[SD] Re: Venus on all Z80 Ti's
[Prev][Next][Index][Thread]
[SD] Re: Venus on all Z80 Ti's
> Van: Tijl Coosemans <tijl.coosemans@mail.be>
>
> > Why not just include the BASIC code to start the kernel?
>
> It might probably safe some space. On the other hand, having no prgm0
> might be a bit more user friendly: no "freaking" programs on your calc.
Well, the code needed for auto-execution in BASIC is 39 bytes when using
the ASE-startup-header. This in comparsion to just "prgm0", which would
only take 2 bytes. I would think twice before including the BASIC starter
in every program...
Maybe I could adapt the ASE header, then it would be 30 bytes, and Venus-82
would need to be 6 bytes bigger. But as far as I've seen the BASIC program
needs to Stop/Return before the new 'keyhandler' catches on :-(
Have already tried hacking the ASE beta2 backup, and things didn't work
without a Stop/Return...
There might be problem in finding out where the TIOS stores the info about
the calling BASIC prog, but that's just a matter of time...
A drawback on ACE's startup is that your program/shell needs to be at least
733 bytes big (+ VAT-entry + sizebytes). But that's pretty small allready,
I think :-) Could off coarse add vnExplorer to prgm0 too.
btw, the stack-area doesn't come in a backup, so putting Venus-82 there is
a little inconveniant (as Dines stated). On the Ti82 you need to do (more)
things yourself than on the Ti83, so I can use the extra space. Though it
would be a bit oversized for a less than 200 byte shell, on the Ti83. But
on that calc (when using the same hack) the nearest entry is at $9400,
which would make a 214-215 byte program (including the 39 startup-bytes).
Plus, if you start prgm0, you are shure that that it's on the calc (else
you get an "ERR:UNDEFINED"). Just letting the calc 'jump' to a certain
memory area isn't quite safe, I think.
Henk Poley <><
Follow-Ups: