Analysis of submitted games

Post all your Kryoflux related topics in here. From questions about the hardware through to disks you've managed to image up and, probably most importantly, write back without any problems :)

Moderators: DrCoolZic, mr.vince, Moderator Team

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

Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 8:32 am

In this thread I would like to provide analysis with Aufit of games submitted to me.

Yogi's Great Escape (Working)
First a picture of the imaging
Yogi's Great Escape (Kryoflux display).png

The N indicates non standard tracks with 5 * 1024 sectors.
The red square indicates KF failure to read the track. For some reason the track could not be imaged a no raw file is output
Yogi's Great Escape-T01.0.png

The quality of the dump is good and no hard protection found. T01-T55 uses 5 sectors of 1024 bytes ans other tracks uses 9 sectors of 512 bytes.
Any good copy program can be used to copy this game.
Side 1 is not used by Atari. Could be Amiga tracks? Not unformatted but not usable under Atari
conversion to stx seems OK
See also viewtopic.php?f=95&t=26160
You do not have the required permissions to view the files attached to this post.

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 9:53 am

Atari 520STE Discovery Xtra DE2 - Final Fight (Program Disk) - Not working
Here is the KF UI Display
Atari 520STE Discovery Xtra DE2 - Final Fight (Program Disk) KF display.PNG

Here you can see that Track 01.1 has a buffering error so the raw file for this track is only few bytes
This indicates that you have a problem with your hardware setup could be one of the following:
- your KF board is connected to your PC through an USB multi port concentrator. You SHOULD not do that connect directly
- your PC is too slow or too busy. It is important that your PC is able to decode the stream protocol from the KF board correctly. I had this problem on a very old very slow PC
- check also that the drivers are correctly installed

In this case the stx file generated does not have any information for Track 01.1 and you should get the following error with Pasti
Atari 520STE Discovery Xtra DE2 - Final Fight (Program Disk) Pasti Error.PNG


Solution correct your HW setup
You do not have the required permissions to view the files attached to this post.

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 10:08 am

Atari 520STE Turbo Pack Games Disk 6 - Impossible Mission II & Human Killing Machine - Not Working
Here is the KF UI display
Atari 520STE Turbo Pack Games Disk 6 - Impossible Mission II & Human Killing Machine KFUI.PNG

Here you have error on T13.0 T57.0 T75.0 Bad stream position. This indicates probably invalid stream protocol
This is probably the same problme as above.
YOU SHOULD CHECK YOU HW SETUP
Talk to SPS people for more info

Track 13.1 to 17.1 Read failed: either the disk is bad or could be related to your HW setup
You do not have the required permissions to view the files attached to this post.

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 10:34 am

Audio Sculpture (Main & Data Disk)\Audio Sculpture (Main Disk)
Track 00.0 Has some protections:
Sector 66 with no data at beginning of track and duplicate sector 66 at end of track. This is a standard protection no problem for Aufit
Audio Sculpture (Main Disk) T00.0.PNG

Track 01.0 CRC error on sector 11. This disk uses 11 sectros per track so it is very close to track limit. KFUI also report read failed
KF has not been able to read correctly the sector either the disk is dirty or not readable so there is nothing can do about this problem because all 5 revolutions are read incorrectly

Track 04.0 KFUI says Good+modified. This seems to indicate that your disk in not an original but a copy
and indeed if you zoom at end of sector 3 you will see that there is a glich at end of sector in the sector splice area. This indicates that most probably the track has been written and after sector write could have been used?
Audio Sculpture (Main Disk) T04.0-zoom.PNG

This is more informational ...

But in summary Track 01.0 and track 03.0 have CRC errors for no good reason (known protection) and therefore your image is bad
Try to clean disk to remove CRC errors
You do not have the required permissions to view the files attached to this post.

User avatar
Mug UK
Administrator
Administrator
Posts: 11179
Joined: Thu Apr 29, 2004 7:16 pm
Location: Stockport (UK)
Contact:

Re: Analysis of submitted games

Postby Mug UK » Thu Feb 27, 2014 12:25 pm

I used my originals of Audio Sculpture but all of the above except Yogi were done last year (same HW setup) before I moved house.

Once I find the disks again I will reimage them for further testing.

Is there anything I can do to improve the Yogi image (even though it works)?

Sent by my Note III
My main site: http://www.mug-uk.co.uk - slowly digging up the bits from my past (and re-working a few): Atari ST, Sega 8-bit (game hacks) and NDS (Music ripping guide).

I develop a free Word (for Windows) add-in that's available for Word 2007 upwards. It's a fix-it toolbox that will allow power Word users to fix document errors. You can find it at: http://www.mikestoolbox.co.uk

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 1:47 pm

Mug UK wrote:I used my originals of Audio Sculpture but all of the above except Yogi were done last year (same HW setup) before I moved house.

