SuperCard Pro

Moderators: DrCoolZic, Moderator Team

User avatar
npomarede
Atari God
Atari God
Posts: 1095
Joined: Sat Dec 01, 2007 7:38 pm
Location: France

Re: SuperCard Pro

Postby npomarede » Wed Jan 01, 2014 5:35 pm

DrCoolZic wrote:Yes I know that macrodos is supported. I was taking this as an example of what needs to be done.
Yes any kind of bytes (or even bits) timing variation can potentially be described. But each time a new type of block needs to be described the lib needs to be modified.
I am not convinced that all protections used by Atari are handled today (at least Dungeon Master is not).

But the same could be said about STX ; file format needed modifications to handle fuzzy bits and bytes' duration change, as well as modification of the corresponding pasti.dll.
Now, as many more games were imaged with STX, we could maybe consider that all kinds of protections were encountered and pasti.dll won't change anymore, but I wouldn't consider this a big advantage against IPF/capslibrary (and as Atarizoll wrote, some games are not imaged with pasti, so changes might still be needed to the STX format)

Nicolas

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Wed Jan 01, 2014 5:43 pm

All the more reason to use flux... it doesn't matter what is discovered, it will always work....
I am the flux ninja

User avatar
DrCoolZic
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2144
Joined: Mon Oct 03, 2005 7:03 pm
Location: France
Contact:

Re: SuperCard Pro

Postby DrCoolZic » Wed Jan 01, 2014 5:49 pm

npomarede wrote:But the same could be said about STX ; file format needed modifications to handle fuzzy bits and bytes' duration change
Why? I think it can handle all we need?

User avatar
npomarede
Atari God
Atari God
Posts: 1095
Joined: Sat Dec 01, 2007 7:38 pm
Location: France

Re: SuperCard Pro

Postby npomarede » Wed Jan 01, 2014 5:54 pm

JimDrew wrote:All the more reason to use flux... it doesn't matter what is discovered, it will always work....

Unless you discover later than the flux you dump was wrong due to a defective media :)
Higher level tools have some drawbacks, as you need to "know" all the protection schemes, but at least it can help to validate/reject a dump if the result contradicts the expected layout for the protection. What I mean is that whatever method you use to store the data, you still need a way to validate that the dump is really correct and not altered by some media errors (maybe the higher resolution used in SCP allows to detect media defaults more easily ?)

User avatar
npomarede
Atari God
Atari God
Posts: 1095
Joined: Sat Dec 01, 2007 7:38 pm
Location: France

Re: SuperCard Pro

Postby npomarede » Wed Jan 01, 2014 5:58 pm

DrCoolZic wrote:
npomarede wrote:But the same could be said about STX ; file format needed modifications to handle fuzzy bits and bytes' duration change
Why? I think it can handle all we need?

What I mean is that Ijor certainly released several pasti.dll version before reaching the current version, adding extra structures to the original STX format to handle more exotic protections. So, it would be the same if IPF needed to be modified to add new formats that no one encounter so far since IPF is not really a big player in the Atari ST emulation so far.

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2387
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: SuperCard Pro

Postby AtariZoll » Wed Jan 01, 2014 6:00 pm

This with flux is very promising. Then in case of Atari you need only good WD1772 emulation in emulators. Size of images is surely not problem now, and 10 MB is really not much. Of course, there is possibility to make combined image with ST format like sector datas, and using flux only for protected tracks. Such format would be still very simple.
Maybe best thing in it is HW floppy emulation - concept with flux based images. Writing onto real floppies is something outdated now.

"Unless you discover later than the flux you dump was wrong due to a defective media :)"
Yes, it is very important, especially now, with aged floppies. Jim, what are experiences with detecting defective media in flux images ?
I'm not against GMO, I'm against that children play with fire.

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Wed Jan 01, 2014 6:09 pm

The reality is that if the data on the disk really is bad, then there is nothing you can do about it. The same result would occur with the real floppy in a real computer. Testing the image afterwards is all that I do to make sure that it works correctly. Tools like my flux display and DrCoolZic's tool can show you when the data being read is bad due to a dirty head or dirty disk. I have only seen cases where the disks were dirty causing a read issue, and its pretty obvious when you look at the flux display... the flux looks "smeared". I have never seen a case where the disk was physically defective - in fact, I had to poke a hole in a 5.25" disk just to make sure bad media could really be seen. I have thousands and thousands of Amiga and C64 disks - and I have not had a single "bad" disk from media failure yet. That's not to say that it can never happen... that's just my experience.

If you have a bad disk, no type of higher level checking is going to fix that - but it could pop up a window telling you "sorry about your bad disk". :)

The notion of re-reading a disk over and over to "finally" get good data is really not accurate. The bits don't magically jump around and change. What does change though is debris on the disk and/or the drive head. So, if the disk and head are clean, then the data will be the exact same every time you read it. I also do not see any difference in gauss strength between a disk formatted 30 years ago and one formatted today, so the bits don't apparently weaken over time either.

By the way, the sole reason for creating SuperCard Pro was for WRITING to real disks. This has always been the biggest request to me over the last few years from the CBM community. I finally looked into the issue and created SuperCard Pro as a result. So, writing to a real floppy is the most important aspect of this product, because that is what people want.
I am the flux ninja

User avatar
DrCoolZic
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2144
Joined: Mon Oct 03, 2005 7:03 pm
Location: France
Contact:

Re: SuperCard Pro

Postby DrCoolZic » Wed Jan 01, 2014 6:39 pm

npomarede wrote:
DrCoolZic wrote:
npomarede wrote:But the same could be said about STX ; file format needed modifications to handle fuzzy bits and bytes' duration change
Why? I think it can handle all we need?

What I mean is that Ijor certainly released several pasti.dll version before reaching the current version, adding extra structures to the original STX format to handle more exotic protections. So, it would be the same if IPF needed to be modified to add new formats that no one encounter so far since IPF is not really a big player in the Atari ST emulation so far.

oops my mistake I though you were saying "file format needs modification..." :oops:

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2387
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: SuperCard Pro

Postby AtariZoll » Wed Jan 01, 2014 8:15 pm

Of course, I did not mean some miracle - to fix defective floppies :D But information about possible bad data is very useful.
I have some interesting experiences with 'dirt' on floppy surface: some floppies produced good images only after 4-5 imagings. And they were on clean, dry place for many years. Some kind of thin film was on surface, what went away after some head passes. Likely, it is vapourization of case plastic .So, nothing about magical bit changes, just some sort of 'auto' cleaning.
I have have some originals which are unreadable now.
I'm not against GMO, I'm against that children play with fire.

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Wed Jan 01, 2014 8:41 pm

Try cleaning the disks that won't read using 91%+ alcohol and cotton swab. You can build "cleaning stations" for 3.5" and 5.25" disks by using old drives and removing the heads. Wire them so the motor spins when powered up. The 3.5" drive is a bit more tricky because you need access to both sides. The 5.25" disks are easier because you can just flip them over.

You use a cotton swab saturated in alcohol. Spin the disk and using just the weight of the cotton swab, drag it back and forth across the media surface. Keep inspecting the cotton swab and rotate it to a clean spot. Once you get the cotton swab so it is no longer changing color, the disk surface will be clean. Don't apply a bunch of pressure, it really won't help speed up the process and you can likely distort the plastic film that the oxide is on, definitely causing problems.

Also, clean the heads frequently. This may require some vigorous cleaning too because when a disk spins it generates friction on the disk, causing heat which can actually "bake" the debris on the surface of the head. The debris has to be dissolved and scrubbed off of the head.
I am the flux ninja

Hippy Dave
Atari Super Hero
Atari Super Hero
Posts: 515
Joined: Sat Jan 10, 2009 5:40 am

Re: SuperCard Pro

Postby Hippy Dave » Wed Jan 01, 2014 11:31 pm

Hi,

I have noticed the Supercard and Kryofluxcard (KFC) connect to a
34-pin Floppy Drive connector.

Note that the magnetic flux on the Floppy Disk is read, amplified,
filtered, differentiated, comparatorized, and smacked twice in a row
with a pulse generator before being send down a long cable to the KFC.

