world. Unprotects do not exist for every protected disk.

So archive ALL the things!

All the things?!

And this is the bit that drives the OCD folks nuts. An archive can contain files, unprotected files, IMG, unprotected IMG, IMD, TD0, C2P, TC, IFP, or whatever, but they are not always applicable. Your driver disk doesn't need to be dumped with a TransCopy board, but your booter game disk doesn't even HAVE files! For your protected word processor, Copy II PC can re-create an operational disk, but ImageDisk and Teledisk can't, so there is no point including those files. Your Non-DOS disk may not even be copy protected, but it needs 256 byte sectors or whatever, so a raw WinImage dump isn't going to do it. Only want to store KryoFlux images? Great, now everyone has to buy this piece of special hardware... and so on. :confused3::rastarolleye::jumping1:

I don't know about anyone else, but at the end of the day, I just want to run an old application and see what it was all about.

 

[Chuck(G)]

I preserve the originals--and the image.

And I agree that just the files may not get you there. An example is Harvard Graphics 1.0 or so--the copy protection is located in the inter-sector gap.

 

[cr1901]

Wow, I have a lost of questions after reading this thread:

They sent an insulting message to one of the other CHM staff members when they questioned their pricing policy
after being shocked at what it was going to cost to buy one.

Source (I believe you, I'm just curious where you found this out)?

I'm really close to buying one. Which is why I was really wondering about just selling off my option board on ebay. These seem like a better option. I'm not really interested in collecting old ISA cards, so I don't need one for a "collection"

I've never heard of an Option Board before this thread, and am still doing research. However, I DO collect old ISA cards, and this seems like an interesting addition- would you be willing to sell it off Ebay? As for Trixter's point... if the software can be found, I can probably use IDA to reverse-engineer it.

EDIT: Since VBulletin apparently never heard of strikethrough, consider the previous paragraph null and void (http://retro.icequake.net/dob/ Up to $300 for one!)... there is probably someone who can put that hardware to better use than I. The only disk I really need copied is 100 Monochrome Mazes.

The exact location of all of the data can vary wildly as the disks may have been produced by hand or at varying times.

Maybe I'm misreading something, but how does one produce a floppy disks magnetic material and prepare it for use in a real disk by hand?

I don't see what's so special about any of these widgets, particularly since most people use them to archive copies.

You can build one yourself using a microprocessor--I've done it using something as primitive as a ATMega AVR and an SRAM chip. In fact, most MCUs have a "capture" timer feature that makes the job easier...

Writing copies is a bit more involved because you have to get rid of the quantization error in your sampling and also account for write precompensation. But there's code out there to help you with that.
...
It really isn't rocket science.

With all due respect, not all of us can be Chuck Norr- err, Guzis .
Okay, bad joke aside, I'm sure most of us here could create a circuit to do such a task if we had the time to look up old literature- it's ALL there to learn the prerequisites, but not necessary right in front of you. You must be of those people who think vi is actually a good text editor (spoiler: it isn't... not with its 9000 year learning curve) !

Also, re: the inside of the floppy disk, why would transitions be spread further apart? Constant Angular Velocity?

 

[Shadow Lord]

Basically to agree with others in this thread your average joe computer user (myself included) is not about to roll his own dedicated FDC/copy protection bypass board. What would sell IMHO something that would meet the following criteria:

1. A board that allows read/write from std. floppy drives on contemporary systems (i.e. today that would be a 360KB drive hooked up to an i7 machine as an example).
2. It would need all the SW in place (no writing your own code) to make it be fully functional (drivers, read/write/analyze program, etc.).
3. It would need to be ready to go out of the box
4. Finally it must have an easy interface with simple commands (i.e. copy to disk, copy to image, write to disk, etc.). Advanced options could be included for fine tuning but that should not be the only way to run it

The Copy II PC OB met all these criteria back in the day and it was insanely successful. From what I have seen of the Catweasel MK4, the Kyroflux and this SuperCard Pro is insanely powerful HW but very poor SW support. The SW is limited, intimidating, or non-existent in many cases. If these guys could come up with a nice SW interface I could see someone (e.g. me) buying a card bolting it into a 5150 (whose innards are gone) and installing 4 HH drive (5 1/4" DD, HD, and 3 1/2" DD, HD). Connect a US cable to a modern system and you would have the ultimate archiving/reproduction machine.

P.S. is it me or most of these seem aimed at copying Amiga/Atari/Commodore disks even though the HW/SW runs on a PC?