KryoFlux Images of my FD

[DrCoolZic]

I am now ready to make stream images of my Atari FDs

I will maintain here a list of FDs done.
I hope to get the IPF back and I will try to post them somewhere

I now have a preliminary Web page on the KryoFlux device: http://info-coach.fr/atari/hardware/dev ... yoflux.php

I do not have yet a list of all my FD but you can have a look at pictures of about 300 FD I will start with http://info-coach.fr/atari/hardware/dev ... g_to_image

PS. I have limited time so be patient

[DrCoolZic]

Done so far - 8 Sept 2011 - 27 Game/Prog (35 FD)

Code: Select all

Program/Game                    Type     Publisher          Year   Disk  Language Boxed  ProtTxt Problem  Info

Awesome                         Game     Psygnosis          1991   3     English
Back to the Future III          Game     Imageworks         1991   2     English
Balade au Pays de BIG BEN       EducationCoktel Vision      1989   2     French          Yes
Barbarian                       Game     Psygnosis          1987   2     English
Blood Money                     Game     Psygnosis          1989   2     English
Bob Morane - Science Fiction    Game     Infogrames         1987   1     French
Cabal                           Game     Ocean Software     1989   1     English
Carrier Command                 Game     Ubisoft            1990   1     English         Yes
Chaos Engine (The)              Game     Renegade           1993   2     English
Chaos Strike Back & Utility     Game     FTL Games          1989   2     English  Yes
Chart Attack (compilation)      Game     Gremlin            1991   2     English                          Ghouls'n'Ghosts/Venus the Flytrap/Lotus Esprit/James Pond
Chess Simulator                 Game     Infogrames         1990   1     Multi                   YES
Chicago 90                      Game     Ubisoft            1990   1     English
Cisco Heat                      Game     MirrorSoft         1991   1     English
Crack Down                      Game     US Gold            1991   1     English
Crossbow legend of William Tell Game     Screen 7           1989   2     English
DE-50 Editor W/MPE              Music    DrTs               1988   1     English  Yes
DM/CSB Cartographer             Game     Dreampark          ?      1     English  Yes
Dungeon Master                  Game     FTL Games          1988   1     English  Yes
Dungeon Master Editor           Game     Softex             1988   1     English  Yes
* Fighter Bomber                Game     Activision         1989         Multi           Yes
Populous                        Game     Electronic Arts    1989   1     English  Yes
Populous                        Game     Empire Software    1992   1     English
Populous                        Game     Ubisoft            ?      1     English
Theme Park Mystery              Game     MirrorSoft         1990   1     English
Turrican                        Game     Rainbow Arts       1990   1     English


Boxed = complete program/game with box and manuals
ProtTxt = Protection by text requested (from manual or other)
Problem = Yes mean could not make it run on STe with French TOS

[DrCoolZic]

Done so far 44 Games (83 FD)

Code: Select all

Program/Game                    Compil  Type     Publisher          Year   Disk  Language Boxed  ProtTxt Problem  Info