See page 1 and 8 of the following MC3470 data sheet:
http://www.gme.cz/img/cache/doc/433/033 ... heet-1.pdf

(this circuit is contained within the onboard FDD IC on the SF354,
and all 3.5" FDD ICs -- modern FDD ICs use internal switched
capacitors and *BBDL IIR LPFs)

* Bucket-Brigade-Delay-Line
* Infinite-Impulse-Response
* Low-Pass-Filter

What matters here is that the Boolean data has a Minimum High-time
and Low-time with Limited Bandwidth (limited time domain resolution).
Thus high resolution Stream files can be decimated (lossy compression).
You will probably find that there is a limited set of high times and a
limited set of low times in a given file.

Thus, lossy Huffman compression; separately fudge the high-time and
low-time a little so that they compress better.

Furthermore data from several disk revolutions should compress very
well as compared with previous revolutions (only differences).

Note that the difference of several revolutions can give some hints
to guess in writing damaged or protected images.

Note that data can be written faster than it can be properly read
(MC3469 FDD Write Controller).

So save the Stream files, they will be compressed some day,
and they will be read/write also.

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Thu Jan 02, 2014 3:03 am

You can't get away with lossy compression if you want the EXACT same data as what was read. If you don't care about that, then sure, you could just decode the data directly to MFM at that point and cut out a ton of data... but you sacrifice densitity detection, weakbits, strongbits, etc. You could use a modified RLE with the flux data. However, the files aren't THAT big to begin with, so storage is not an issue. If you start adding even simple compression it really takes away from the simplicity (and speed) of a pure flux image. I think speed might be the biggest concern for emulations where the floppy drive emulation is happening in the background of the emulation.

With flux data there is an identical number of high times and low times. Flux transitions are measured from the high to low transition of each pulse. There is a low to high transition that occurs, and that width varies from drive to drive. So, a flux stream is like this:

P = flux transition length
X = sub pulse (typically 500ns-1000ns)
You do not have the required permissions to view the files attached to this post.
Last edited by JimDrew on Thu Jan 02, 2014 9:48 pm, edited 1 time in total.
I am the flux ninja

Hippy Dave
Atari Super Hero
Atari Super Hero
Posts: 515
Joined: Sat Jan 10, 2009 5:40 am

Re: SuperCard Pro

Postby Hippy Dave » Thu Jan 02, 2014 4:33 am

:D
Last edited by Hippy Dave on Thu Jan 02, 2014 4:54 am, edited 1 time in total.

Hippy Dave
Atari Super Hero
Atari Super Hero
Posts: 515
Joined: Sat Jan 10, 2009 5:40 am

Re: SuperCard Pro

Postby Hippy Dave » Thu Jan 02, 2014 4:53 am

The data on the floppy drive Read Data line (pin 30) is not pure flux information.
For example the X = sub pulse period (130 - 1100 nS) when written will be read
back as no shorter than 400 nS +/- 20% otherwise it is rejected as too short
before it shows up on pin 30.

The data has digital high time and low time. I assume the data stream files
consist of sequential high time and low time counts. Intelligently adjusting
these time counts +/- a small amount to make a file compress better will
not make much difference. The objective is to compress the stream file
within the inherent system tolerances, such that it is indistinguishable from
data collected by different floppy drives... nothing more, nothing less.

JimDrew wrote:...
With flux data there is an identical number of high times and low times...
There is a low to high transition that occurs, and that width varies from drive to drive.
...

These will compress well because they are repeated. Sample one and paste it
everywhere it is needed.

Here is another excellent data sheet:
http://www.mirrorservice.org/sites/www. ... ystems.pdf
See pages 3 and 10.

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Thu Jan 02, 2014 5:45 am

The data on the RDData line is the flux reversal duration. Pulses can be anywhere from a few hundred nanoseconds to several milliseconds. Every drive treats certain types of reversals differently. Depending on the data, you could have thousands of different flux lengths. The sub pulse is just for resetting the pulse and is not used. It's the time between high to low transitions that is measured, with the low to high pulse completely ignored.

I am sure that you could try to encode the flux into fewer bits and pack the data, but there is no need and this would kill the speed advantage of using flux.

I just realized that my ASCII art is screen width dependent. I will make a picture.
I am the flux ninja

User avatar
DrCoolZic
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2144
Joined: Mon Oct 03, 2005 7:03 pm
Location: France
Contact:

Re: SuperCard Pro

Postby DrCoolZic » Thu Jan 02, 2014 1:51 pm

Back to possible format for saving protections ...

I just realized that what people are asking for: compact format for non protected track, and detailed up to flux level for protected track is just whet Richard Adam (pretty smart guy) has defined more than 20 years ago for the Discovery Cartridge :)

If you are interested read http://info-coach.fr/atari/hardware/dev ... backup.pdf that describe the format
http://info-coach.fr/atari/hardware/devices/dc.php

I have software to read and decode the file...

User avatar
DrCoolZic
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2144
Joined: Mon Oct 03, 2005 7:03 pm
Location: France
Contact:

Re: SuperCard Pro

Postby DrCoolZic » Thu Jan 02, 2014 2:10 pm

JimDrew wrote:Try cleaning the disks that won't read using 91%+ alcohol and cotton swab. You can build "cleaning stations" for 3.5" and 5.25" disks by using old drives and removing the heads. Wire them so the motor spins when powered up. The 3.5" drive is a bit more tricky because you need access to both sides. The 5.25" disks are easier because you can just flip them over.

You use a cotton swab saturated in alcohol. Spin the disk and using just the weight of the cotton swab, drag it back and forth across the media surface. Keep inspecting the cotton swab and rotate it to a clean spot. Once you get the cotton swab so it is no longer changing color, the disk surface will be clean. Don't apply a bunch of pressure, it really won't help speed up the process and you can likely distort the plastic film that the oxide is on, definitely causing problems.

Also, clean the heads frequently. This may require some vigorous cleaning too because when a disk spins it generates friction on the disk, causing heat which can actually "bake" the debris on the surface of the head. The debris has to be dissolved and scrubbed off of the head.

I did not thought cleaning was so important on 3.5 FD?
I have bough the following kit http://www.ebay.fr/itm/Nettoyage-disque ... 19e6e0992f
Do you think this will help?

User avatar
Brume
Red eyes
Red eyes
Posts: 4022
Joined: Mon Apr 22, 2002 10:16 am
Location: France
Contact:

Re: SuperCard Pro

Postby Brume » Thu Jan 02, 2014 2:42 pm

DrCoolZik: yes, it's important.
Also, there's a video on youtube where both dudes explain how to clean Amiga floppy disks. I've tried yesterday and... it works! I manage to recover many old Atari disks, that FastCopy didn't manage to image previously. Of course it doesn't work each time (I'd said I obtained 80 % of good results).
See there:
http://www.youtube.com/watch?v=ZRT05QwHqPE

Btw, I've bought the same kit from the same seller a few days ago ;) I'm waiting for it.

User avatar
DrCoolZic
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2144
Joined: Mon Oct 03, 2005 7:03 pm
Location: France
Contact:

Re: SuperCard Pro

Postby DrCoolZic » Thu Jan 02, 2014 4:07 pm

Brume wrote:DrCoolZik: yes, it's important.
Also, there's a video on youtube where both dudes explain how to clean Amiga floppy disks. I've tried yesterday and... it works! I manage to recover many old Atari disks, that FastCopy didn't manage to image previously. Of course it doesn't work each time (I'd said I obtained 80 % of good results).
See there:
http://www.youtube.com/watch?v=ZRT05QwHqPE

Btw, I've bought the same kit from the same seller a few days ago ;) I'm waiting for it.

Considering that the 3.5 diskette where protected by the "metal cover" I did not thought it would be useful to clean them?
But I knew head had to be cleaned. By the way seems like the cleaning kit is for 15 or so cleaning so I wonder at what frequency the head needs to be clean?