Once I find the disks again I will reimage them for further testing.

Is there anything I can do to improve the Yogi image (even though it works)?

Sent by my Note III

Yogi is good it has no problem.
but the other two needs to be reimaged.

Falcon game to come next .... very interesting ;)

User avatar
Mug UK
Administrator
Administrator
Posts: 11179
Joined: Thu Apr 29, 2004 7:16 pm
Location: Stockport (UK)
Contact:

Re: Analysis of submitted games

Postby Mug UK » Thu Feb 27, 2014 1:57 pm

Ok. I have found two sets of Falcon disks so I can do the other set if necessary :-)

Sent by my Note III
My main site: http://www.mug-uk.co.uk - slowly digging up the bits from my past (and re-working a few): Atari ST, Sega 8-bit (game hacks) and NDS (Music ripping guide).

I develop a free Word (for Windows) add-in that's available for Word 2007 upwards. It's a fix-it toolbox that will allow power Word users to fix document errors. You can find it at: http://www.mikestoolbox.co.uk

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 2:25 pm

Falcon from Spectrum Holobyte (2 disk)

This is a very interesting case
The first possible protection is on track T79.0. This track is a short track with a clock consistently pushed to 2100µS resulting in a track with only 5970 bytes.

But here is the interesting part: all the tracks on Side one of the FD seems to be unformatted as there is a no sector present on any of the tracks.
There is "bug" in Aufit: when none of the tracks on side 1 contains sector then the FD is considered as single sided ... but apprently for this game it is not the case ...

If you try to load the game with Steem using Pasti. The Pasti code warn you that you are trying to read a non imaged track 00.1
This indicates that the program checks the content of this track. If you use a program like my Panzer program you will see that in that case Pasti returns a track full of 00 and so the program loads normally
Note: to make it work you need to place the prog file in drive A and the data file in drive B

If we look at track 00.1 we can see that the track is not really unformatted but quite strange.
falcon-t00.1.PNG

This image has been generated with Aufit 0.4b and the this track is displayed in blue: this indicates that the track is not unformatted (as computed by the entropy) and do contains sync bytes but however does not contain any sector.
This version of Aufit also allow you to force Double Sided FD. So if you check this option it writes the track information for tracks of the second side (so you do not get the warning in Pasti)
But the cure is worst than the problem because now the program fails to load.
The reason is that the reading of the bytes before the sync is somewhat random ... and unfortunately the first revolution does not decode the first bytes as zero ...
So you have to select the second revolution and generate the stx file with data from this revolution ... then it works

Of course this a not acceptable use model and I need to think of a way to recognize the situation automatically so user do not have to do anything.
The mechanism has to be general and not hard coded to this specific situation .... so it is not easy to consider all parameters ...
You do not have the required permissions to view the files attached to this post.

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 3:43 pm

Hades HW setup problem = buffering error
Super Cycle several read track errors

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2760
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: Analysis of submitted games

Postby AtariZoll » Thu Feb 27, 2014 3:45 pm

There are at least 3 diverse Falcon Spectrum Holobyte releases. Please specify exact version, if possible.
And all what I see at AM are listed to have single sided floppies.
English language is like bad boss on workplace: it expecting from you to strictly follow all, numerous rules, but self bending rules as much likes :mrgreen:

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 3:53 pm

AtariZoll wrote:There are at least 3 diverse Falcon Spectrum Holobyte releases. Please specify exact version, if possible.
And all what I see at AM are listed to have single sided floppies.

I do not know what version I do not own the FD
This might be the trick it looks like the disk are single sided but in reality the program reads track 0 side 1
You do not have the required permissions to view the files attached to this post.

User avatar
dlfrsilver
Atari God
Atari God
Posts: 1398
Joined: Mon Jan 31, 2005 1:41 am
Contact:

Re: Analysis of submitted games

Postby dlfrsilver » Thu Feb 27, 2014 4:28 pm

DrCoolZic wrote:
AtariZoll wrote:There are at least 3 diverse Falcon Spectrum Holobyte releases. Please specify exact version, if possible.
And all what I see at AM are listed to have single sided floppies.

I do not know what version I do not own the FD
This might be the trick it looks like the disk are single sided but in reality the program reads track 0 side 1


Blue color in CTA means Copylock track, if i remember well.
Now SPS France representative since the 19th of June 2014. Proud to be an SPS member !

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 4:36 pm

dlfrsilver wrote:
DrCoolZic wrote:
AtariZoll wrote:There are at least 3 diverse Falcon Spectrum Holobyte releases. Please specify exact version, if possible.
And all what I see at AM are listed to have single sided floppies.

