[Freeciv-Dev] Re: (PR#12833) consider_settler_action simplification

["Jason Short" <jdorje@xxxxxxxxxxxxxxxxxxxxx>]

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=12833 >

Brian Dunstan wrote:

>>Actually, that's wrong.  You can't just amortize the
>>difference in tile
>>value, because that difference is not a one time
>>payment.  
> 
> Are you sure?  The value, according to
> evaluate_improvements(),  is the difference, before
> and after, of city_tile_value().  In a republic
> government, a square that is plains with road an
> irrigation has a city_tile_value() of 93, according to
> current weightings.  The question is, 93 of what? 
> What are the units being used?  If the units are gold
> pieces equivalent, the 93 looks like that is the value
> of the perpetuity in lump sum today.

If you read the text of the other ticket you'll see the current code is
(IIRC) buggy in this regard.

The actual math for calculating the payoff in current-day value of a
payment in perpetuity is pretty simple (simply the inverse of the
interest rate, as you say).  But it's made more complex by all the
complicated rigaramole that's added on (like the 4x factor for inuse tiles).

-jason