new player frustrated about losing leveled units

[Zerovirus]

Perhaps certain healer-mage style units could have an extra special, like cure and heal. Revive? If any unit could revive a fallen unit that's much more incentive to go for a spam attack instead of using units that cost more.

This idea is cool, and might turn out to be the first major change to Wesnoth gameplay we've seen in like ever.

(Of course, it's Dave here talking about it, so yeah.)

[Zarel]

This isn't an RPG this is a strategy game. Even on some of the harder campaigns you will rarely need a field full of level 3s. When you do it is generally because you are playing with a highly limited amount of gold. Most campaigns give you enough experience that with the possible exception of the start of the campaign and the final battle you would never be using all your level 3s anyway so you effectively have backups.

This may be a strategy game, but its leveling system plays like an RPG, albeit an RPG with rather low level advancement.

Actually what I'm in favor of is making perma-death much more possible to avert. When a unit's hitpoints reach 0, it should be marked as 'dying', and the hex it is on marked with some kind of symbol. The unit will die within a certain number of turns, unless you move a unit onto the hex, in which case you are able to rescue/heal it, and it will be returned to your recall list for use in later scenarios. It is possible that for a certain number of scenarios afterwards it may be unusable, marked as 'recovering' in the recall list.

Hrm. I think making it impossible to recall a "dead" unit until the next scenario is a good idea. It neatly solves the problem of what to do about multiplayer.

I don't think moving a unit onto a hex to have it is the best solution to averting perma-death. There could be inadvertent situations in which a unit could die without another unit being able to save it. Then again, I don't have a better idea.

Perhaps certain healer-mage style units could have an extra special, like cure and heal. Revive? If any unit could revive a fallen unit that's much more incentive to go for a spam attack instead of using units that cost more.

The problem is that this changes gameplay depending on whether or not your faction has healer units.

[Edwarfin]

Actually what I'm in favor of is making perma-death much more possible to avert. When a unit's hitpoints reach 0, it should be marked as 'dying', and the hex it is on marked with some kind of symbol. The unit will die within a certain number of turns, unless you move a unit onto the hex, in which case you are able to rescue/heal it, and it will be returned to your recall list for use in later scenarios. It is possible that for a certain number of scenarios afterwards it may be unusable, marked as 'recovering' in the recall list.

This concept is likable and undeniably fascinating, but I have a few questions about how it can fit in with the existing game, and am slightly concerned with its level of realism.

As you all know, campaign heroes can't be killed, or else the campaign immediately comes to an end. What happens, though, if this feature is implemented? Say that Prince Konrad is fighting it out with some orcs, and he gets 'killed', but as he's dying, some elf comes up and rescues him. Since Konrad isn't actually dead (so the campaign to save Wesnoth can continue), what happens then? Do the elves lose the ability to recruit new troops for the rest of that level? Is Konrad out on illness for the next few scenarios, and you "fill in" as the highest-ranking unit, or so, while he's out? If so, would we have to make new dialogue for any characters that might be taking the leadership role while the hero's gone? It seems to me that that would require a fair amount of work.

On realism, it seems that if some loyal unit can come up and rescue a dying soldier, that the enemy could also finish the guy off, in just as much or less time. I know that realism isn't always the major focus of Wesnoth, but this seems counterintuitive. It also eludes understanding in some cases -- for instance, if a unit with 1 HP got blasted by Delfador's lightning, or smashed by a troll's club, or had a lance driven through his heart, how he could reasonably be expected to survive that?

What would happen if an enemy unit entered the hex that the dead soldier was in? Does he finish killing him, or is the guy left intact? If the latter, why? If the former, it seems that it'd be so easy to finish a guy off that we might as well have left the original death system in.

How are AI units done away with? Do you have to wait a few turns for them to die off? If so, that could mess up the balance of some strategies; just trying to kill the leader isn't effective, since you also have to survive their followers long enough that the leader dies. If such a leader is defeated, say, Asheviere, and they are then rescued, does this mean you have to kill every unit in the area to finally slay her?

I'm not against this idea, I'm just curious as to how these issues can be overcome.

[catwhowalksbyhimself]

Honestly, I don't see a problem here in the first place. Especially in large campaign losing the odd level 3 isn't going to be that big of a deal as long as you level new units to replace your loses. Wesnoth isn't a role playing game, it's a strategy game where losses should be expected and accepted. Obviously, you don't want to lose too many and/or certain ones but that doesn't mean we should suddenly remove an entire lair of strategy (managing your losses) just for those who might struggle with this concept. We might as well well remove the day night cycle, or resistances, or attack types, or anything else some new players have to learn.

