Technoball and all TLK Games Originals

[dlfrsilver]

Hello,

I have a meeting with the technoball, technoball Z, equality and all game programmer in Paris on Thursday july 24th.

Most of those games use a custom protected format, and i will preserve games and source codes for him, and as such i will create extended ADF, WWARP files, and RAW dumps for SPS to preserve

There is prototype materials in them i hope the disks will be still working wish me luck

[Retroplay]

Good luck.

[dlfrsilver]

thanks

[prowler]

Good luck, Denis!

But surely you mean Tuesday July 24th?

[dlfrsilver]

yes absolutely ^^ ! Tuesday !

[clenched]

EAB is fortunate to have Captain dlfrsilver always on the lookout for new stuff. Good luck.

[mr.vince]

Indeed. Don't forget to dump the sources, too, please. We've started putting those into the archive as well.

Good luck!

[dlfrsilver]

Confirmed by a phone call at 12.00 , i meet the guy at 6.00 pm in paris, so everything will be finished at 20.30, the disk images will be made

Here is what he gave me :

Equality (3 disks) :
- Equality V1 (revision feb 19th 1992) in TLK-DOS format (16 sync words + 6332 bytes per track LOL !)

This is the considered LOST release !!!! and it is the full release !

- Equality V2 (revision dec 10th 1993) in amigados format, this is the version we have in HOL

- Equality music modules => Dumped OK

- TLK Music disk => only disk 2 of this demodisk made by The Lady Killers (TLK) for Ubi Soft

- Tecnoball OCS (TLK-DOS) Master disk => Disk OK
- Tecnoball Z (TLK-DOS) Master disk => Disk OK
- Tecnoball AGA Master disk => Disk OK

- TLK-DOS Masterdisk creator/copier => Everything about the disk format is described in the tool, the source in even included !
It looks like a DMA copier, with copper and blitter involved !

Tronic (2 disks) :
- Tronic AGA v0.36 Shareware release (Master disk)

- A game called Vade Retro Alienas (a prototype) => It's a shoot'em up ; This game is using the TLK-DOS format

- A Prototype called Mad Race (Preversion and demo disk) => The Preversion use the TLK-DOS format and the april 1990 version is in dos format

- A game called MENTAT for amiga 1200 (i have sources and the game).

- Ebonstar94, TechnoballZAGA source codes Yataaa !!! Tronic has an error, i must try to salvage the disk.

[dlfrsilver]

Post updated

[dlfrsilver]

Please check this part of the source code, showing the TLK-DOS format :

a disk is composed (under amigados)
----------------------------------------

of 80 cylinders (0 to 79) with 2 sides each
or of 160 tracks (0 to 159) with 11 sectors each
or of 1760 blocks (0 to 1759)


taille d'une piste (disposition non codé : codé mfm
------------------- (sous amigados) :
:
block header...................... 32*11 : 32*11*2
block data ...................... 512*11 : 512*11*2
track interval approx. ........ 348*01 : 348*01*2
:
----------:----------
6332 : 12664

block header scheme
------------------------

bytes before synchro $0000 : $aaaaaaaa
syncmark : $44894489
indentificator dos ($ff), tracl n°,
n° sector, nber sector remain. $ff?????? : $????????????????
unused $00000000 : $aaaaaaaaaaaaaaaa
unused $00000000 : $aaaaaaaaaaaaaaaa
unused $00000000 : $aaaaaaaaaaaaaaaa
unused $00000000 : $aaaaaaaaaaaaaaaa
auto-contr. block header $???????? : $????????????????
auto-contr. block of data $???????? : $????????????????

- - - - - - -

In a block header 2 long words following each other mfm coded
are resulting an uncoded long word.
In a data block 2 long words mfm coded of the same sector located
at 512 bytes from each other are resulting in an uncoded long word

- - - - - - -

- note about tlkdos 2.0 disks -

by creating disks 100 % non-dos, my goal was already to gain
space, then reading speed access, protection too BUT mainly
against lamers, an experimented pirate will crack my disks
quite easily...

A tlkdos formatted disk is made of 80 cylinders, so 160 tracks
BUT NO SECTORS.

tlk-disk size is varying from 984,38 kb all MFM
to 1687,50 kb all gcr coded (not yet operational)

a track containt : - 12600 bytes MFM or GCR coded
- 4 bytes for the syncword
- 62 bytes of track interval (GAP ?)

