[crossfire] Summary, was Re: The future of Crossfire

[Nicolas Weeger]

> - Better maps/quests - more on the thinking side, less on the 'you just

Agreed.

> - Remove some of the old/bad maps - this can be harder to determine - what

I'd rather recycle them if possible. We don't have that many map makers :)

As a related note, I'd use the 'lore' field of maps to describe what quests 
they are part, things like that (lore is never displayed except by DMs using 
the dumpmap command). Splitting the author field (as a feature request says) 
could also help, maybe even a special "quest" field. But that's really not 
that an urgent thing.

> - Better NPCs.  NPCs should be more visible - some wander about town, some
> perhaps in the wilderness - as of now, they pretty much all just hang out
> in the tavern, which doesn't add much color

*nods*

> - Better NPC conversations - for many NPC's, trying to figure out what they
> know and getting them to say it can be frustrating since it is specific
> word match and many times players don't know what word

Yann's CFDialog will help, we just need imagination for creating the 
dialogs :)

> - Better NPC interaction - in many cases, giving/taking items is a bit
> confusion, since it is an altar nearby the NPC which is doing the actual
> work, so that if the NPC moves away, or you just can't figure out the right
> space, the transaction doesn't work.

Scripting will help. And/or use a trick (in Lursendis quest, a NPC asks the 
player to put ingredients in a stove. A hook on its close event makes the NPC 
check the contents).

> - Slow down combat - combat is too fast so there is no tactics.  Combat
> should be slowed down so that there are tactics and time to react to
> events.  IMO, crossfire should be an adventure/RPG game more than an action
> game.  Note that slowing down combat will likely have other effects -
> reduce gain of exp, reduce gain of wealth, etc.

Agreed, to the condition we don't become a turn-based game :)
Let's say a reasonable reaction time and planning should save one's life 
often, unless really going in weird places or really unlucky. I'd add stats 
(wc/ac/...) rebalancing, while we're at it :)

> - Add better (some?) sound/music support.  Music files are on the tracker,
> and so map level music probably isn't hard.  The game is missing lots of
> potential sound effects (just things like gates opening), so would be nice
> to be able to add more to give a more immersive feel.

Agreed. We need client support, though.

> - Races & Classes:  Need to be more distinctive.  Perhaps several races
> should be removed (not especially distinctive from one another- was more
> relevant back when race & class was the same thing).  Classes need to be
> redone so that there is more importance in choosing the class you take
> beyond starting skills, so that high level wizards and fighters actually
> look quite a bit different.

Yep.

> - Coherent world - rules should apply that make sense.  A lot of this may
> be in backstory on the wiki - it might be nice to move more of this (or
> copies of this) in game, in books in the library, etc.

Agreed.

> - Add missing documentation - everything about archetypes should be
> documented. Some code cleanup likely in order to make objects behave in a
> consistent fashion.

Agreed, but shouldn't be the priority. Documenting when fixing, yes. 
Documenting just to document, well, rather do something else if you can 
(what? I did document like crazy just for the sake of it? No, you're just 
dreaming ^_-)

> - Better testing - make more use of testing framework.

Harder to do. Writing real meaningful tests takes almost as much patience as 
writing the code to be tested, and is error-prone too.
I'd rather add assert() every place it makes sense.
More important, document the intended behaviour (granted, a unit test is used 
for that, but sometimes it's not worth writing). What are abilities supposed 
to do? What should be the behaviour of fleeing NPCs, especially those that 
have a bow/horn/rod to fire with?



IMO, this should be our medium to long term. Of course if we could do all this 
in a reasonable delay, it could be into CF 2.0, but it shouldn't delay the 
release too much just for that. Better release incremental things if 
possible.


As usual, we also depend on people actually doing the code/work at some 
point :)


Nicolas
-- 
http://nicolas.weeger.free.fr [Petit site d'images, de textes, de code, bref 
de l'aléatoire !]
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mailman.metalforge.org/pipermail/crossfire/attachments/20070606/81acdfb5/attachment.pgp