Re: A83: Re: Assembly-83 Digest V1 #842

[Prev][Next][Index][Thread]

Re: A83: Re: Assembly-83 Digest V1 #842


> > any shell (which is, you have to admit, the only fitting environment on
>       ^^^^^                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > TI-calculator to write ASM programs in). So if you change data in your
>   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
> Bzzzt! I've never written any shell-based programs for the TI-83. Never
felt
> the need for a shell, not even to use one. I know some people don't agree,
> but this is my opinion, and the fact that I have that opinion nullifies
your
> statement.

Okay I agree this was a


> > - Instead of using BIT 0,A use RRA or RRCA
>
> Might be useful at times, but it trashes a.

Often you only need to test. Especially on super-fast interruptroutines.
Okay you don't need those on the TI I think but it's still a valid example.


> > - Use EX AF,AF' and EXX!!! Only 4 T-states while PUSH and POP are 11 and
> > 10!!!
>
> And crash on every interrupt. The TI-83 interrupt handler uses af', bc',
de'
> and hl'. I wouldn't recommend trusting that they stay the same if you
change
> them from your main program.

Yes??? WHAT A GOOFY INTERRUPT!!!
One can't use IY as well... what a stupid routines TI has made!

Can I COMPLETELY hook the interrupt routines??? I know there is a hook, but
if I bent it, changing it to a plain RET, will it still execute the BIOS
routines? Or are those accessed _through_ this hook?

Would be very convenient.


> > just keep staring at your source for an hour
>
> Agreed! =)

Hehe...


> > 1. programs are 'squished' (that means they are in binary form; each
byte
> > represents the corresponding code: an un'squished' program on the TI
takes
> > at least two bytes for every byte), and
>
> True for zasmload too, and isn't very hard to do yourself if you want to
have
> your own loader.
>
> > 4. shells have nice looks, can run programs very easily (no
> > send(9prgmBLABLA),
>
> Which is easier: prgmGAME
>               or prgmA -> select Game -> [Y=] ?

Who says you have to quit to TI-os everytime? And the prgmGAME thingie needs
an extra file doing the send(9 trick... space is quite important I think.

Okay you've got me SOS also needs space, you're right. But it makes it good
because of the argument you have skipped to comment on... SOS (or ION) only
needs a few bytes to run a program. ZASMLOAD needs as much free mem as the
size of the program itself. TI-OS needs about half of the size of the
program's size but because of the way the program is stored it also takes
twice as much space.


> > and most people instantly dump games NOT written for a
> > shell (I have to confess I am one of them).
>
> I do it the other way around.

A shame!


> > 5. ZASMLOAD is THE program to lock your calculator. Ever tried to run a
> > Basic-program with it? Since I have no ZASMLOAD anymore my calculator
hasn't
> > crashed for... well, a very long time.
>
> Any facts to back this up? I reckon zasmload is one of the most stable
ti-83
> programs there is.

Hmmm, well I have had other experiences.


> > Yuk. Z80 is tha bomb!!! 30 years old but still up and running.
> > PS. can you show me some example of 68k code?
>
>         move.l    4.w,a6
>         suba.l    a1,a1
>         jsr    _LVOFindTask(a6)

Yuk... I prefer Z80.


~Grauw


--
>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<
          email me: laurensh@geocities.com or ICQ: 10196372
             visit the Datax homepage at http://datax.cjb.net/
MSX fair Bussum / MSX Marathon homepage: http://msxfair.cjb.net/
>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<
References: