[Freeciv-Dev] Re: wonder balance (was: Blitzkrieg patch)

[Reed Meyer <rdm@xxxxxxxxxxxxxx>]

> Yes, but they are only well balanced if experienced players still feel
> they have a choice to make.  For the Pyramids, I feel you have to
> adjust a few settings for this to be the case, but not very much.
> For the Great Library, and non-eternal wonders in general, it
> suffices to slow down research speed, but this also lengthens
> the game considerably.

The Great Library is powerful, but again I feel it is decently balanced,
at least in a standard Civ II game.  Now, if you set aifill to 30 in
freeciv, then YEAH, probably Great Library is too powerful.  Its
definition, in the long term, probably has to be modified.  Maybe to:
"Your civilization automatically gets a tech advance already known to at
least 1/3 of all the other civilizations."  How about that?

> >      (I understand your argument, but I personally still feel that the
> > Pyramids are too important to pass up.
> 
> I used to feel like that, until I met players who could beat me.
> Play a few games on civserver.freeciv.org sometime :)

Of course, I don't ALWAYS build the Pyramids; circumstances change from
game to game, which is what I'm talking about re: game balance and
variety.  Circumstances might demand quick expansion, or a rival might
have too much of a head start building the Pyramids.
     (Incidentally, "reproducing like rabbits" is certainly a powerful
general plan towards winning the game, but that strategy is somewhat
overrated.  It's possible to win by being "civilized".  I attempt a blend
of the two, utilizing one method more than the other based on
circumstances.)

> > [...]  To go strictly by your definition, then yes, I would have to
> > concede that there are Wonders which could make an already-powerful player
> > an instant winner.  But at the SAME token, ANY good decision by the
> > powerful player, not JUST building a Wonder, could be enough to finally
> > guarantee victory [...]
> 
> OK, but if at that point going for the wonder is sufficient, in any
> game with those settings, then the game is basically reduced to a race
> to get that wonder.  This is the problem that plagued generator 2 and 3
> games with Magellan's before 1.11.0.

What I'm saying is in a game with a good "game balance", such a situation
should be exceedingly RARE.  Unless you match up two players which have
almost exactly the SAME strategy and the SAME circumstances have befallen
them over the course of the game, an obvious winner should have emerged
well before that winner builds a Wonder which is "the final straw".  This
means that that Wonder is basically irrelevant, because the winner would
have won ANYWAY, whether that Wonder existed or not; it was just a matter
of time.  The Wonder just helps in getting the game over with quicker. :)

Regarding the problem with generators > 1, I would argue that that
represents a problem inherent in the GENERATORS more so than the Wonder.
When the folks at Microprose playtested Civ II, they didn't utilize world
generators like generators 2 and 3 in freeciv, so the possible game
imbalance due to Magellan's was never seen.  (I haven't played generators
> 1 in 1.11, so I don't know how good they are now, but in 1.10 I
personally disliked those generators, and not just because of Magellan.  I
felt they could've used a lot of improvement.  Maybe that improvement is
there now.)

> OK.  Well, in that case, I think it's more important to reduce the chance
> of a player who already has more than equal chance to turn into an instant
> winner, since that is by far the more common situation.  The wonder should
> be an important strategic option, not a surefire way for the leader to
> seal victory.

Agreed, as long as we understand the subtleties of this requirement
("final straw" elaboration above and in previous posts).

> It depends on the wonder, of course: the Pyramids' importance increases in
> importance with map size and land mass, Magellan's increases with map size
> and decreases with land mass, for the Great Lib I really wouldn't know.

Since it only depends really on number of civs, Great Lib wouldn't depend
on map size except insofar as you would be more likely to put more civs
into a large map than a small one.

> > But, the expansion stage is usually near completion by the time Magellan's
> > is built, right?
> 
> With standard settings and at least 6 players, yes.
> 
> > I guess, then, a question is whether the LIGHTHOUSE (as
> > opposed to Magellan's!) is an incredibly powerful Wonder.
> 
> I feel it is in a position similar to the Pyramids: not worthwhile with
> standard settings, but I've seen it used to great effect in games with
> slightly different settings.  The effectiveness of the Lighthouse probably
> depends greatly on timeout: many-island empires are impossible to defend
> with a low timeout.

Well, as I guess I should have elaborated more, my experience is almost
entirely Civ I / Civ II (not freeciv) and STANDARD settings (or nearly
standard; the main variation besides map size being percent land mass).
I have no idea to what extent game balance breaks down when you play
non-standard settings.  One thing's for sure: Microprose obviously spent a
LOT of time fine-tuning the game balance on Civ II, so I would be quite
hesitant to make significant tweaking to game rules that try to emulate
Civ II, unless you want to completely abandon the Civ II model (which
hopefully will be done eventually, though, if freeciv is to be anything
but a Civ II rip-off).

A quasi-related topic:  someone remarked that the Freeciv AI is pretty
good, better than Civ II's anyway.  I don't think it is, not significantly
anyway, although it's true that the Civ II AI cheats outrageously and the
freeciv AI cheats very little.  Obviously, of course, human opponents are
the best challenge in either game.

Cheers,
---Reed