Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2002:
[Freeciv-Dev] Re: Bombers/Fighters no longer obsolete (PR#1628)
Home

[Freeciv-Dev] Re: Bombers/Fighters no longer obsolete (PR#1628)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: Raahul Kumar <raahul_da_man@xxxxxxxxx>
Cc: "Ross W. Wetmore" <rwetmore@xxxxxxxxxxxx>, freeciv-dev@xxxxxxxxxxx, bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: Bombers/Fighters no longer obsolete (PR#1628)
From: Raimar Falke <rf13@xxxxxxxxxxxxxxxxx>
Date: Fri, 28 Jun 2002 12:12:02 +0200

On Thu, Jun 27, 2002 at 07:35:31PM -0700, Raahul Kumar wrote:
> 
> --- Raimar Falke <rf13@xxxxxxxxxxxxxxxxx> wrote:
> > On Wed, Jun 26, 2002 at 09:56:41PM -0400, Ross W. Wetmore wrote:
> > > At 09:45 PM 02/06/25 -0700, Raahul Kumar wrote:
> > > >Bomber                     Stealth Bomber
> > > >
> > > >build_cost    = 120        build_cost    = 160    + 33%
> > > >attack        = 12         attack        = 14     + 17%
> > > >defense       = 1          defense       = 5      + 400%
> > > >hitpoints     = 20         hitpoints     = 20
> > > >firepower     = 2          firepower     = 2
> > > >move_rate     = 8          move_rate     = 12     + 50%
> > 
> > So you get +17% attack, +400% defense and +50% speed for only 33% more
> > cost.
> 
> A rebuttal
> 
> Ross, Raimar: I offered the attackpower formula in order to avoid a long drawn
> out discussion. It is not an ah-hoc formula, or random, or equivalent to any 
> 20
> other formulas anyone may choose to make up. It analyses the most important
> parts of the combat effectiveness of any unit. Now I will compare and contrast
> the formula I stole from the AI (This is a first. I never thought I would ever
> defend any piece of code in the AI!) to Raimar's, as the given alternative.
> 
> This is an example of why Raimar's weights method is flawed. In order to 
> figure
> out whether a unit is a good attacker or defender, you consider the two
> seperately. 
> 
> ===============================================================================
>   Legion -> Musketeers
>               build_cost:  40  -- >   30  -25%
>          attack_strength:   4  -- >    3  -25%
>         defense_strength:   2  -- >    3  +50%
>                       hp:  10  -- >   20  +100%
>  
>  We may not want that the attack_strength is reduced even is the
>  build_cost is also reduced by the same amount. This wouldn't be a
>  problem if we would have real stack-vs-stack fight. But for the
>  current code it may be a disadvantage because you can only enter the
>  fight with one unit.
> ================================================================================
> Here is one example that Raimar's approach and mine clearly differ. 
> 
> Musketeer attackpower = 3 * 1 * 20 = 60 
> Legion                 = 4 * 1 * 10 = 40
> 
> Defence power is Musketeer 60 vs Legion 20. A even bigger mismatch.
> 
> And the legion's build cost is greater! There is never a time when you would
> want a legion as opposed to a musketeer. Raimar's method introduces extra
> complexity, ignores the relationship between firepower/hps/attack. And it
> produces the wrong answers. It also fails to emphasize why musketeers are so
> powerful in the game. My method also answers what would be needed to make
> legions as powerful as musketeers.

You are correct. I missed the doubling of the HPs.

> So to be blunt: Stealth bombers are weak. Bombers are far better attackers for
> the shield cost. Remove the obsolete tags. It makes no sense to "upgrade" to a
> weaker unit. The move rate and defence are irrelevant, because bombers are 
> used
> exclusively in a attacking role. 

So what about increasing the HPs to 30 and reduce the defense strength
from 5 to 3?

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "It is not yet possible to change operating system by writing
  to /proc/sys/kernel/ostype."              sysctl(2) man page


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