My thoughts about randomness in Wesnoth

[Romla]

I know there was a lot talking about this theme and developers don't want to decrease randomness in the game, because they like it. I like it too, but I don't like one thing, which happens sometimes:

I had two relatively healthy Knights (elite of my small army - A Tale of Two Brothers) and I wanted to kill one badly injured unit. Both my Knights had 64% chance to kill it and both had 6% chance to die. And yes - they both died and enemy unit stayed unharmed.

I'm trying to play without loading, but when this thing happens, I don't have another choice, but to load a game or start a new one, regardless how smart I am playing - because you must take little (sometimes bigger) risks to prevent larger losses. And this thing (in my opinion) it's not realistic, it's not fair and above all it's not fun (at least for me).

Life (war) is about randomness but not about wonders. I especially don't like the fact, that one unit can kill two much stronger units without even taking damage.


I know that developers don't want to decrease randomness, so my solution to this cause is only hypothetical and to my liking:
Change the damage calculation system in the way, that units will do damage in every attack, but they will not do full damage, but only random part, depending on terrain (defense).

Example:

Wesnoth style:
Knight with 8x4 attack can do (let's omit other factors, resistances etc.) 0-32 damage on virtually every terrain (there are only different probabilities he will do 0 or full damage on different terrains).

My style:
Knight with 8x4 attack can do (again let's omit other factors) on terrain with 50% defense only for example 30-70% of damage with every attack. So he will do at least 8 (|8x0.3|x4) damage and at most 24 (|8x0.7|x4) damage on terrain with 50% defense. And for example 0-40% damage on terrain with 80% defense. The formula is simple: +-20% damage on every terrain (can be more or less).

Advantages:
1. One badly injured unit cannot kill many stronger units without even taking damage.
2. Attacks are not so much unpredictable (but they are still random).
3. Perhaps others.

Disadvantages:
1. Maybe not so simple, but definitely not much more complicated.
2. Not so random how developers like it.
3. Surely others...

It was only my idea and I will play Wesnoth with current damage calculations. You are free to comment on this, why is this good, why is this bad, or you can ignore it. I'm sorry if something very similar was posted before.

Thank you for your attention.

[Jozrael]

Something very similar has been posted before. Once a mod sees this, any topic on luck gets locked on sight so I'm trying to give you a bit of an explanation before this since it seems you're rather polite.

The luck system won't be changing. It's been brought up LITERALLY hundreds of times by new players. It is the foundation of wesnoth. If you personally want to code/create/release a different version with your luck system, there is -no one- stopping you. But there is not a single dev that will put an ounce of effort towards changing the luck system. They've each spent countless hours debating it with different people (each of whom haven't heard the other discussions so they need to go over the entire subject again) and they are frankly just tired of it.

I wish you well

[Radament]

Hi,

Personally i'd be interested to see how your proposal would work out. Still Wesnoth is fine as it is; the probability of your example happenning is lower than 5/10000. I guess thats roughly the probability of me getting hit by a car when i cross the street at certain places:)

It's a problem with the Campaigns. Luck changes violently, and in multiplayer you're gonna loose (and win!) due to excessive randomness, but after many many games it evens out. In single player all is connected, and you're definitely gonna need to restart/reload on higher difficulty levels.

[thespaceinvader]

Jozrael provides the serious response, i get to provide the less serious one =D

In the form of this elegant and finely crafted link.

[Radament]

Btw, you might wanna check this out:
http://saurons-mod.zor.org/

[Wintermute]

Jozrael is correct. I'll just add: Consider playing poker. It might bother some people that you can have several bad hands in a row (high card jack, say). But you could also have 2 full houses in a row. Imagine if we change the rules for poker: now we can have no hand less than two pair, and no hand better than a flush. This is a different game than poker. Some people might like it - but it is not poker. It is a different game, and anyone who wants to play POKER will not like it.

So it is with Wesnoth. You are free to create such a mod, and many players may like it.

[Romla]

Yes, I understand. Thank you very much for your kind replies, especially to Radament - the mod can maybe solve my problem (or sometimes I will try to do my own, but this can save me time).

[dontano]

That link is awesome! ^^ Strange that I never seen it before... whatever

I was thinking about a new luck system that involves a "Luck" parameter, without considering terrain resistances, etc... This could have some similarities to the various RPG-like games that have the same feature (Accuracy - Evade).
For instance, Luck increases the chances of getting unharmed while in combat, and the more you have it the better it is (say... 0 to 255). If a Spearman has 10 of luck, then there is only a minimum chance to avoid an enemy attack, whenever he is sinking in deep water or is on a village. That means less frequent "shift tide-medal" and frustrating situations, including survivals and more so. How many times did you face nearly untouchable bats in Orocia, speculating that they were using a cheap cheating device like Gameshark?
About Accuracy... well, I'll say that it's quite the same situation as the Evade, but reversed. Magic attacks have, for example, 50 points out of 255, and other regular weapons have 20. I dunno though how to calculate the various situations, since they're based on RPGs typical combat mechanics. ...And since Wesnoth isn't an RPG but instead a TBS (Turn-Based Strategy game), then forget what I said.

PS: I like those methods of calculating luck, Romla.

[zookeeper]

Sauron's mod is two years old, there's not much you can do with it now.

If someone wanted to make a luck mod today, the best way would be to implement a couple of relatively simple WML features into the main game which would then enable one to write most of the actual luck mod in WML. That way, one wouldn't need to compile a different client to play the mod (which really is a huge, huge obstacle).

[dontano]

So, that means that it would imply more than just a few macros right?

[Romla]

If someone wanted to make a luck mod today, the best way would be to implement a couple of relatively simple WML features into the main game which would then enable one to write most of the actual luck mod in WML. That way, one wouldn't need to compile a different client to play the mod (which really is a huge, huge obstacle).

So there is maybe one thing which developers could write down to new release plans...

[Jozrael]

I think he was giving you tips on how you could do it. As far as I know, no dev is currently willing to put in work for changing how luck affects the game.

[zookeeper]

That's right. The devs have little to no interest in luck mods, but a clean minimal patch that'd enable something like that to be done in WML would probably be accepted. At least in order to enable the stock CABD answer to luck change/mod suggestions too, if nothing else.

[Noy]

No, For reasons I've stated here

Locked

[Dave]

I had two relatively healthy Knights (elite of my small army - A Tale of Two Brothers) and I wanted to kill one badly injured unit. Both my Knights had 64% chance to kill it and both had 6% chance to die. And yes - they both died and enemy unit stayed unharmed.

So you're saying that you risked your two healthy knights by charging at an almost dead enemy that did have some chance to kill you?

Perhaps you should have considered a more prudent approach and attacked with swords? You'd probably have still had a chance to kill, would have had a higher chance of doing at least some damage, and would have had no chance of your knights being killed. Your knights would also take less damage during the encounter.

Knights, and other units with charge, are by their very nature chancy units. If you don't want the chance of things going sour, then use less chancy units. Making some change to the game to make everything non-chancy would just reduce the distinction between units.

David