Excess and Potential Changes in Experience.

[Dave]

Changing experience gained to 6 + 4(level) has the following effects:

1) It becomes easier to advance units near the beginnings of many campaigns.
2) It becomes less trivial to advance units near their ends.

Both of these would be good things, no?

No, I think both of these would be bad things.

At the moment playing a campaign well is already too dependent on doing well in the early scenarios, and then minimizing losses after that.

With this change, it'd become even more like that: to win you'd have to learn to milk your units up to level 3 as soon as possible, and then protect them later on, and that'd be it.

The design of the game is intended for people not to be able to level units up too fast early on, and even if units are lost later on, for there to be potential for recovery. Wesnoth is already unforgiving enough, this change would only make it more unforgiving.

I find that in latter scenarios, such as Northern Winter, my level 1 units can practically just walk around killing level 2 units and level up a ton.

And that is entirely the point of Northern Winter: to give the player the opportunity to level-up some first level units. This gives players who arrived at Northern Winter with a recall list decimated by previous difficult scenarios a chance to recover a little.

David

[Elvish_Pillager]

This gives players who arrived at Northern Winter with a recall list decimated by previous difficult scenarios a chance to recover a little.

Unfortunately, in order to execute this strategy properly, in my experience, you need quite a few advanced units to weaken the enemies first.

-----------

The current experience system is a major reason why SotBE is such a hard campaign: the enemies mostly get level 1 units. And it's true: Enemies getting higher level units makes the game easier for you. Isn't it weird?

[Dave]

This gives players who arrived at Northern Winter with a recall list decimated by previous difficult scenarios a chance to recover a little.

Unfortunately, in order to execute this strategy properly, in my experience, you need quite a few advanced units to weaken the enemies first.

You do need some, but you don't need a huge number.

The current experience system is a major reason why SotBE is such a hard campaign: the enemies mostly get level 1 units. And it's true: Enemies getting higher level units makes the game easier for you. Isn't it weird?

The enemy getting a few higher level units makes the game easier, yes. But sometimes that's precisely what the scenario designer intends: to give the player a chance to build up some experience.

SotBE is a hard campaign because it's designed in such an unforgiving way.

David

[Elvish_Pillager]

I suppose that's right... it just seems really weird and counterintuitive.

[stillnotelf]

I agree that the kill-bonus tends to generate some pretty bizzare tactics, like hoping Delfador will miss with his lightning bolts...

Has anyone ever attempted to implement an exp system based on damage? Perhaps:

(damage dealt/total damage possible) x 5 exp

and then no exp kill bonus? This would mean that the unit is only learning when it's actually successfully doing something, I always thought it was a little silly for my Shamans to get exp for missing with both attacks. I can see several serious problems with the idea, just off the top of my head (magic attacks gaining exp faster, exp being based on terrain), but I was wondering if anyone had perhaps tried this in a previous version?

[Doros]

It's seems like there's a resounding "no" in response to changing the experience formula, but what about EP's other proposal? (experience carrying over) It is really disappointing when a unit that needs a couple experience points accidentally kills the leader, and that really is just luck. I know chance is an important part of Wesnoth, but it seems really counterintuitive to worry about killing a unit by mistake, and I think it just leads to save-loading anyway.

[turin]

It's seems like there's a resounding "no" in response to changing the experience formula, but what about EP's other proposal? (experience carrying over) It is really disappointing when a unit that needs a couple experience points accidentally kills the leader, and that really is just luck. I know chance is an important part of Wesnoth, but it seems really counterintuitive to worry about killing a unit by mistake, and I think it just leads to save-loading anyway.

yes, it definitely leads to saveloading... I probably saveload more to stop a unit from getting the kill when it shouldn't than I do to stop a unit from dying...

But, I don't think it should carry over. I don't think it fits the balancing of the game.

[Dave]

It's seems like there's a resounding "no" in response to changing the experience formula, but what about EP's other proposal? (experience carrying over) It is really disappointing when a unit that needs a couple experience points accidentally kills the leader, and that really is just luck. I know chance is an important part of Wesnoth, but it seems really counterintuitive to worry about killing a unit by mistake, and I think it just leads to save-loading anyway.

yes, it definitely leads to saveloading... I probably saveload more to stop a unit from getting the kill when it shouldn't than I do to stop a unit from dying...

I don't see why people feel so compelled to do this......Wesnoth is intentionally designed so that you can't go around getting every bit of experience the way you want it. You can try to get a certain unit experience, and sometimes it will work out....and sometimes it won't.

David

[markus_2]

I don't see why people feel so compelled to do this......Wesnoth is intentionally designed so that you can't go around getting every bit of experience the way you want it. You can try to get a certain unit experience, and sometimes it will work out....and sometimes it won't.

In my experience, this is not a issue for units which tend to get experience anyway. I know my archers and fighters will eventually get enough experience. But when it comes to mages and shamans, taking a cheap killshot is the major source of experience for them, since they won't survive a round in the front line.
One might speculate, that saveloading is just the next step in the contortions usually required to get an unit a safe killshot. Then again, that might just be a convenient excuse for cheating.

[Doros]

yes, it definitely leads to saveloading... I probably saveload more to stop a unit from getting the kill when it shouldn't than I do to stop a unit from dying...

I don't see why people feel so compelled to do this......Wesnoth is intentionally designed so that you can't go around getting every bit of experience the way you want it. You can try to get a certain unit experience, and sometimes it will work out....and sometimes it won't.

David

It's not that we want every bit of experience we can get, it's just that you need a lot of experience for a certain unit sometimes (particularly at the beginning of the game), and if you don't level enough units in the first few scenarios, you can either play them over, lose consistantly in the later ones, or save-load. I tend to dislike the first few scenarios of every campaign because I have to spend most of it painstakingly making sure that I level units quickly, and I have to reload every time one of my important units gets cheated out of 8 to 24 experience points at random.

[drachefly]

I have to agree -- wrangling things so the weak units can make a KILL is a powerful requirement, especially as the level 2 versions of weak units are SO much more powerful (compare Shaman to Druid, Mage to White Mage).

2 + level XP just for fighting would probably preserve the current XP balance well, and greatly reduce support-unit wrangling, by encouraging them to participate in a protected niche, but not requiring the line of battle to be distorted to arrange for them to kill.
But pressure to put healers into combat will continue to produce weird effects... though hopefully not as weird.


Stillnotelf:
In respect to your XP only on success idea: this is totally backwards. I think it would make more sense to get EXP for failure. In real life, I successfully did many things today, but I did not gain any experience from them since they were easy. On the other hand, when I was a baby, I found those things difficult. What is the best way to estimate difficulty? Failure rate.

In Wesnoth, of course, hitting isn't based on your skill, it's based on the defender's defense % in the terrain they're standing on, unless you're a marksman or using magic. So, I don't think that's a good direction to go.

[stillnotelf]

Stillnotelf:
In respect to your XP only on success idea: this is totally backwards. I think it would make more sense to get EXP for failure. In real life, I successfully did many things today, but I did not gain any experience from them since they were easy. On the other hand, when I was a baby, I found those things difficult. What is the best way to estimate difficulty? Failure rate.

In Wesnoth, of course, hitting isn't based on your skill, it's based on the defender's defense % in the terrain they're standing on, unless you're a marksman or using magic. So, I don't think that's a good direction to go.

This is a problem I see, it's backwards to do it one way, but it's also kinda backwards to do it the other way. Wesnoth defense is a hit percentage, implying that it's really the defender's ability to hide that matters, not the attacker's ability to hit (excepting Marksman and Magic). It doesn't make any sense to give someone exp because they couldn't see their target, they won't learn how to hit better because they couldn't locate the opponent. Exp for hitting the target is like positive reinforcement; they learn when they do it successfully. I think it has to do with how you consider the unit's learning abilities.

Perhaps it would be good to distinguish exp and learning: you LEARN from failure, but you become expert at a set of motions through repeated successful attempts. Thus, exp from success, learning from failure.

It might not be a bad idea to make it exp-on-misses just to make the randomness a zero-sum-game, you get either damage on the opponent OR exp, you're happy either way

[edit was for clarity, I initially was very rushed when writing]