Re: Rare game uploaded to Asimov

[Hot Rod <hotrod.apple2@gmail.com>]

On Wednesday, March 6, 2013 1:05:31 PM UTC-6, BLuRry wrote:
> On Wednesday, March 6, 2013 9:09:54 AM UTC-6, Hot Rod wrote:
> 
> > This is interesting.  I've been away from Apple stuff for the last few months, but did take a peek at this.
> 
> 
> > It looks like the disk is formatted as a DOS 3.2 13-sector, but only tracks $00-$0A (or was it $09?  Rats - forgot to jot that down), and is a single-load game.
> 
> 
> > The boot is somewhat interesting; copies itself from $800 down to $200, then loads into $300, eventually winding up at $7700 to do the final load.  Looks like it uses a DOS 3.2 RWTS relocated from $B700 down to $7700.  I haven't spent the time to decode which exact track/sectors are loaded, but it seems like a single load from there.  Afterwards it covers its tracks by erasing the RWTS from memory.  Goes to $3700 to set up things, then bails out to $401 (which in turn goes to $50E0) to start the game (via some stack manipulation and RTS).
> 
> 
> > Anyway, I grabbed the RWTS and used Advanced Demuffin to convert the 13-sector data into DOS 3.3 format, with the thought of putting this into a more convenient single file (and to see what else is on the disk), but I thought I'd mention that I get a read error on track $00, sector $0A.  So there may be a problem with the .nib image.
> 
> 
> > I don't know yet if track $00, sector $0A is used in the load though (and the program does seem to run), so it might be moot.
> 
> 
> > I did a quick scan of the converted disk image data, and I do see the author's name, and what appears to be fragments of source code here and there (and fragments of an Apple DOS Master disk I think).
> 
> 
> > Should be a fairly straightforward project to file this game, which would allow it to be used more conveniently.  If no one else gets to it, I might.
> 
> > Cheers.
> 
> > ]HR
> 
> 
> 
> Yes, please do.  The single-load version I have in Apple Game Server is very hit-or-miss and doesn't seem to always work correctly.  Sometimes sound is disabled, sometimes it doesn't start unless cold-booted.  Very wonky.
> 
> 
> 
> -B

OK, I'll have another look at it.

I did confirm a few more things.  Only tracks $00-$08 are formatted/have data (and are only 13 sectors each) in the .nib image.  I'll assume the original disk like this too.

But also, track 0, sector $0A is clobbered in the .nib image.  There's just the address marker for it (D5 AA B5 FF FE AA AA AF AA AF AA) and then just a small fragment after that before sector 0 data is laid right on top of it.  Hopefully that data isn't used (still need to follow the track/sector list to see what's really loaded).

[It would really be best to run Advanced Demuffin against the original disk to get the data, and hopefully track 0, sector C.  And then use ADT to transfer the converted data cleanly.]

Other than using a standard DOS 3.2 13-sector format, the only change to the RWTS is at $798E (which would be $B98E in a normal DOS 3.2 RWTS load), which changes a LDA #$00 to a LDA #$2F; they seed the checksum byte with the volume number instead of a #$00 (so basically a #$FE instead of a #$00).  That's what prevents a normal DOS 3.2 RWTS from just reading the sectors.  So between the boot and the format, someone put some time into trying to protect it.

The program itself seems pretty small though; most of memory seems to be shape data.

Haven't done a very thorough look at it yet though.

]HR