Awesome                                 Game     Psygnosis          1991   3     English
Back to the Future III                  Game     Imageworks         1991   2     English
Balade au Pays de BIG BEN               EducationCoktel Vision      1989   2     French          Yes
Barbarian                               Game     Psygnosis          1987   2     English
Blood Money                             Game     Psygnosis          1989   2     English
Bob Morane - Science Fiction            Game     Infogrames         1987   1     French
Cabal                                   Game     Ocean Software     1989   1     English
Carrier Command                         Game     Ubisoft            1990   1     English         Yes
Chaos Engine (The)                      Game     Renegade           1993   2     English
Chaos Strike Back & Utility             Game     FTL Games          1989   2     English  Yes
Chart Attack (compilation)      Yes     Game     Gremlin            1991   2     English                          Ghouls'n'Ghosts/Venus the Flytrap/Lotus Esprit/James Pond
Chess Simulator                         Game     Infogrames         1990   1     Multi                   YES
Chicago 90                              Game     Ubisoft            1990   1     English
Cisco Heat                              Game     MirrorSoft         1991   1     English
Crack Down                              Game     US Gold            1991   1     English
Crossbow legend of William Tell         Game     Screen 7           1989   2     English
DE-50 Editor W/MPE                      Music    DrTs               1988   1     English  Yes
Deuteros                                Game     Activision         1991   2     Multi
DM/CSB Cartographer                     Game     Dreampark          ?      1     English  Yes
Dragon Ninja                            Game     Imagine            1989   1     English                 YES      Not working on STe French TOS
Dragon Ninja                            Game     Ocean Software     1991   1     English
Dungeon Master                          Game     FTL Games          1988   1     English  Yes
Dungeon Master Editor                   Game     Softex             1988   1     English  Yes
Electronic Pool                         Game     Microdeal          1986   1     English
Elvira                                  Game     Accolade           1990   5     English
Epic                                    Game     Ocean Software     1992   2     English
E-SWAT                                  Game     US Gold            1991   2     English
Explora III                             Game     Infomedia          1990   3     French
Ferrari Formula One                     Game     Ubisoft            1990   2     English
Fighter Bomber                          Game     Activision         1989   2     Multi           Yes
Fighting Soccer                         Game     Ubisoft            1990   1     Multi
Fire & Forget                           Game     Infogrames         1990   1     French
Fondations Waste                        Game     Exocet Software    1988   2     English
Football Manager II                     Game     Ubisoft            1990   1     English
Forgotten Worlds                        Game     US Gold            1989   2     English                 YES     
Games Galore                    Yes     Game     Mandarin Software         2     English                          Skystrike/Yomo/Mouthtrap/Skatetribe
Gauntlet II                             Game     US Gold            1989   2     English                 Yes     
Gazza Soccer Mania                      Game     Prism Leisure      1990   1     English
GhostBusters II                         Game     Activision         1989   4     English
Gods                                    Game     Renegade           1991   2     English
Golden Axe                              Game     US Gold            1991   2     English                 YES     
Guerriers Ninja (Les)           Yes     Game     Virgin                    4     French                           Shinobi/Double Dragon II
Gunship                                 Game     Microprose         1986   2     English
Hostages                                Game     Infogrames         1989   1     English
Populous                                Game     Electronic Arts    1989   1     English  Yes                     Only works on ST US TOS
Populous                                Game     Empire Software    1992   1     English
Populous                                Game     Ubisoft            ?      1     English
Theme Park Mystery                      Game     MirrorSoft         1990   1     English
Turrican                                Game     Rainbow Arts       1990   1     English


[DrCoolZic]

If you own a KryoFlux board and you want to make images of Atari FD I recommend that you use my batch files

I have modified the original doDump.bat batch file provided by SPS. This new batch file is called AtariDump.bat and is tailored for Atari users. It fills for you many default values suitable for Atari and it uses the -i4 switch so you get perfect images of Atari FD. By default the name of the game that you enter is used to create a directory and stream files (the spaces are removed) but you can specify another name if you want. The stream files uses the following name <game_name>(current_disk-max_disk)track.head.raw and a <game_name>.txt file that contain the information you have entered is created in the same directory.
For example if you enter "Maniac Mansion" as the name of the game with two discs a directory called ManicMansion is created and in this directory you will find the following files:
[*] ManiacMansion.txt with all the information you have entered
[*] ManiacMansion(1-1)00.0.raw to ManiacMansion(1-1)83.1.raw for all the tracks of disk 1
[*] ManiacMansion(1-2)00.0.raw to ManiacMansion(1-2)83.1.raw for all the tracks of disk 2

The DTC program display a lot of useful information on the screen but unless you stay stuck in front of your screen all this information is lost! Therefore I have created a second script file called AtariDumpLog.bat that dumps all the messages (both stdout and stderr) in files that you can read later. These files are created in the game directory and have the format <game_name>_dump(current_disk-max_disk).txt

The two batch file are attached and have been tested on more than 70 games

For example with the above example two files will be generated: ManiacMansion_dump(1-2).txt & ManiacMansion_dump(2-2).txt

[DrCoolZic]

Done so far 71 games for a total of 123 floppy diskettes !