I do not know what version I do not own the FD
This might be the trick it looks like the disk are single sided but in reality the program reads track 0 side 1


Blue color in CTA means Copylock track, if i remember well.

New coloring scheme with 0.4b still experimental

Code: Select all

Once the FD image is loaded the information for the first track of the disk is displayed in the different graphs. The program also displays two track grids composed of selection buttons: one grid for each side of the floppy. At this stage a quick analysis of the track content is done and is used to determine the background color of the buttons:
•   light green for normal tracks,
•   light gray for unformatted tracks, and
As soon as a track is fully analyzed the background color will turn to:
•   Dark green for normal tracks,
•   Red for protected tracks,
•   Blue for track with no sector that has data that can be read (3+ sync),
•   Yellow for track with possible data but hard to read,
•   Dark gray for unformatted/empty tracks

Also Shannon entropy is used. the number indicates how many bytes to code all values in the histogram. For example 4 means all values can be coded on 4 bits and 9 means all values can be coded using 9 bits.
If value is low then track is "consistent" if value is high track is probably unformatted. This is cheap way of knowing randomness of tracks

IFW
Captain Atari
Captain Atari
Posts: 181
Joined: Fri Jul 22, 2011 4:53 pm

Re: Analysis of submitted games

Postby IFW » Thu Feb 27, 2014 4:50 pm

On Falcon 0.1 it could be simply duplicator info.
The only reason for the game to check 0.1 would be to see automatically if the game is on a single-sided or double-sided disk; if it's readable then it wouldn't require disk 2 or 3 or whatever is written on side 1.

In CTA blue is known protection density (e.g. Copylock, Speedlock etc), except for short or long tracks which use different colour.

IFW
Captain Atari
Captain Atari
Posts: 181
Joined: Fri Jul 22, 2011 4:53 pm

Re: Analysis of submitted games

Postby IFW » Thu Feb 27, 2014 4:58 pm

An image library should not bail on an unused tracks requested (unless for optional debugging); that is pretty normal and happens for games checking the disk type. Instead, unformatted data should be supplied, so checks like these wouldn't fail.

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2760
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: Analysis of submitted games

Postby AtariZoll » Thu Feb 27, 2014 5:16 pm

Yes - if there is no user data on side B, then is really pointless to put there some protection - lose users with single sided drive. So, most likely, as IFW says it is test, and may be that floppy 2 side B contains floppy 3 side A .
English language is like bad boss on workplace: it expecting from you to strictly follow all, numerous rules, but self bending rules as much likes :mrgreen:

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 5:21 pm

The fact that Pasti report the request for 0.1 indicates that the program is trying to read this track.
If you look at the scaterred chart the track does contains MFM transition even though there is no sector.
If the read track returns a value other than 00/FF on a byte located before first sync (apparently tested several times) the test fails and program do not load.
This seems very similar to what is happening with track 78.0 of barbarian (reported in another thread).
But what makes the protection in Falcon "more hidden" than in Barbarian is that it is located on a side of the FD where all other tracks are unformatted.

Pasti does not choke on request it just kindly tell you (this can be suppressed as an option) that your image probably does not contains all the information required. This is a very nice informative feature :)

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 5:23 pm

AtariZoll wrote:Yes - if there is no user data on side B, then is really pointless to put there some protection - lose users with single sided drive. So, most likely, as IFW says it is test, and may be that floppy 2 side B contains floppy 3 side A .

single sided drive is a myth for Atari users. Might have been on very early 520STf ?
My suspicion is it is there because it is well hidden. At least it fooled me :mrgreen:

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

Re: Analysis of submitted games

Postby DrCoolZic » Thu Feb 27, 2014 6:01 pm

Actually again I am not sure how this track has been generated but the scattered chart and the histogram are very informative
falcon-5rev.PNG

As you can see the flux transition follow a Gauss curve centered on 8µs
Here the scattered chart shows the transitions for the 5 revolutions. As can be seen with the sync line the results are quite different for each revolution. This happen because there are a lot of transition located on the border of the inspection window. But all transitions are centered around 8µs this decode into 100010001000100010001 .... and depending on first sampling this give 1111 2222 4444 or 8888 that decodes into 00 55 AA FF so there is a very chance that you read one of these character before first sync.

I have looked at the way it s decoded by CTA (although track is detected as noise at 58%) in the different revolution and indeed we find the values mentioned above.
I still have to figure what to do in Aufit. My best chance to detect this kind of pattern is to look at MFM rather than decoded data.
You do not have the required permissions to view the files attached to this post.