[Dave]

As you all know, campaign heroes can't be killed, or else the campaign immediately comes to an end. What happens, though, if this feature is implemented? Say that Prince Konrad is fighting it out with some orcs, and he gets 'killed', but as he's dying, some elf comes up and rescues him.

In general when your leader is incapacitated (falls to 0 hitpoints) you would lose the scenario instantly.

Whether you lose when other 'hero' units lose all their hitpoints or actually die is up to the discretion of the scenario designer.

On realism, it seems that if some loyal unit can come up and rescue a dying soldier, that the enemy could also finish the guy off, in just as much or less time. I know that realism isn't always the major focus of Wesnoth, but this seems counterintuitive. It also eludes understanding in some cases -- for instance, if a unit with 1 HP got blasted by Delfador's lightning, or smashed by a troll's club, or had a lance driven through his heart, how he could reasonably be expected to survive that?

What would happen if an enemy unit entered the hex that the dead soldier was in? Does he finish killing him, or is the guy left intact? If the latter, why? If the former, it seems that it'd be so easy to finish a guy off that we might as well have left the original death system in.

He survives in such a case. I find the realism argument fairly irrelevant, but even if we want to talk realism, we can. There is no real statement in Wesnoth that a unit represents one physical character on one physical hex. Wesnoth is simply an abstract representation of fighting units on a battlefield. A defeated unit might simply be 'scattered' within the hex they are in, which might be miles across. Hunting down the members of the unit and killing them would be time consuming in the middle of a battle.

Indeed this is very realistic. In real world battles, it is uncommon for even the losing side in a battle to have more than 20%-30% of their combatants killed -- especially in the ancient world.

Furthermore, since the units involved in the battle are not going to have any further affect on the current scenario, the unit defeating them is more motivated to simply incapacitate them than wipe them out for future battles. After all, the motivation is to win the battle, not do damage to the enemy army to hurt them in future battles.

How are AI units done away with? Do you have to wait a few turns for them to die off?

Remember that units that are defeated can only be 'rescued' for use in future scenarios. They can have no further affect in the current scenario.

Since the AI doesn't have a recall list and can't reuse units from one scenario to another, the concept of an AI rescuing fallen units is irrelevant and doesn't apply. As soon as an AI unit falls to 0 hitpoints it is removed from the battlefield permanently as happens now.

David

[tsr]

How are AI units done away with? Do you have to wait a few turns for them to die off?

Remember that units that are defeated can only be 'rescued' for use in future scenarios. They can have no further affect in the current scenario.

Since the AI doesn't have a recall list and can't reuse units from one scenario to another, the concept of an AI rescuing fallen units is irrelevant and doesn't apply. As soon as an AI unit falls to 0 hitpoints it is removed from the battlefield permanently as happens now.

David

I like something about this general idea. But this really caught my attention. Can the AI handle recalls at all or is it technically crippled in this aspect? I mean is it just "the way we do it" that makes the AI-sides lack a recall list?

I get nice (albeit slightly blurry) images in my head when I think about your latest post David especially how you will be able to fight throughout a whole campaign against "the same" enemy forces (considering your angle of "not dead, just scattered").

/tsr

[zookeeper]

Actually what I'm in favor of is making perma-death much more possible to avert. When a unit's hitpoints reach 0, it should be marked as 'dying', and the hex it is on marked with some kind of symbol. The unit will die within a certain number of turns, unless you move a unit onto the hex, in which case you are able to rescue/heal it, and it will be returned to your recall list for use in later scenarios. It is possible that for a certain number of scenarios afterwards it may be unusable, marked as 'recovering' in the recall list.

I think this system would add some more strategy to the game -- the overhead of having to plan to save fallen units. It would also allow you to fairly easily save units from death if you had some unlucky rolls amidst a fight you were winning. If you mis-strategize and leave some unit isolated against overwhelming odds, however, you risk perma-death.