Code: Select all

Program/Game                     Compil  Type     Publisher          Year   Disk  Language Boxed  ProtTxt

Awesome                                  Game     Psygnosis          1991   3     English
Back to the Future III                   Game     Imageworks         1991   2     English
Balade au Pays de BIG BEN                EducationCoktel Vision      1989   2     French          Yes
Barbarian                                Game     Psygnosis          1987   2     English
Blood Money                              Game     Psygnosis          1989   2     English
Bob Morane - Science Fiction             Game     Infogrames         1987   1     French
Cabal                                    Game     Ocean Software     1989   1     English
Carrier Command                          Game     Ubisoft            1990   1     English         Yes
Chaos Engine (The)                       Game     Renegade           1993   2     English
Chaos Strike Back & Utility              Game     FTL Games          1989   2     English  Yes
Chart Attack (compilation)       Yes     Game     Gremlin            1991   2     English                          Ghouls'n'Ghosts/Venus the Flytrap/Lotus Esprit/James Pond
Chess Simulator                          Game     Infogrames         1990   1     Multi                   
Chicago 90                               Game     Ubisoft            1990   1     English
Cisco Heat                               Game     MirrorSoft         1991   1     English
Computer Hits Volume 2           Yes     Game     Beau-Jolly         1988   2     English
Crack Down                               Game     US Gold            1991   1     English
Crossbow legend of William Tell          Game     Screen 7           1989   2     English
DE-50 Editor W/MPE                       Music    DrTs               1988   1     English  Yes
Deuteros                                 Game     Activision         1991   2     Multi
DM/CSB Cartographer                      Game     Dreampark                 1     English  Yes
Dragon Ninja                             Game     Imagine            1989   1     English                 
Dragon Ninja                             Game     Ocean Software     1991   1     English
Dungeon Master                           Game     FTL Games          1988   1     English  Yes
Dungeon Master Editor                    Game     Softex             1988   1     English  Yes
Electronic Pool                          Game     Microdeal          1986   1     English
Elvira                                   Game     Accolade           1990   5     English
Epic                                     Game     Ocean Software     1992   2     English
E-SWAT                                   Game     US Gold            1991   2     English
Explora III                              Game     Infomedia          1990   3     French
Ferrari Formula One                      Game     Ubisoft            1990   2     English
Fighter Bomber                           Game     Activision         1989   2     Multi           Yes
Fighting Soccer                          Game     Ubisoft            1990   1     Multi
Fire & Forget                            Game     Infogrames         1990   1     French
Fondations Waste                         Game     Exocet Software    1988   2     English
Football Manager II                      Game     Ubisoft            1990   1     English
Forgotten Worlds                         Game     US Gold            1989   2     English                 
Games Galore                     Yes     Game     Mandarin Software         2     English                          Skystrike/Yomo/Mouthtrap/Skatetribe
Gauntlet II                              Game     US Gold            1989   2     English                 
Gazza Soccer Mania                       Game     Prism Leisure      1990   1     English
GhostBusters II                          Game     Activision         1989   4     English
Gods                                     Game     Renegade           1991   2     English
Golden Axe                               Game     US Gold            1991   2     English                 
Guerriers Ninja (Les)            Yes     Game     Virgin                    4     French                           Shinobi/Double Dragon II
Gunship                                  Game     Microprose         1986   2     English
Hard Drivin                              Game     Tengen                    1     English
Hostages                                 Game     Infogrames         1989   1     English
Hound of Shadow (The)                    Game     Electronic Arts    1989   3     English
Ikari Warriors                           Game     Elite              1987   1     Multi
Indiana Jones & the Fata of Atlantis     Game     Lucasfilm Games    1992   1     English
Indy The Graphic Adventure               Game     Lucasfilm Games    1989   6     English
International Ice Hockey                 Game     Impulze            1992   1     Multi
Iron Lord                                Game     Ubi Soft           1989   3     Multi
Ishar                                    Game     Silmarils          1992   2     English                 
Katate Kid II                            Game     Microdeal          1988   1     English
Kick Off                                 Game     Anco               1989   1     French
Kick Off 2                               Game     Anco               1990   1     English
Kick Off 2                               Game     Empire Software    1992   1     English
Lombard RAC Rally                        Game     Mandarin           1988   2     English
Loom                                     Game     Lucasfilm Games    1990   3     French          Yes
Maniac Mansion                           Game     Lucasfilm Games    1989   2     French
Manoir de Mortevielle                    Game     Lankhor            1988   1     French
Mega Traveller - Zhodani Conspiracy      Game     Empire Software           2     English
Mig 29 Soviet Fighter                    Game     Codemasters        1990   1     English
Mr. Heli                                 Game     Firebird           1987   2     English                 
Night Shift                              Game     US Gold                   2     English
Operation Neptune                        Game     Smash16            1990   1     English
Populous                                 Game     Electronic Arts    1989   1     English  Yes               
Populous                                 Game     Empire Software    1992   1     English
Populous                                 Game     Ubisoft            ?      1     English
Theme Park Mystery                       Game     MirrorSoft         1990   1     English
Turrican                                 Game     Rainbow Arts       1990   1     English

