Real Time Crossfire Mailing List Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CF: What do you want?

On Fri, 28 May 1999, Mark Wedel wrote:

> David Sundqvist wrote:
>  The decision to be an action game or adventure/exploratory gain is certainly a
> major point.  I think that right now, it can move to either direction without
> requiring a rewrite - right now, that split could pretty much be determined by
> the maps.

>  To me, both seem to be viable games.  I don't think an action game would have
> quite as much long term staying power - to me, an action game means going back
> closer to the gauntlet type games - very fast action, more limited option of
> items and spells, and I would probably make a more direct path of the maps.
>  I think the game has been moving towards being more of an adventuring game in
> the past several years.  Most features and maps that have been added have been
> in that flavor.
>  Note that the code is under the GPL, so in theory anyone can grab it and make
> whatever game they want out of it - so no matter what decision is made, that
> doesn't limit what can happen - it just limits what some people (those that
> agree with the direction chosen) will work on.

Yes. IMO, an adventuring game is also more fun to work on for the
developers. As an action game, I'd argue that crossfire is already fairly
complete. With some graphics fixups and some map fixups it would be pretty
much finished, because I dont think it's possible to balance with
reasonable means. The amount of monsters and the speed of the game makes
it too unpredictable to judge how changes affect the gathering of money
and experience; relations and question becomes a minor point in the game.

So, IMO, we should work towards the adventure/exploration type game.

>  I agree that map set needs to be revisited.  A couple years, the mapset was
> pretty much dump and the 'chico' maps were chosen instead (the chico set
> integrated some of the older mapset), on the basis the chico maps were much
> better.  However, I don't think the maps have kept up with code improvement. 
> This isn't the clarification of just bugs/features, but additions to the code
> can change the map in many ways.  New skills or spells can make a map that used
> to be difficult much easier.  If you look at developement, it is new code
> features that tend to get added - the maps tend not to get updated (although the
> cleanup of the city maps recently was a nice change from that.)

Agreed. There are still many maps that lack floors, many that lack
lighting info, etc. If we make a push to change the game we could also
include good guidelines for mapmakers (including a thought out world with 
history, political, racial and religious direction of different areas,
flora, fauna etc (extendable, of course, for those who want to add new
regions)), especially if we start moving towards balancing the game. 

Wether this should be pre-1.0 or post-1.0 I'm not sure yet. The changes
should be well thought out, and as I said, IMO, as an action game
crossfire is pretty stable as is (0.95.3 with the disappearing archetype
patch and the caching fix has been stable for me). Some functionality in
the pipeline could be used in both game types, but that could be
backported if people feel like it.

>  One of the ultimas (or maybe many of the ultimas did that).  However, there
> were still at least a few scales - a town was still only a single space on the
> larger world map.

Yes, and current games are moving towards a single-scale approach. The
issue of performance isnt as important for the server anymore; we should
have a fairly large leeway even to play a large amount of people on a
server these days.

>  I think to do that for crossfire would require some improvements.  In ultima,
> shops were typically just the shopkeeper and some furniture - in crossfire, a
> shop is a large area with stuff on the various squares (ala nethack.)
>  The idea of a shopkeeper that has the stuff is I think nice.  With a better
> scripting language, or perhaps just improvement in the archetypes, the
> shopkeepers could have different wares and different personalities.  The full
> plate might not be available in just any shop - you may have to go to the
> shopkeeper in that other town to get it.  but the shopkeeper in the dwarven city
> may have some superior armor and weapons.

Yes, we should move a bit towards smaller houses; this would be good
anyway if we balance monsters to be a good challange as 3-4 rather than no
challange in hundreds. Realestate would be more 'expensive', but I
dont really think that's a problem, since we can always build up or down.
Remember, for example, that old-city is already virtually a 'town' in
pretty much such a scale. 

If we implement more detailed shopkeepers, we could also figure in such as
racial and religious group liking (see earlier mail), to affect pricing in
shops. A dwarf of the same religion would probably sell cheaper to another
dwarf, altho he might even refuse to sell to an halforc.
>  I think in any case, you still probably want at least 2 scales - an outdoor
> scale and an indoor scale (otherwise, a 40x40 city in the world is basically a
> 40x40 block literally in the world.)  I don't see a larger outdoor scale that
> big an issue - things like furniture don't really show up there, and putting a
> house image beside the road I think is still a nicer touch than wondering upon a
> 20x20 area integrated into the world map.