I'm not sure what to say, frankly that sounds like a not such a good idea. In fact, that sounds like a huge can of worms to me. I'll just list some of the things which would be problematic:

1. What would happen with plague and such?
2. Your regular units could safely die and be rescued, but you'd immediately lose if your leader dies? Sounds very annoying.
3. A much more viable strategy would be to just rush the enemy and consciously sacrifice even your high-level units as long as you have enough quick and cheap units to go around and rescue everyone who's dying. Having those high-level units be out of commission for the next scenario or two isn't a problem if you have enough of them - get half of your high-level units disabled in one scenario, the other half in the next one, get the first half back again in the next one, repeat.
4. Scenario design becomes much more difficult because when a unit "dies", it might not actually die. You can't really write a piece of dialogue for an enemy who kills one of your units, because your unit isn't actually dead. And if it really dies on the next turn when you fail to rescue it, there's no killing unit associated with it anymore. Or it might itself be dead.
5. Exceptions to the rule because of scenario design problems (such as it being impossible to rescue special unit X after getting "killed" by their nemesis, because there's some special dialogue or something happening when that occurs) would be really annoying. Firstly the exceptions be clumsy to communicate to the player, and secondly it'd still be annoying that sometimes some of your units would be severely handicapped in that they couldn't be rescued.
6. It would simply alter campaign gameplay considerably, and doesn't sound like it'd be for the better. The risk of permanently losing a high-level unit would effectively completely disappear in many situations so you could safely have your low-hp mage shoot the enemy leader even if he has a chance to die if you don't really need him in the next scenario. Basically, it'd be really easy to abuse. You rarely have so few high-level units later on in a campaign that you'd recall even most of them every scenario, so retreating damaged high-level units back to heal in a village wouldn't really make much sense when you could simply suicide them while dealing lots of damage to the enemy, rescue them, have them be recovering for the next scenario or two, and then repeat.

Of course, some of those might have simple solutions.

[silene]

Actually what I'm in favor of is making perma-death much more possible to avert. When a unit's hitpoints reach 0, it should be marked as 'dying', and the hex it is on marked with some kind of symbol. The unit will die within a certain number of turns, unless you move a unit onto the hex, in which case you are able to rescue/heal it, and it will be returned to your recall list for use in later scenarios. It is possible that for a certain number of scenarios afterwards it may be unusable, marked as 'recovering' in the recall list.

1. What would happen with plague and such?
2. Your regular units could safely die and be rescued, but you'd immediately lose if your leader dies? Sounds very annoying.
3. A much more viable strategy would be to just rush the enemy and consciously sacrifice even your high-level units as long as you have enough quick and cheap units to go around and rescue everyone who's dying.

That looks a lot like the system in the Final Fantasy Tactics games: dead units can be revived by allied units in the three turns following their death. So I can comment a bit on these three points from experience:

1. No idea, there is no "plague" ability in FFT. But there is a "zombie" ability that is somehow related: zombies don't need anybody to help them, they auto-revive after three turns.
2. Right, in some FFT games, the death of your leader is an instant loss, and this is indeed quite annoying in practice. Especially since the leader is quite exposed.
3. No, this strategy doesn't work. In order to revive a unit, you need to use an uncommon item (depending on the FFT game, you may not even be able to buy it) and it can be used by units at the expense of other abilities. Some other units have instead revive abilities, but again at the expense of other abilities. In the context of Wesnoth, you could see reviving as a trait with a big downside.

[Zarel]

3. A much more viable strategy would be to just rush the enemy and consciously sacrifice even your high-level units as long as you have enough quick and cheap units to go around and rescue everyone who's dying. Having those high-level units be out of commission for the next scenario or two isn't a problem if you have enough of them - get half of your high-level units disabled in one scenario, the other half in the next one, get the first half back again in the next one, repeat.

There would be a penalty for doing so - the price of the unit and then some, maybe some experience loss... enough to make it so intentionally sacrificing high-level units would be a fairly bad strategy.

4. Scenario design becomes much more difficult because when a unit "dies", it might not actually die. You can't really write a piece of dialogue for an enemy who kills one of your units, because your unit isn't actually dead. And if it really dies on the next turn when you fail to rescue it, there's no killing unit associated with it anymore. Or it might itself be dead.

Rewrite "kill" to "defeat". Seems simple enough.

5. Exceptions to the rule because of scenario design problems (such as it being impossible to rescue special unit X after getting "killed" by their nemesis, because there's some special dialogue or something happening when that occurs) would be really annoying. Firstly the exceptions be clumsy to communicate to the player, and secondly it'd still be annoying that sometimes some of your units would be severely handicapped in that they couldn't be rescued.

There are already exceptions to the rule in the other direction, anyway - witness Li'sar being "killed" 2-3 times before joining you in HttT.

