[TIB] Re: TI-83 Basic Stumper


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

[TIB] Re: TI-83 Basic Stumper






On Tue, 27 Aug 2002, Listar wrote:
> Date: Mon, 26 Aug 2002 21:07:09 +0200 (MEST)
> From: robvanwijk@gmx.net
> Subject: [TIB] Re: TI-83 Basic Stumper
>
> > "That shouldn't be a problem"?!?  You perhaps expect dialogue
> > like the following...
> >
> >   (Player: Hit the orc!)
> >   YOU HIT THE ORC.
> >   (Player: Get the gold!)
> >   LA LA LA LA... SCROLL...
> >   YOU GET THE 85   GP.
> >   (Player: Quit!)
>
> Eehm, yes? What did you have in mind then?
>

The point of that sample dialogue was two-fold:
  (a) The game would have to make sure that two lines of text
      remained on the screen at all times, as you pointed out
      in an earlier message ("That shouldn't be a problem").
      Hence the "LA LA LA LA... SCROLL..." message, which was
      meant as a sarcastic indication that adding extra text
      to the screen for no reason other than to make the
      internals of the game easier is not a Good Idea in
      general.
  (b) The extra spaces in "YOU GET THE 85   GP." were intentional,
      and meant to indicate that all the "Disp tricks" to date
      have involved extra spaces in the output.  I repeat again,
      correct output is more important than... well, anything.
      If you can't show the player the correct output, then why
      show him any output at all?

Using Robert Mohr's idea (sorry, Robert, that I keep harping on
your solution), the dialogue would read (correctly):

  (Player: Hit the orc!)
  YOU HIT THE ORC.
  (Player: Get the gold!)
  YOU GET THE 85 GP.
  (Player: This is cool!  Keep playing!)

> > I agree that Robert Mohr's trick is very original, and not terribly
> > practical in a production game, but I re-iterate that IT SEEMS TO BE
> > THE ONLY CORRECT SOLUTION SUBMITTED THUS FAR.  (I don't really "get"
> > the Omnicalc idea, but it sounds like it involves including a big
>
> Do you happen to use an 83-? Omnicalc is an application, that's a special
> type of assembly program that runs straight from flash rom, so it will only
> work on calcs that have that flash rom (duh), in this case the 83+ / 83+SE.
> There is no easy way to use this function on a standard 83, so let's just
> forget about it.

Fine by me.  And yes, as indicated by the original posting, I do use
a TI-83 exclusively, not a TI-83+.  The 83 has no Flash rom.

> > 27K memories.)  I repeat, ROBERT'S IDEA WORKS!  NOBODY ELSE'S DOES!
>
> Yes, it works, however, it is SLOW. Calculating regressions is not as fast
> as just adding a bunch of numbers. Especially if you're gonna use a lot
> of numbers, this will slow the game down noticeably.

The game will probably only use the number-printing code once or twice
(at most) per game turn, and in a roguelike game, the length of a turn
depends on the player's input.  That is, Rogue is more like chess or
checkers than Doom or Asteroids; the only limit on processing time is
the player's patience.  I hope that made sense.


> The only "problem" I see with Romman's solution is the extra space, but
> I fail to see how a quick, short piece of code with one slight imperfection
> would be worse then a rather slow, long piece of code, that uses a score
> of extra variables.

I've said this above, and I'll say it again: good programs produce correct
output.  Your programs produce wrong output.  Give me a correct program
and I'll give you a candy bar.  Or something.

> Could you please tell exactly what your problem with our solutions is,
> because I'm not sure I get it.

The "Disp tricks" require the player to subordinate his sense of
aesthetics to the programmer's laziness.  That's a Bad Thing.

Hope that helped (sigh),
 -Arthur




Follow-Ups: