<< Back to General Forum   Search

Posts 1 - 17 of 17   
1 attacker vs 1 defender ?: 11/20/2011 02:42:36


[中国阳朔]Chaos 
Level 49
Report
is it possible that when you attack with 1 army and target a territory with 1 defending army, that both die? it seems it doesnt happen. It would leave zero armies in the defending territory, while still being neutral.

also, kinda related, why is it that if both attacker and defender have 100% kill rate, the attacker never kills the defending 1. (again in same scenario as above)

just curious
1 attacker vs 1 defender ?: 11/20/2011 03:20:43


Richard Sharpe 
Level 59
Report
No, it is not possible. In the wiki it says something about this and how it reduces the success rate of such an attack since if both 'win' then the defender automatically survives.

After all, you can't have a territory with 0 units in it.
1 attacker vs 1 defender ?: 11/20/2011 03:23:45


Richard Sharpe 
Level 59
Report
"If equal numbers fight and both sides kill the maximum number of armies possible, and since every territory must have at least one army, there would be no armies left to control the captured territory. In this event, WarLight simply resurrects one defending army and allows it to retain control of the territory. For this reason, it's rarely a good idea to attack a 1 with 1 or to attack a 2 with 2, etc. "

Wiki on 'Analyze Graph'
1 attacker vs 1 defender ?: 11/20/2011 04:03:04


[中国阳朔]Chaos 
Level 49
Report
ok thanks, but it is possible to have 0 armies on neutrals as setting though, so it could make sense it remains a neutral but without any defending armies
1 attacker vs 1 defender ?: 11/20/2011 15:39:35

SebCorps
Level 3
Report
No
1 attacker vs 1 defender ?: 11/20/2011 15:45:22


angel 
Level 3
Report
I have joined a game with all neutrals as 0 before.
1 attacker vs 1 defender ?: 11/20/2011 22:05:30

RvW 
Level 46
Report
You have a point, the game engine is indeed capable of dealing with zero armies in a (neutral) territory. However, what would happen if a player-controlled territory is reduced to zero armies? I'm not sure how the engine would handle that (allow it, or automatically revert it to neutral??).

I attempted to test this scenario by setting the number of starting armies per territory to zero (which could make sense; after all, you would still get reinforcements), but got an error message when attempting to do that:
"Initial player armies must be between 1 and 1000."
Note the slider only goes from 1 to 15, so this check has definitely been added separately. On the other hand, it is possible to have more than 1000 armies in a territory (I've seen 10k+ on USA big), so we cannot blindly assume the "at least one" constraint is there because the engine would screw up with zero armies.

Since this situation is special-cased already, it should be relatively easy (from a coding perspective) to add additional "solutions" for it. Allowing a territory to go to zero armies could be a setting (note this would change strategy quite a lot!) and deciding whether player-controlled territories revert to neutral when they lose all their defending armies.
I do not expect this to cause too many other new feature requests. I think people will understand they "usually" need at least 1 controlling army, which takes care of "I want to be able to move my last army out of a territory". Since only an attack by enemy forces can reduce a territory to zero defenders, it must by definition be on the front lines; preventing the need for an "all territories must be reinforced to one army before other territories can receive reinforcements" option.

I did a little testing with a low offence kill rate (10%). Each and every single 1 vs. 0 attack (out of many dozens) resulted in the "empty" territory to be taken (which makes sense of course, but I wanted to test it anyway, just to be sure).

To the best of my knowledge, it is currently impossible for a player-controlled territory to have zero armies on it. Also, a neutral territory with more than zero armies can never be reduced to zero armies; it can only have zero armies if it had zero armies at the start of the game.
1 attacker vs 1 defender ?: 11/21/2011 14:04:53

Darkruler2005
Level 7
Report
Would actually love to see this changed a bit. The game could do with the option to allow 0 armies on a captured territory (just like you can change fog, wastelands, etc). I never understood why you're forced to have 1 army on it, and especially now that I know neutrals are capable of it it should be possible for players.
1 attacker vs 1 defender ?: 11/21/2011 14:08:42


Richard Sharpe 
Level 59
Report
Darkruler... consider it like a captured territory in the real world. You would have to post a garrison to hold the territory and police it. The 0 for neutral would imply that it was simply unpopulated and undefended.
1 attacker vs 1 defender ?: 11/21/2011 14:48:54

Darkruler2005
Level 7
Report
That's a certain logical reason behind such a decision, yes, but that mostly just is an arbitary restriction. An option for this never hurts anyone. I feel there is no difference between this and changing the number of neutral armies in every territory to maximum.
1 attacker vs 1 defender ?: 11/22/2011 04:07:57

RvW 
Level 46
Report
@Darkruler:
You want to know the real reason every territory needs to have at least one occupying army...?

This game (and many others) were inspired by (the board game) Risk. Back then, the only easy way to keep track of who controls which territory is by using some kind of marker. And, guess what, Risk came with a whole lot of markers; the armies. It's just one of those things that gets taken over without too much thought going into it. But, that is not yet a reason why it can't be changed any more.

In Warlight, even more than in classical Risk, it is much harder to capture a territory defended by one army than it is to capture an undefended territory, especially when using only a single attacker.

Risk:
15/36 ~= 42%
[ties favour the defender]

WL (general case): Offence luck * ( 1 - Defense luck )
WL (default settings): 0.60 * ( 1.00 - 0.70 ) = 0.60 * 0.30 = 0.18 = 18%
[when both the attacker and defender die, one defender is magically resurrected, exactly to prevent the zero-armies-left scenario]

With no defenders, both games will have a 100% success rate, even when attacking with a single army. This would actually be quite a change in how the game works. Also, many settings on here have been very carefully fine-tuned (number of starting armies, values of bonuses, etc.). By freeing up the armies usually tied up in "guard duty", you'd ruin that balance.

So, even if it would be an easy fix (it very well might be), I doubt it will improve the game. If nobody likes the option, then there's little point implementing it, no matter how easy. (Then again, I'm not Fizzer, so you could always make it a feature request and see how he likes it.)
1 attacker vs 1 defender ?: 11/23/2011 13:49:42

Darkruler2005
Level 7
Report
I've played Risk more than I want to know, so yes, I understand in that game it's a necessity. As you said, not a reason here, since colour of the territory changes to that of the player.

I disagree with your reasoning about the ease of capturing a territory. Is it not a player's choice how much armies he leaves behind? I rarely, if ever at all, have more troops than 1 in every spot, since they're generally useless when not at the front line. I would definitely use all of those troops if I could set each territory to 0 minimum. But there's no risk of them being attacked when not at the front line. And when they break through, you are able to place troops as defense as much as you like (remember that it's still your choice how many defenders you left behind). On the default setting, you need 5 troops to certainly capture a territory with 1 defender, 8 for 2, etc. What matters is how many defenders you have on the territory during the turn the person attacks (which happens to be the same turn as the one in which you can place armies).

Don't get me wrong, I agree developing time should not be put into a setting if nobody wants it. I don't know if nobody wants it, but I guess that can be researched.
1 attacker vs 1 defender ?: 11/24/2011 02:57:27


[中国阳朔]TexasJohn 
Level 35
Report
RvW, it's possible to have millions of armies in a territory. Chaos created a game on the Star Wars map in which 20mil armies were crashing against 30mil armies. Is there an actual limit to how many armies a territory can support? I wouldn't think so, but you never know....
1 attacker vs 1 defender ?: 11/24/2011 03:53:31


[REGL]Hellgunner23
Level 21
Report
I am with Darkruler, I would love a setting where I could leave territories controlled with 0 armies, those troops would be of great use on the frontlines.
1 attacker vs 1 defender ?: 11/24/2011 04:22:58

RvW 
Level 46
Report
@TexasJohn:
Seeing how the game is supposed to "normally" work with a few dozen, maybe a couple hundred armies, if a couple million are allowed, it sounds like Fizzer didn't put in any explicit limit. In that case there's a good chance you're only limited by the internal representation, probably a little over 2 billion, or a little over 4 billion (note: "billion" in the sense of "thousand million"). If you can put more than 4.3 billion armies in a territory, the next likely limit would be at either 8 or 16 billion * billion.
Either way, I think that for all practical purposes we can say there is no limit. :p

On a slightly related note, just thought of something. When using the army cap, for it to have the "same" effect in a game where the minimum number of armies is one, the cap should be lower than in a game where the minimum number of armies is zero. Just something to keep in mind.
1 attacker vs 1 defender ?: 11/25/2011 17:58:06


Matma Rex 
Level 12
Report
Unless it was changed, the hard limit is 2^31 = 2147483648 armies per territory, which is the largest signed integer number most computers can internally represent and store without fuss. (You can also use 2^63 with still reasonably little fuss. Larger numbers are difficult to work with for reasons I won't dive into here :) )

Some time ago there was a thread where someone hit it in a game (by using sanction cards with huge negative values and huge abandons) and got an exception. It was since kind of fixed by limiting the maximal value of sanction card.

Here's the thread, and some further explanation: http://warlight.net/Forum/Thread.aspx?ThreadID=969
1 attacker vs 1 defender ?: 11/25/2011 19:10:25

RvW 
Level 46
Report
How exactly did we get from "1 vs 1" (pretty much the smallest battle you can fight) to "so many armies the integer overflows"...?

I think this is a critically important bug and should be fixed. As in, properly fixed. Switching to 64 bit integers is not a real solution; those still overflow, only a little later. The real solution was described, more than eleven years ago already, in [RFC2795](http://tools.ietf.org/html/rfc2795#page-5). I suggest this bug be fixed before RFC2795 turns 12 years old. ;)

Please click [here](http://www.rot13.com/index.php?text=Ybbx%20ng%20jung%20qngr%20ESP2795%27f%20oveguqnl%20vf.) if you don't get it.
Posts 1 - 17 of 17