6. It would simply alter campaign gameplay considerably, and doesn't sound like it'd be for the better. The risk of permanently losing a high-level unit would effectively completely disappear in many situations so you could safely have your low-hp mage shoot the enemy leader even if he has a chance to die if you don't really need him in the next scenario. Basically, it'd be really easy to abuse. You rarely have so few high-level units later on in a campaign that you'd recall even most of them every scenario, so retreating damaged high-level units back to heal in a village wouldn't really make much sense when you could simply suicide them while dealing lots of damage to the enemy, rescue them, have them be recovering for the next scenario or two, and then repeat.

Like I said, a high penalty for death would prevent most problems along these lines.

[jb]

Jetrel

At this point, I'm completely convinced that the perma-loss of L3 units is a broken design; it's a similar thing as I've seen in a number of RPGs. With that much state invested in the character, only a handful of freaks actually let them die.* The other 95% of us immediately save-load the game.

Actually, I consider save-loading cheating. If I fail a scenario, i'll restart from turn one. Maybe MP guys are just more "hardcore", but I'd never stop my entire war engine for a single unit. This seems to be the default attitude of the circle of players I chat with, not the "handful of freaks" attitude. Unless you are calling us the handful of freaks....wait a minute!

I hear what you are saying though. I do. It's frustrating.


Dave

Actually what I'm in favor of is making perma-death much more possible to avert.

I've got to agree with zookeeper on this one. I don't think it's a good idea. The concept of perma-loss is what gives each high-level unit special value. Death should be painful and have consequences. I put perma-loss avertion akin to training wheels of a bicycle, where you should fall, you don't fall.


Please, let's not consider making this the norm.

[pauxlo]

Can we say this CABD with WML, and so scenario/campaign designers can themselves decide whether they want this?

[Daracos]

hi im totally new to wesnoth but im a competitive strategy gamer for a long time.
the most important point is:

keeping your units alive is part of the strategy!!!!!!

if your units die, it is because your strategy failed. it is really just that simple. believe me.
it does not matter how much luck there is involved. if your strategy is superior, luck is just a side effect.
lets consider a mp game. of course you can loose due to luck. but will it happen twice against the same player? maybe but not more often than that. in the end what counts is your skill, game-experience and creativity, plus mood, plus everyday ups and downs and so on. there are a lot of factors and the luck factor of the damage calculation is just one of them. and luck is important to make thing unpredictable and funny.

on perma death: i like it. really. when you get used to it, it is just the spices in your meal. it is exactly what makes the awesome coolness of my elvish sword master even more valuable. he is really ownage, but he is no super hero. he may die. this is the reason why i sometimes withdraw an entire army behind a river or what ever, to avoid dying and avoid having to restart the scenario.

i have a friend who always talks about the exact thing when we play a strategy 1on1. and it annoys me to death. really. what is more frustrating than a player you have beaten, who afterwards complains about the horrible luck factor and takes away all your credit??? its not about the luck factor.
this whole discussion is about people who cant admit to have lost.
i get frustrated aswell when i loose my cool lvl 3 units. but then i try to improve my strategy and it mostly works.

also remember some braveheart sequences when you think about losses in a medieval battle

edit: another point: today i read the article about the philosophy of wesnoth ... it was the first time that i heard about this basic design principle behind the game (simplicity), and it made a lot of sense to me... so i really believe those considerations of reviving units (at least those i read above) sound really complicated and imply lots of chain-reaction-like changes of other ideas

[fozzy]

I am a new poster, but long time lurker. And I'm very pleased that this topic has been brought up by a developer. It's been something I've been thinking about a great deal lately.

(Bunch of anecdotal and personal rationale following, feel free to skip to my lists if you wish)

I had a discussion with a bunch of coworkers recently about Wesnoth. As can be expected from developers who like to game, they have all played it at one point or another. The response was not what I expected though; it was pretty much unanimous frustration with the campaigns. The friends that I play MP games with all love it.

Full disclosure, I had not played a campaign in a long time. I have shifted to almost exclusively multiplayer games (sometimes vs the AI). When I did play 6-7 campaigns in the past, I was a shameless save-loader. I recall it being frustrating at times, but I continued because it was free and I had alot of time on my hands.

I started playing campaigns again to refresh my memory and to see if I could apply my MP experience to the campaigns and avoid save-loading. And it IS frustrating. And don't misunderstand, it's not frustrating in a "this is not possible" sense, but rather in a "this hard in a way that can often be unrewarding". This experience was on medium and hard difficulty (medium can often prove to be too trivial).

Campaign scenarios have to be approached in an entirely different way that is frankly less rational.