User avatar
DrCoolZic
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2144
Joined: Mon Oct 03, 2005 7:03 pm
Location: France
Contact:

Re: SuperCard Pro

Postby DrCoolZic » Thu Jan 02, 2014 4:08 pm

DrCoolZic wrote:
Brume wrote:DrCoolZik: yes, it's important.
Also, there's a video on youtube where both dudes explain how to clean Amiga floppy disks. I've tried yesterday and... it works! I manage to recover many old Atari disks, that FastCopy didn't manage to image previously. Of course it doesn't work each time (I'd said I obtained 80 % of good results).
See there:
http://www.youtube.com/watch?v=ZRT05QwHqPE

Btw, I've bought the same kit from the same seller a few days ago ;) I'm waiting for it.

Considering that the 3.5 diskette where protected by the "metal cover" I did not thought it would be useful to clean them but watching your video it seems that it works?
But I knew head had to be cleaned. By the way seems like the cleaning kit is for 15 or so cleaning so I wonder at what frequency the head needs to be clean?

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Thu Jan 02, 2014 4:31 pm

That guy was pretty aggressive with the cotton swab! You don't want to warp, stretch, or distort the plastic film that the oxide is on, so be gentle during cleaning!

Mold and mildew are attracted to water molecules and apparently iron as well, and don't forget that iron rusts when exposed to water. So, disks stored in high humidity environments will have a film of rust and mold/mildew on them, even with the metal cover closed. I have some disks here that I got in lots on eBay where you can physically see mold/mildew that have grown on the surface of the disk.
I am the flux ninja

User avatar
dlfrsilver
Atari God
Atari God
Posts: 1269
Joined: Mon Jan 31, 2005 1:41 am
Contact:

Re: SuperCard Pro

Postby dlfrsilver » Thu Jan 02, 2014 7:56 pm

alcohol did the trick, i have cleaned my atari ST educative software from génération 5, and the result is good so far :)

Once i get my Supercard pro, i will generate some dumps of problematic games :) (included one on amiga which has GCR tracks, read back as MFM lol)
Now SPS France representative since the 19th of June 2014. Proud to be an SPS member !

User avatar
Brume
Red eyes
Red eyes
Posts: 4022
Joined: Mon Apr 22, 2002 10:16 am
Location: France
Contact:

Re: SuperCard Pro

Postby Brume » Thu Jan 02, 2014 8:54 pm

dlfrsilver wrote:alcohol did the trick, i have cleaned my atari ST educative software from génération 5, and the result is good so far :)

Once i get my Supercard pro, i will generate some dumps of problematic games :) (included one on amiga which has GCR tracks, read back as MFM lol)


You still have your large collection of ST games?

User avatar
dlfrsilver
Atari God
Atari God
Posts: 1269
Joined: Mon Jan 31, 2005 1:41 am
Contact:

Re: SuperCard Pro

Postby dlfrsilver » Thu Jan 02, 2014 9:19 pm

Brume wrote:
dlfrsilver wrote:alcohol did the trick, i have cleaned my atari ST educative software from génération 5, and the result is good so far :)

Once i get my Supercard pro, i will generate some dumps of problematic games :) (included one on amiga which has GCR tracks, read back as MFM lol)


You still have your large collection of ST games?


No brume, only a few a them remains. The big part of them have burned in the house.
Now SPS France representative since the 19th of June 2014. Proud to be an SPS member !

JimDrew
Atari Super Hero
Atari Super Hero
Posts: 545
Joined: Mon Nov 04, 2013 5:23 pm

Re: SuperCard Pro

Postby JimDrew » Thu Jan 02, 2014 9:52 pm

DrCoolZic wrote:Back to possible format for saving protections ...

I just realized that what people are asking for: compact format for non protected track, and detailed up to flux level for protected track is just whet Richard Adam (pretty smart guy) has defined more than 20 years ago for the Discovery Cartridge :)


The problem here is the same. Some one or some thing has to decide what is non-protected and what is protected. With flux, it does not matter - its the exact same data that is output by a real floppy drive... which is why I can make SCP into a floppy drive emulator by just clocking out a flux file that was clocked in and saved as an image.
I am the flux ninja


Social Media

     

Return to “SuperCard Pro Disk Copier”

Who is online

Users browsing this forum: No registered users and 1 guest