I'm not sure yet, and I think I'll have to test out the concepts to see
how they work out. If we had a same-scale approach, it would be different.
If it's different good or different bad, I havent decided yet :). You'd
lack the 'overview' aspect you get now, where the outside world is mostly
a transport from one place to another rather than part of gaming, but on
the other hand you might get more of a live-in experience. 
Considering the reasons for rapid transport between cities, they are
mostly that shops dont have what you want and all shops are pretty much
the same, so you can run between all cities and check the shops. Or, you
are looking for some special (or unbalanced) dungeon.

With the shops that could be solved by more specialized shops. If you're
looking for the best armour or weapons there are only some shops you'll
find it at all (but have a good chance of doing so), so you know where to
go. Commonly used things should be more common, like waybreads, and
available in higher quantities, etc. Again, this is part of balancing and
creating a more deeply evolved world.

The outside world would become more of one huge dungeon. Some areas would
be safer like patrolled roads between cities, and you could find
interesting things there too (random travelling merchants? bandits?),
while you might not want to go hitchhiking with your elf character in orc

>  I think a 'map scale' attribute would be useful.  A dimension door on the world
> map should do less than it does on on an indoor scale map, as should many other
> things like missiles and other spells.  Depending on preferance, this scale
> could also increase travel time (but not below some point so the character still
> moves with at least some reasonable speed)

Yes, it's possible, but it's also complexity. As I said, I'm not sure
about the scale concepts yet, but I'll have to test it and see what it
does. There is the problem of large houses in the country side, but if
it's good or bad I dont know yet :).

>  There are a few issues with big maps:
>  1) memory usage.  Probably not as big an issue now, but with how crossfire
> deals with memory and the maps, I am not sure if the OS would actually swap out
> a portion of the map, since the program may access areas of it.
>  2) CPU usage - currently, crossfire animates all objects that are active, so
> having huge maps with potentially huge number of monsters would slow things
> down.  With faster cpu, this may not be as much an issue.  However, what may
> work better is to only process areas of the map - for example, maybe  only stuff
> within 22 spaces of the player (double view area so if the player wanders back
> in, the monsters have had at least a few ticks to move around so things don't
> appear identical to how the player left).  But how to handle a player leaving a
> map is still a problem - you probably don't want to freeze that entire map if
> the player steps out, but if large maps are standard, you probably don't want to
> start processing all the objects on the map.
>  That problem could probably be handled by adding a list of the exit the player
> used and when it was used, so you know what areas of the map to animate.

Yep, this requires some thinking. But I think it's possible to solve.

>  Yeah.  The type of monster to generate could more easily be done by something
> like a treasurelist.  A treasurelist of forest monster types or hill monster
> types or the like.
>  One big thing that was not mentioned would be to increase the viewable area. 
> 11x11 (or is it 13x13) seems downright small - in many cases, you can't even see
> accross a relatively small room.  If a major change is done, then increasing it
> it something larger would probably be in order.  For some areas, let the
> darkness and/or line of sight limit viewable areas.  For other things, the map
> could perhaps have a viewable limit set

Agreed. It's currently 11x11, but I think that wouldnt be too hard to
extend. LOS limiting view area instead (and brightness (by the way, there
was a patch sometime for day/night cycling, what happened to that? With
some time-scale like 24 cf hours per 8 hours realtime it could be pretty
cool. Affect things like generation of undeads, appearance of some sorts 
of NPC characters, shop opening times (altho that should be most of the
time :) and it could be interesting)). Perhaps 16x16 would be a good area
to choose. 
>  I think if we are going to evolve the game, we really need to decide what
> features should this new game have.  If we can identify what we want the game
> like, then we can identify if we can take crossfire there and perhaps what to
> get rid of.  If nothing else, it might at least identify what features are
> important and should be coded soon and what stuff isn't really a big deal and
> not to worry about it.

Definitely agreed.

Best regards,

[you can put yourself on the announcement list only or unsubscribe altogether
by sending an email stating your wishes to ]