User avatar
Mug UK
Administrator
Administrator
Posts: 11179
Joined: Thu Apr 29, 2004 7:16 pm
Location: Stockport (UK)
Contact:

Re: Analysis of submitted games

Postby Mug UK » Thu Feb 27, 2014 6:09 pm

I have found a different 3.5" drive. I shall redo Falcon using this drive so you can compare disk images/streams.

Sent by my Note III
My main site: http://www.mug-uk.co.uk - slowly digging up the bits from my past (and re-working a few): Atari ST, Sega 8-bit (game hacks) and NDS (Music ripping guide).

I develop a free Word (for Windows) add-in that's available for Word 2007 upwards. It's a fix-it toolbox that will allow power Word users to fix document errors. You can find it at: http://www.mikestoolbox.co.uk

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2760
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: Analysis of submitted games

Postby AtariZoll » Thu Feb 27, 2014 7:51 pm

I looked supplied STX img. converted from ... It actually tries to load regular sector from side B, via XBIOS call. If there is 'Size' at begin then loads $E004 bytes in RAM, again with XBIOS calls, and there is 10 sector/track on calls.
All in all, I think that it is some special diagnostic thing - likely used only by testing of SW. Commercial release did not have that data - as is visible by lack of sectors on side B.

FalconSidB.png


Period. :D
You do not have the required permissions to view the files attached to this post.
English language is like bad boss on workplace: it expecting from you to strictly follow all, numerous rules, but self bending rules as much likes :mrgreen:

IFW
Captain Atari
Captain Atari
Posts: 181
Joined: Fri Jul 22, 2011 4:53 pm

Re: Analysis of submitted games

Postby IFW » Sat Mar 01, 2014 10:49 am

DrCoolZic wrote:
AtariZoll wrote:Yes - if there is no user data on side B, then is really pointless to put there some protection - lose users with single sided drive. So, most likely, as IFW says it is test, and may be that floppy 2 side B contains floppy 3 side A .

single sided drive is a myth for Atari users. Might have been on very early 520STf ?
My suspicion is it is there because it is well hidden. At least it fooled me :mrgreen:


It was a myth that publishers and Atari itself believed then ;)
Most early games try to squeeze everything on side 0 of a disk, and if that's not possible there is a second disk.
Publishers only value one thing: profit, and adding an extra disk costs money.
So they'd have lost a lot of customers if they only released double-sided games, otherwise it wouldn't have justified the cost of making each side a separate disk.
Unfortunately we don't have sales figures for the initial release of ST, but it had to be quite significant to support single-sided release for such a long time.
e.g. even Dragonflight has sides separated - and uses side 1 to reduce swapping if the drive supports it.

AtariZoll
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2760
Joined: Mon Feb 20, 2012 4:42 pm
Contact:

Re: Analysis of submitted games

Postby AtariZoll » Sat Mar 01, 2014 11:00 am

Absolutely agree with IFW. There was for sure big number of machines with SS drive sold.
Myself bought 520ST with SS too, because it was 200 DEM less than with DD drive. For that money I bought little later DD drive, so at end had 2 drives without extra expense.
I'm not sure that it was good idea to produce machines with SS drive at all. But who knows ... Price was surely big factor in choosing computer.
The consequences we may see: lot of SS floppies. Or very slow loading of Dungeon Master - because they used extra level-2 packing to make it fit on 400K floppy.
English language is like bad boss on workplace: it expecting from you to strictly follow all, numerous rules, but self bending rules as much likes :mrgreen:

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

Re: Analysis of submitted games

Postby DrCoolZic » Sat Mar 01, 2014 11:25 am

I was referring to drives and not to disks but you probably right!

I have about 20 Atari floppy drives (including the one in many Atari machines I own) and there are ALL double sided so I assumed that most machines had double sided drives but I may be wrong 8O

However for publisher making single sided floppy DISKS still make sense as they would be created faster by Trace replicator ...

IFW
Captain Atari
Captain Atari
Posts: 181
Joined: Fri Jul 22, 2011 4:53 pm

Re: Analysis of submitted games

Postby IFW » Sat Mar 01, 2014 1:17 pm

Not much faster, PLUS you'd have to duplicate twice as many disks... adding the drive spinup time etc, plus media cost itself releasing something on 2 disks instead of one was probably a big blow for major publishers financially.
Another thing is the choice of FDC: 1772 is for single-sided drives, I think 1773 or something similar natively supports side selection and checking and was the same price. The point is the double-sided FDC as well as drives were available at the time, it was a cost saving measure to go with single-sided for both and adding a side select separately had to be an afterthought.

"640 kB ought to be enough for anybody" - A single disk side should be enough :lol:


Social Media

     

Return to “Kryoflux”

Who is online

Users browsing this forum: No registered users and 1 guest