[Freeciv-Dev] Re: (PR#7045) Double trade route revenue

["Arnstein Lindgard" <a-l@xxxxxxx>]

<URL: http://rt.freeciv.org/Ticket/Display.html?id=7045 >

On Fri, 5 Dec 2003 11:59:54 -0800 Jason Short wrote:

> Are you sure doubling it isn't overkill, though?  If you divide by 5 or 
> 6 instead of 8, is that enough to make trade comparable?
> 
> Ideally the trade and production of a vertically-developed civilization 
> should be balanced to that of a horizontally-developed civilization. 
> Tweaking this is pretty hard, I think.  It may depend on the ruleset 
> (terrain production/trade values).

It may be overkill. Testing is required to find a balanced
formula of course. I just tried tripling, and that was too much :)

While tweaking, we should remember that smallpox has a clear
production advantage. So assuming "waste" is not implemented as
default, the veritcal builder needs more money to compensate. Buying
a unit costs 1-4 gold pr shield, depending on how finished it is.
Buying improvements has a flat rate of 2, which is good but not
enough.

Traditionally, the smallpoxer will build many early units and attack
the builder. Also, the time required to build and grow a trade city
is a factor, so the future profit should be greater for the player
forfeiting early units.

Since, after the tweaking, time will be on the builder's side, we get
even stronger incentive for the smallpoxer to attack early! Now, if
we add a way to defend properly, as described in PR#7047, and make it
harder to steal techs without the prerequisites, the warmonger may
have no option but to grow his own vertical cities, or else he'll
loose out in the long run. I think this is the way to combat ICS,
balancing by incentives.

On Fri, 5 Dec 2003 15:57:24 -0800 raven@xxxxxxxxx wrote:

> what's up with the 4
> trade route limit?

It's a balancing tool. Tweaking the number of trade routes is almost
the same as increasing the value of one, but not quite. If
traderoutes became too profitable, one way to counter-balance could
be reducing the number to 3, like in Civ2. The purpose would be
giving incentive to build more large cities. Else you have smallpox
with one big city.

The purpose of increasing the max number of traderoutes, as opposed
to increasing profitability, would be to increase the effort required
for a city to reach max output. That would make sense if we otherwise
tweak too much in favor of the builder, and want to give the warmonger
a bigger time window to decide the game early.

Remember that the expensive "Marketplace" gives a 50 _percent_ bonus
on top of the base trade. Each trade route increase the base trade.

One almost useful strategy currently is to invest lots of money to
complete one specialized trade city quickly. You probably can't afford
another until you have profited from the first.

Now, if profits increase, it almost becomes appropriate to start
sending caravans to your neighbor. Currently, the only correct
strategy is to attack your closest neighbors (one at the time, unless
they're newbies), because you may only win Freeciv by growing faster
through invasion, than everyone else. Low corruption guarantees the
viability of unlimited horizontal expansion.

The remaining problem is that you simply don't have the time/
resources to build and grow your cities. You may only build units
because of the gross (but fun) attack/defense distortion, see
PR#7047.

If trade becomes required to compete, it opens up future possibilites
for sophisticating the game. Embargos, and thereby diplomacy, could
be made critical. Then again, Freeciv would probably never be as good
as Civ3 in this area, so it might be best not to go there and avoid
comparisons.

Cracking Freeciv is just a big mathematical formula, and there's
no reason for the defeatist myth of incurable smallpox to be so
persistent.

At any rate I suggest we aim to implement the syrup game style mode
that the Apolytonites love, since the current situation is the worst
possible: city improvements are a significant part of the code, but
utterly useless. Except maybe a few city walls in some cases.


Arnstein