Re: CF: Are these bugs?

[Mark Wedel <mwedel@scruznet.com>]

David Andrew Michael Noelle wrote:

> >  Word of recall was never meant as an escape spell.  It was meant as a 'I am in
> > the middle of no place and just completed this adventure and what to get back to
> > town'.
> 
>     Oh, okay.  As long as it's intentional.  There were some other games,
> probably diku muds, where I used to use it for escape.  It generally had
> some unpleasant side-effect in those games, though.  A seperate escape
> prayer might not be a bad idea, as long as it did something like dropping
> all your coins and anything that's not inv-locked before yanking you out of
> the dragon's mouth.  Maybe even locked items might have to make a saving
> throw to go with you.  Given a choice between losing stuff and losing
> experience, I'll leave behind as much as I need to.

 Since a player can lock their items, and will lock their best items, ignoring
locked probably doesn't make a sense.  A locked item is really a game
abstraction so when you do a dropall, it ignores those objects - it doesn't mean
that the character is necessarily doing anything special with them.

 As far as get away fast spells, dimension door comes to mind.  You have to be
careful with get away really fast spells - if you can always escape from death,
then that really decreases the challenge quite a bit doesn't it?  Even removing
items might not do much good, since any decent player is likely to have backups
and most of his money stored is his apartment.

> 
>     While I was testing my modified Word of Recall, I noticed something in
> spell_utils.c that I don't understand.  When certain spells are cast by an
> item instead of praying or spellcasting skill, the item's level is not a
> factor in the strength of the spell.  Instead of the item's level,
> SP_level_strength_adjust uses the player's level in the skill used to apply
> the item.  For most, this is zero, since use_magic_item is a neutral skill,
> so the power of the effect would not be increased at all, regardless of the
> item's level.  For scrolls, however, Intelligence skill is used to calculate
> the effect.
>     Is it really supposed to work that way?  What then is the difference
> between a Rod of Large Fireball (level 0) and a Rod of Large Fireball (level
> 42)?

 It isn't supposed to work that way.  The spell stuff needs to be redone quite a
bit - the spell code has been hacked up quite a bit for various enhancements and
the like (for example, a temporary pseudo player object is created because some
spells need a player to work, so this pseudo object is merged with the spell
casting object to work around that - the correct way would be to pass both the
player and spell casting object to the functions that need it)

 Spells are likely to be redone a bit in the 96.x release (make objects taht
describe the spells instead of having the spell tables get compiled in)


>     How about a teleport spell with a command line option?  If the first
> word of the argument is "set", add an invisible exit-type object to the
> player with the name set to the rest of the argument and the destination set
> to the current location.  If the first word is not "set", look for an exit
> in the player's inventory with a name that matches the argument.  Keep a
> special exit named "TeleportBack" that is set every time teleport is cast,
> so that if the argument is missing or not matched, you appear in the place
> you last teleported from.

 The question is how many such objects should be allowed, and how precise should
they be?  One current issue is the fact that it is very easy for players to
visit all the shops in the world hunting for good items very quickly.  If they
could teleport between them, this makes it even easier (a planned fix for the
quick visiting is to increase the world scale - that was also going to be done
for other reasons)


>     Exactly what I was suggesting.  I /think/ it would be enough to abort on
> detection of anything hostile, success of the repeating action (for
> search/disarm) or having your path bloocked (for movement).  It would be
> useless for combat, but that's okay.  "Success" should be defined by
> per-skill stop conditions, like stopping prayer or meditation when you can't
> recharge any more, hiding when you're as hidden as you can get, stealing
> when you've taken everything they have, and so on.  Just common sense
> reasons that would stop the action from doing anything useful anyway.  I
> can't imagine that would take much more than a switch on the skill being
> used (assuming movement has already been handled) and checking certain
> conditions that have already been coded elsewhere.

 The slightly tricky part is that the successful use of the skill needs to
trickle back up (or something does) saying stop doing this command.

 This may get looked at more later, as some commands should take longer than
other (you should not be able to swap armor in one tick, but you can right
now).  If some real time for those actions are added, then some interruption
method is needed there, but it probably needs to be controllable (for example,
if you were attacked and just started taking off your armor, you could argue
that you should stop taking off your armor (you want to wear it for this battle
than take it off), or maybe you really do want to take off that armor is being
attacked by something like a rust monster.)

 Those scenarios start to get trickier.
-
[you can put yourself on the announcement list only or unsubscribe altogether
by sending an email stating your wishes to crossfire-request@ifi.uio.no]