Complete.Org: Mailing Lists: Archives: freeciv-dev: July 2003:
[Freeciv-Dev] Re: (PR#4645) Silent Hunter and Sonar code.
Home

[Freeciv-Dev] Re: (PR#4645) Silent Hunter and Sonar code.

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: undisclosed-recipients: ;
Subject: [Freeciv-Dev] Re: (PR#4645) Silent Hunter and Sonar code.
From: "Rafa³ Bursig" <bursig@xxxxxxxxx>
Date: Tue, 22 Jul 2003 07:31:36 -0700
Reply-to: rt@xxxxxxxxxxxxxx

Dnia 2003.07.22 13:22, Per I. Mathisen napisa³(a):
> 
> On Tue, 22 Jul 2003, Rafa³ Bursig wrote:
> > 1) SILENT_HUNTER
> > basicaly it is submartine functionaliy and work like Partial_Invis (
> > replace it ) with differences that sub is not detected until any
> unit
> > enter on tile with sub. When this happend and unit onwer is in war
> with
> > sub owner then sub automaticaly attack first (if can) if not then
> unit
> > stop.
> > In both cases sub is dettected and visilbe untill end of turn.
> ...
> > 2) SONAR
> > abbility to detect SILENT_HUNTER units (added to Destroyers and
> Heli)
> > all cities have this functionality too.
> 
> I'll just point out that this one big disadvantage: Right now we have
> relied all over the codebase on the assumption that we see all units
> adjacent to us. Breaking this assumption is going to cause all kinds
> of problems.
> 
Primo: this is next civ2 feature that I want have in freeciv.
(In Civ2 you can find sub by Destroyers or entering on tile with this 
sub not moving on adjacent tile)

Secundo: This will make sub and destroyer units usefull again, as I 
know current we have problem with it.

Tercio :
We are here to fix problems :)

> You will also have all kinds of strange problem cases. For example,
> what will happen if you try to move into a tile with one allied unit 
> and one at_peace silent hunter unit. For that matter, what to do if 
> you try to move into an at_peace silent hunter.
> 
Primo:
- server know about all units.
- SH attack only when you find SH unit by the unit that can be attacked 
by SH unit (other way only find this SH unit).
- if SH has been found then it is normal unit (you can normal attack 
it).
- patch is unfinished and may of that may not impemented and not tested.

Secundo: what happen when you try enter on tile with one ally unit and 
one at_peace unit (you know about this unit) ?

Tercio :

if you don't know about SH unit in ally stack then your unit will stop 
and you get msg that you found SH unit there. IMHO this problem is 
impossible becouse the only way that SH unit can be on the same 
position that your ally units is that SH onwer must be ally with your 
ally and in this case you will know about this SH unit.

if you attack units stack with SH unit inside then normaly best 
defender is find (it can be SH) and start normal combat.
If you destroy defender then all units on stack are destroyed (SH too)

Cities find SH units in full city range.

other question ?

> The "defender attacks" idea also has problems. For example, ground
> units cannot attack air units. What if an air unit attacks your 
> silent hunter unit?
> 
if your air unit try enter on tile with unknow SH unit (no fighter/sam 
abbility) then your air unit stop and you get msg that you found SH 
unit here.

> Finally, the AI will have to see these units with its omniscience.
> Otherwise a lot of AI code must be rewritten.
> But that removes much  of the point of adding them in the first 
> place, since they cannot be  used for single-player. (The same is 
> already the case for F_PARTIAL_INVIS, BTW, but I plan to fix that one 
> day. That won't be hard.)
hmm... why it can be use with single-player ?
you can always use it agians AI.

I'm not a AI specialist but...
- almost all new functionality requre AI changes.
- first we must finish adding this new functionality then "teach" AI 
use it.
- if AI work on server then it know all units positions ?

btw how many current futures can be used by AI ?
( I know about new trade code )

fixing AI should be our primary target ?

> The concepts are interesting, but I do not think we should be
> introducing them at this time. We have enough code-breaking patches 
> queued up for the next release as it is.
I know about it and I plan this change after our primary 1.15 changes :)

Rafal




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