Reading only appears when the sync word is detected, which can vary for each track.

(indeed the case, i found in the source 16 different syncs)

2 words MFMF coded of the same track separed by n word(s) result in
an uncoded word. n can take the 36 values which follows :

6300 3150 2100 1260 1050 0900 0700 0630 0450 0420 0350 0300 0252
0210 0180 0150 0140 0126 0100 0090 0084 0070 0060 0050 0042 0036
0030 0028 0020 0018 0014 0012 0010 0006 0004 0002


* once decoded, a track contain

- in mfm, 6300 bytes
- in gcr, 10080 bytes

The track is decomposing as such :
s+0 : identificator tlkdos idtlk
s+2 : n° track complement to 1 $????
s+4 : lowest word checksum $????
s+6 : strongest word checksum $????

n+0 : identificator free $0670
n+2 : identificator free $0670
n+4 : identificator free $0670
n+6 : nber bytes/track $????
n+8 : nber bytes/track $????

n+0 : identificator prg idprg
n+2 : name $???????? (4 chars maxi)
n+6 : end flag $0000/$ffff if last track
n+8 : size $????

[DeafDaz]

Any chance u can upload them to the zone? (adf)

[dlfrsilver]

Not yet, the original are a pain in the ass to image :S

[DeafDaz]

will look forward to it when u do mange m8

[prowler]

Thanks for taking on this task, Denis. And good luck!

[dlfrsilver]

i have imaged all the amigados disks. The problem remains with the TLK-DOS format.
i'm actually seeing the problem with the coder

[prowler]

Good work.

Please post more on the technical details of the TLK-DOS format and the problems of imaging those disks as you solve them (if the programmer is happy for you to do so). It's all very interesting.

[kaffer]

Did you dump with kryoflux?

[dlfrsilver]

Yes i dumped them !

[lesta_smsc]

BTW, and excuse my ignorance, but if these are the original developers of the material, why did they also have protected disks lol - or why can't they remove them?

[dlfrsilver]

The coder doesn't have anymore the original tecnoball and equality source codes.
He gave me the masterdisks of the games, for which i'm doing disks images and also SPS dumps

anyway the games will need to be cracked, as those never were back in the day.
Not counting the unreleased ones and prototypes.

Stingray will get the images to whdload install them and also crack them if he wants.

[kaffer]

It's a straightforward disk format, the only twist is that the sync word and MFM-block size varies by track, according to a lookup table. Once you have the table, it's easy enough.

[kaffer]

The sync words used include some with consecutive 1s, such as 2aad and 2a8b. Bad idea, as the consecutive flux transitions affect each other during reading, weakening and pushing apart the induced voltage peaks. Particularly on tracks closer to the spindle, a spurious 0 can end up shifted into the bitstream, between the 1s.

[mr.vince]

Dennis, did you put them on the FTP?

[dlfrsilver]

I have not put them on the ftp yet, because i spent most of the week to image all the disk from Bruno Ethvignot's (A2000 quarterback archive hard drive, A1200 QB hard drive archive, etc etc. He gave all his amiga disks to me

But yes, i have dumped each original masterdisks in kryoflux stream format and will upload them very soon

EDIT : it's now done

[kamelito]

I remember having met of of TLK games guys, not the 68k coder, he told me that it took ages to develop the ball simulation. He though it would have been a lot faster to do the game in C. This guy then went developing game for PC.
Kamelito

[kaffer]

Originally Posted by kaffer

The sync words used include some with consecutive 1s, such as 2aad and 2a8b. Bad idea, as the consecutive flux transitions affect each other during reading, weakening and pushing apart the induced voltage peaks. Particularly on tracks closer to the spindle, a spurious 0 can end up shifted into the bitstream, between the 1s.

Interesting that I found some old-style Copylock tracks have the same problem (sync marks 65xx). They seem just about impossible to write back with Kryoflux well enough to pass the Copylock check. A re-dump and -analysis always finds some sector sync marks undetected, including sector #4 (#4-6 are checked by the Copylock routine, sync missing on one of those means endless retry and protection check hangs).