Complete.Org: Mailing Lists: Archives: freeciv-dev: April 2005:
[Freeciv-Dev] Re: (PR#12905) RFC: normalizing the effects of happiness b
Home

[Freeciv-Dev] Re: (PR#12905) RFC: normalizing the effects of happiness b

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
Subject: [Freeciv-Dev] Re: (PR#12905) RFC: normalizing the effects of happiness buildings
From: "(Eddie Anderson)" <saywhat@xxxxxxxxxxxx>
Date: Fri, 29 Apr 2005 19:10:13 -0700
Reply-to: bugs@xxxxxxxxxxx

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=12905 >

First this:

To whom it may concern:
    I haven't received much mail from the Freeciv-Dev list in the
last 36 hours or so.  However, I received a new "subscribed" msg for
that list in my last packet.  So I assume that means that:

   1) there is a problem with the list, and

   2) the problem is being worked on.

    In the meantime, if you posted a msg (that you intended me to
see) anytime during the last two days or so, then you should assume
that I did not receive it.  Maybe when the Freeciv-Dev list is fully
functional again, any missing posts from the last two days will be
recovered and then sent out to the subscribers.  If so, then I'll
catch up eventually.


"Peter Schaefer" <peter.schaefer@xxxxxxxxx> wrote:
>
><URL: http://bugs.freeciv.org/Ticket/Display.html?id=12905 >
>
>On 4/29/05, (Eddie Anderson) <saywhat@xxxxxxxxxxxx> wrote:
>> 
>> <URL: http://bugs.freeciv.org/Ticket/Display.html?id=12905 >
>> 
>> "Peter Schaefer" <peter.schaefer@xxxxxxxxx> wrote:
>
>>     Even with the change I proposed, unhappiness caused by military
>> units still has an effect - it prevents celebration. 
>
>I find that cheesy - players leading a war will not care much about a
>little trade lost, but a city in disorder commands their attention,
>because you can't buy military units in it and it is good for nothing.
>I also dislike changing celebration into more trade - people don't pay
>that much attention to micromanaging their trade. Also, under this
>model, the celebration will never stop, unlike city growth, which is
>negative feedback limiting celebrations to a few special turns.

    You may be right.  Obviously, I haven't playtested the idea.

    As to celebration not stopping, I don't see a problem with it as
long as the celebration bonus doesn't upset the game balance too
much.  E.g. I once played a game where I built Hanging Gardens (HG)
in a size 6 city (under Monarchy).  IIRC, with just a garrison unit
and a temple I kept that city in continuous celebration until the
discovery of Railroad made HG obsolete.


    There are several interrelated proposals here.  Please let me
try to defend or refute them individually.  As I see it, these are
some of the proposed ideas involved:

1) Eliminate celebration city growth.

2) Convert the "celebration growth" model to a migration model.

3) Replace the celebration growth effect (under Republic and
   Democracy) with something else.

4) Replace the celebration effect under Republic/Democracy with an
   additional 50% trade bonus (on tiles already producing trade)

5) Make "contentment-making" buildings eligible to cause a
   type 3) effect (i.e. make a content citizen happy) when
   appropriate.

6) Consider the effect of change 5) on managing cities.

   a) It makes maintaining order (i.e. preventing disorder) easier.

   b) In wartime, it does *not* (AFAICS) make it easier to satisfy
      the third condition for causing celebration.  The conditions
      needed for celebration are, IIRC:

      1) The city must be size 3 or greater.

      2) At least half of the citizens must be happy.

      3) There must be no unhappy citizens in the city.

      4) Conditions 1-3 must be true for at least "rapturedelay"
         consecutive turns preceding this one.


Ideas 1), 2), 3), and 4):

  1) I think 1) is a good idea with or without ideas 2), 3), or 4).
     I've already implemented 1) in some of my games against the AI
     and I like it.

     But if 1) is the only change that's acceptable, then no code
     need be changed at all.  That's because the existing server cmd
     "set rapturedelay=x" is sufficient to implement it.

  2) To me, 2) seems like a reasonable substitute for 1).  Plus IMO
     it solves some other problems.  Unfortunately, 2) is
     complicated and much harder to implement.

  3) and 4) - I proposed the 50% trade increase for a three reasons:

    a) It's somewhat similar to Monarchy's and Communism's existing
       celebration bonus.

    b) AFAICT the math is the same as that already used for a
       railroad's effect on food and shield production.

    c) I was hoping to narrow the gap in desirability between
       Monarchy and Republic (and between Communism and
       Democracy).  Unfortunately, that idea was ill-conceived.

           The mistake I made was in thinking that Monarchy and
       Communism got a 100% per tile trade bonus during celebration.
       If that was true, then the proposed additional 50% for
       Republic/Democracy's would have been a smaller bonus by
       comparison (0% vs. 100% when not celebrating; 100% vs. 150%
       when celebrating).

           But obviously that's not true.  In fact, Monarchy and
       Communism only get a +1 per tile trade bonus (and only on the
       tiles that are already producing trade).  That's
       significantly different from a 100% bonus on any tile
       producing more than one trade (e.g. wine, gold, any ocean
       tile, etc.).

  (sound of a can of worms opening)

    But what if Monarchy/Communism's bonus *was* 100% per tile?
    Then 50% more for Republic/Democracy's celebration bonus would
    be smaller by comparison.

    In that case, when in celebration mode:

       1) Monarchy/Communism would get 100% per tile trade bonus
          while Republic/Democracy would get 150%.

       2) Monarchy/Communism would still have the martial law
          effect (which may make it easier to maintain the
          celebration mode).

       3) Monarchy/Communism would still *not* have to cope with
          unhappiness caused by military units away from home (which
          also should make it easier to maintain the celebration
          mode).


Ideas 5) and 6):

    The beauty of 5) (as I see it) is that it changes only one
aspect of the code (i.e. allowing the type 3) effect) and lets the
rest of the code work exactly as it does now.  (Unless I'm missing
something.  If so, then please point it out to me.)

    I'll continue this response in a later email.  This is my
seventh post over the last ~24 hours and they're all starting to run
together in my head.  I need to sort them out before continuing.
Sorry.


>Let's just start coding my migration algo, I'll try to help :-P ;-)
>After that migration can replace the city growth part of celebration :)

    I'd like to try your algo.  As I said, I've already dumped
celebration growth from my games.  So having a working migration algo
would be a bonus.  :-)

>> 
>> >So you might get away with 4 steps(city status' ) instead of 5 as they are 
>> >now:
>> >- get initial unrest and determine military unrest
>> >- calculate luxuries and content modifiers and include wonders in these.
>> >(steps)
>> >- set up city for initial unrest
>> >- apply content modifier
>> >- apply luxuries modifier
>> >- apply military unrest starting from lowest content citizen
>> 
>>      Maybe I'm misunderstanding what you wrote, but I don't see how
>> this produces a better result.  In fact, as I read it, this would
>> also render Shakespeare and Bach as impotent as Michelangelo.
>
>I did not suggest putting all wonders into the "content" category, but
>placing some of them into the "luxuries" category(which btw. follows
>your master plan partially). As long as a city has as many happy as
>discontent citizens, it won't revolt, so that allows you to have more
>military units. Wonders of this kind would be more generic and
>scriptable, they would simply have an effect vs unrest and an effect
>vs luxuries. Please forgive me that I haven't worked out all details,
>but I'm sure your plan to convert all to luxuries would also have
>problems. Also, to point that out again, the effects of military law
>just cannot be converted to luxuries, so that your plan to convert all
>to luxuries just can't work.

    OK, I think I see what you meant now - convert the effects of
Temple, Colosseum, and Cathedrals into luxuries at the beginning of
the calculations.  Is that right?

>
>Btw, I think Republic should also have a military law of 1 unit/city.
>
>Greetings,
>   Peter
>
>
    More later.


-Eddie





[Prev in Thread] Current Thread [Next in Thread]