Games: 71      Disks: 123


[Maartau]

... great work !

[DrCoolZic]

not maintained i now have about 200

[Maartau]

not maintained i now have about 200

You gave up your dumps ? Or maybe already have send a couple of them ? (Just for curiosity, some titles interest me, specially compilations) ...

[DrCoolZic]

Actually I was a bit optimistic I have something like 150 and some of them comes from others.
but I have no problem to give anything I have

My problem is that currently my image directory is a mess because I have used a lot of games in KF and SCP format to test Aufit.
I really need to cleanup because I have a mixture of my good/bad dumps done by me or provided as test cases etc ...

I am still hoping that we will have an FTP site where we can store preserved Atari FD.
I am not too fan of putting games on Jim's FTP (retrobackup) because I would prefer a more neutral place.
Also I am trying to push for a PRESERVATION repository where images are preserved with 5 revolutions and validated.
There are currently a lot of images floating around that do not meet preservation requirements as this requires lots of time and knowledge to validate.

[JimDrew]

Retrobackup isn't "Jim's FTP".. it does belong to one of my customers though.

[Brume]

Also I am trying to push for a PRESERVATION repository where images are preserved with 5 revolutions and validated.
There are currently a lot of images floating around that do not meet preservation requirements as this requires lots of time and knowledge to validate.

Does it mean you dump all games with 5 rev with SCP? Even the one that aren't protected?

[DrCoolZic]

Also I am trying to push for a PRESERVATION repository where images are preserved with 5 revolutions and validated.
There are currently a lot of images floating around that do not meet preservation requirements as this requires lots of time and knowledge to validate.

Does it mean you dump all games with 5 rev with SCP? Even the one that aren't protected?

This is a good question

In theory a non protected game can just be saved as an .st file as this would be sufficient for playing it on an emulator and to recreate a FD on an Atari.
However this is not really considered as preservation (at least by Istvan and me) because their may be some information lost in the process.
For example how do you know that the disk you just imaged is an original? You can only find this kind of information by doing a very detail analysis of the disk.
Istvan is currently travelling but hopefully next week I hope he will provide some information on how to test if a disk is an original or a copy.
In the last run of test I did for my protection document I have found some information stored on the FD that are not part of any protection. Most probably this information might have been used by manufacturer to store may be info about the mastering machine ,the company or whatever. Not saving them is probably not relevant for most of people.

I am working a new improved version of Aufit and I hope to be able to add
- quality checking,
- original checking
- auto recovery

As you know Aufit has a very powerful DPLL to read the stream bits. This is especially efficient for non regular rotation speed of a floppy disk (this happen if the disk is not perfectly plane ... difficult to explain). But even with this there are some bits that are read incorrectly. What I have found on several of my floppies is that for some unknown reasons (for me at least) bits are "spreaded" from their original position and if you sample for example 5 revolutions you may end up with these bits read correctly in one revolution and not in the next (this is the auto recovery I am thinking of providing).

Jeff is also doing very detail analysis of the track content and this analysis requires several rotations see viewtopic.php?f=99&t=27185#p262342

So in summary you do not need 5 revolutions specially if the disk is not protected and in good condition, but personally I always sample 5 revolutions (hope Jim will fix the bug that prevents to sample 5 rev on several of my disks) and preferably using SCP *AND* KF so we multiply the chance to get a perfect preservation.
But I am probably too perfectionist

[IFW]

The absolute minimum number of revolutions you need to sample is 2 for commercial disks.
The reason is really simple: on some track(s) data may pass over the index signal. There is no foolproof way of reconstructing such data from a single revolution. Imagine a pattern like this (for simplicity): 1101010101010....11 Now you break this pattern with the index signal. Add index signal jitter (up to 400us on some drives!) and try to tell me exactly how many 01 pairs were there in the sequence in the original data without having multiple revolutions sampled... It's impossible.
Before you say, but no, most disks are purely index synced - funnily enough many of those disks do have one track that is not index synced on purpose - the one that is the copy protection
Another problem as I said is index signal jitter. Usually not so bad with most 3.5 drives, can get very bad with 5.25 drives.
With a single revolution you risk not sampling the entire track if the index signal slightly varies (jitter).

So if you honestly want to make sure (considering the index signal jitter) that you have a whole track sampled index-to-index you want to sample actually 2 revolutions for real - this would take care of whatever jitter your drive has.
But you want to make sure that data over index signal is captured correctly as well, which adds another revolution.

So for safety reasons we are already at 3 revolutions to be sampled as a minimum.
For analysis and to have multiple good reads of the same data sampled continuously 5 revolutions is ideal - having this amount of data makes it possible to extract good data from poorly readable tracks usually, without having to resample it, thus saving a lot of effort.

To summarize: if you want to do preservation use at least 3 revolutions, but if you want to avoid lots of redumps later and want to allow proper analysis of the data for e.g. randomness (weak bits) use 5.
For your own home written disks (strictly speaking of ST disks here!) 1 revolution is enough, for Amiga 2 revolution is the minimum as Amiga tracks are not index synced at all.

[Hippy Dave]

Here is some fun for tracks and index pulses:

Autocorrelation. This is just polynomial multiplication of a number with the mirror of its digits. Thus, polynomially multiply one or two revolutions of data with the reverse of the data, then graph the result to get Blown away.

Try 1111 * 1111 on your calculator (gives same result as polynomial multiplication because of no carry)
notice the middle digit is the greatest.
Now try 12 *21.

Now autocorrelate all the tracks side by side, and take it to the next level.

[IFW]

er... yes that sounds like fun

[DrCoolZic]

Here is some fun for tracks and index pulses:

Autocorrelation. This is just polynomial multiplication of a number with the mirror of its digits. Thus, polynomially multiply one or two revolutions of data with the reverse of the data, then graph the result to get Blown away.

Try 1111 * 1111 on your calculator (gives same result as polynomial multiplication because of no carry)
notice the middle digit is the greatest.
Now try 12 *21.

Now autocorrelate all the tracks side by side, and take it to the next level.

Can you elaborate

From what I understand autocorrelation "is a mathematical tool for finding repeating patterns" (http://en.wikipedia.org/wiki/Autocorrelation)
So this could be useful to find unformatted track and potentially fuzzy bits.

First of all I do not understand your definition "polynomially multiply one or two revolutions of data with the reverse of the data"? Is this autocorrelation?.
What data? The flux values? what is the reverse of data?
Graph the result?
Algorithm?

[Hippy Dave]

Here is some fun for tracks and index pulses:

Autocorrelation. This is just polynomial multiplication of a number with the mirror of its digits. Thus, polynomially multiply one or two revolutions of data with the reverse of the data, then graph the result to get Blown away.

Can you elaborate

So this could be useful to find unformatted track and potentially fuzzy bits.

First of all I do not understand your definition "polynomially multiply one or two revolutions of data with the reverse of the data"? Is this autocorrelation?.
What data? The flux values? what is the reverse of data?
Graph the result?
Algorithm?

Code: Select all

Imagine some track bit-cell data:

345, 234, 326, 152, 221

Now make a new list of reverse data for the autocorrelation:

221, 152, 326, 234, 345

now perform the convolution (polynomial multiplication):

/**** Aside ********************************************************
345(z^0) + 234((z^-1) + 326(z^-2) + 152(z^-3) + 221(z^-4)

221(z^-4) + 152(z^-3) + 326(z^-2) + 234(z^-1) + 345(z^0)

> (z^0) is timezero sample, (z^-1) is the sample before that...
> the + simply sums the terms of the polynomial
  of the form: a(z^0) + b(z^-1) + c(z^-2) + ...
  where: a,b,c,... are the polynomial coefficients. eg. 345, 234,
*********************************************************************/
so:    (polynomial multiplication of the polynomial coefficients)

                         a b c d e
                      *
                         e d c b a
     ----------------------------- 
                    aa ab ac ad ae
   +
                 ba bb bc bd be
   +
           ca cb cc cd ce
   +
        da db dc dd de
   +
     ea eb ec ed ee
     -----------------------------
   = N0+N1+N2+N3+N4+N5+N6+N7+N8+N9 --> plot this data N0, N1, N2 ...

   (note that the data is symmetrical in this case. eg. N0=N9, N1=N8, etc.)

eg.

another example shown with numbers instead of the abc's

     345 234 326 152 221
x    221 152 326 234 345
------------------------
     76245, 104154, N2, ... N7, 104154, 76245

When you plot the data, the histogram will be symmetrical from
left to right, and may give more hints to the track structure.

*** Note that you can perform a correlation of one revolution the
    track with the same revolution (autocorrelation of the same data).

*** Note that you can perform a correlation of one revolution the
    track with a DIFFERENT revolution (autocorrelation of the same track).

*** Note that you can perform a correlation of one revolution the
    track with a TEMPLATE.

*** Note that you can perform a correlation of one revolution the track
    with another track to see how similar (or dissimilar they are)


[DrCoolZic]

First just want to make sure we use the same terms

You talk about

bit-cell data

. I assume you are talking about the flux values read from “decoded stream data” (for example from raw KF or SCP files). With DD MFM these values are generally close to 4000, 6000, or 8000 ns. For example
3620 5993 7990 3953 4037 6034 5951 ...
If things were perfect the exact same series of values would appear again in the second revolution exactly 200 ms later.

You are talking about

a(z^0), b(z^-1), … where (z^0) is timezero sample, (z^-1) is the sample before that... here: a,b,c,... are the polynomial coefficients

Not sure why you say sample "before that" when you actually use the value "after that"?
I assume that a(z^0) indicates that the first sampled value ‘a’ is equal to 3620 not 3620 multiplied by (z^0) ?
So here we have a(z^0)=3620, b(z^1)=5993, c(z^2)=7990 etc.

Now about your

polynomial multiplication.

Usually if we have two series of samples s1=[a b c d e] and s2=[f g h i j] the correlation is defined as
C = af + bg + ch + di + ej (as shown here https://www.youtube.com/watch?v=_r_fDlM0Dx0)
And not the strange polynomial multiplication as you define?

Now back to the purpose of the ‘autocorrelation’. From what I understand correlation allows to express similarity of two time series by one number (normalized version of covariance).
The auto-correlation is defined as the correlation between two parts of one series (for example between two halves of a series) – why called auto – to find out if the series repeats itself. In our case the series of flux transitions repeats itself after one revolution. So for a “normal” track we should get a number close to 1 (not exactly 1 due to variations). For an unformatted track the second series would not be related to the first one so the correlation would be close to 0.

Nowhere I have seen indication of taking a series and correlate with the “reversed” series? This might be useful if you want to show that a series has symmetrical values which is obviously not the cases with a normal track? How can this gives information about the track structure?

But the concept of correlation is interesting with track flux data to find out about unformatted track (no correlation between two revolutions), similar track (correlation between two tracks or track and template). Not sure if this can also be used to find out about fuzzy bits?

From what I understand a similar concept is used in HxC (Jeff are you looking?). A track is sliced in several pieces and similarities of slices between revolutions is made using a sliding pattern. But I do not know if the actual comparison is done using correlation computation or pure value comparison with margin?

interesting reference http://dsp.stackexchange.com/questions/ ... re-similar

[Hippy Dave]

http://www.dspguide.com/
You can read the book on the web page, or purchase it.
This site is a keeper.
BTW. If you follow the directions I gave, you will get the results.

[Jeff_HxC2001]

From what I understand a similar concept is used in HxC (Jeff are you looking?). A track is sliced in several pieces and similarities of slices between revolutions is made using a sliding pattern. But I do not know if the actual comparison is done using correlation computation or pure value comparison with margin?

interesting reference http://dsp.stackexchange.com/questions/ ... re-similar

yes exactly : i do comparaison with value +/- a margin. The timing differences due to the various factor (mainly mechanical...) are compensated by the compare function.
I am not sure if a direct correlation can be used in this application. Some data arrangement/transform may be needed.
I will have a look to the book.