• As opposed to focusing on the scenario at hand, 95% of my effort is focused on keeping a few precious leveled units alive.
• This is unrealistic and awkward because my scenario opponent has no apparent concern about the lives of their units. Likewise, my damage to my opponent's leveled units is meaningless in the context of the campaign as a whole, which is completely at odds with the player's situation. This results in stilted gameplay.
• The experience in campaigns is totally at odds with that of MP, and likely not a great place to prepare.
• Occasionally, no matter how conservative I'm playing and no matter how much I've tried to make sure the odds are overwhelmingly in my favor, with enough dice rolls eventually a catastrophic turn of luck will occur and I'm forced to start over.
• Related to the last point, if I don't start over, still complete the scenario, but lose critical resources, I might hit a wall further in the campaign. I have no idea whether this wall is coming unless I've played the campaign before at this difficulty level. Hitting this wall means going back several scenarios to make sure my gold/level rollover is sufficient, which is absolutely frustrating.
• Based on fear of the "wall", I will resort to obsessive compulsive play where I will only complete a scenario if I did so in a nearly flawless manner.
• Since I'm not save-loading (I do restart a scenario which I consider to be a failure), I resort to another tactic that's less outwardly cheating but still silly -- using level 1 sacrificial lambs as decoys and abusing the AI. The AI loves to pound on a level 1 that's stranded somewhere, seemingly no matter how far the unit is from it's objective and how much it decreases the AI's odds of victory in the scenario.
• It also occurs to me that many of the scenarios seem to be balanced against preserving the player's units rather than the player achieving the scenario goals. This results in frustrating objectives because of the luck based factors in this game. Bad luck is easier to account for when your objective is as broad as completing a whole scenario/MP skirmish -- the rolls will usually balance out in the long run. It is much harder to account for when your objective is much more acute in scope, like preserving a handful of units. The anvil to the hammer here is that the player in a campaign does not benefit nearly as much from an incredible lucky turn of events as it loses from an incredibly unlucky turn of events; getting lucky and wiping out an enemy unit when the odds were heavily against is not as great boon as getting unlucky and having a critical unit wiped out is a misfortune. This is due to the persistence in player units and the lack of persistence in opponent units.

Really, I think the root of the frustration I experience is that the explicit goals of the scenarios does not seem anywhere as significant as the implicit goals of keeping units from perishing and the resultant gameplay that this leads to is often shallow.

I like the idea of being able to recover casualties via unit actions in scenarios, but it might be in conflict with the principle of K.I.S.S. It might be simpler to just allow the recall of all units after a scenario victory (albeit at a cost). My proposal would be:
Allow the recall of all units after a victory, but units that were casualties in the previous scenario cost more to recall. Keep an "Ironman" difficulty available that allows players to play campaigns with persistent casualties.

• Allowing the recall of all units would allow a player to use these actual units to accomplish the scenario goals rather than stockpiling them for the grand finale. The player could play a more conventional MP style of game within a scenario.
• Likewise, the player could play on a more even footing with the AI opponent -- it would not need to be as stilted as suicidal AI vs OCD conservative player.
• The added cost of recruiting a casualty would still make it unlikely that a outright suicidal strategy would be effective.
• The focus of a scenario could be more strongly tied to the completion of the scenario's explicit goals, and balanced as such. Goals could then be more varied, instead of nearly always being "preserve and level".
• Acute instances of terrible luck would be much more forgiving.
• It could potentially be less likely that player would need to go back several scenarios and "snowball" better, but this would depend on scenario balance.
• The presence of an "Ironman" difficulty where the original style of gameplay persists should hopefully result in little to no alienation of players that prefer it. Different strokes for different folks.
• This gameplay change could be rationalized in terms of realism, if required; as mentioned in this thread, often casualties to a unit in a real war where the side was victorious are not that great. These can be seen as soldiers either being taken prisoner and then forfeited, or routing and hiding behind the line of battle, or not dead but unable to fight in this battle. The extra cost to recall a casualty can be seen as spending money recruiting veteran soldiers to fill out the ranks.

The downsides that I can see right away:

• This could be difficult to rebalance, both in a "we have all the time in the world to do this but it's still difficult" and in a "we don't have the resources to make a big change like this" sense. That said, perhaps this would be a change that could be introduced gradually, campaign at a time.
• Including both this and an Ironman style of play would lead to two radically different types of balance. This might be overwhelming.
• Units might get too experienced. This could probably be balanced by increasing the percent scaler on experience required though.
• Likely MANY unforseen quirks and side effects that would need to be accounted for, that would only show up after being implemented.
• Wesnoth might become something other than Wesnoth. I don't know that this is really a bad thing, since it would become more like Wesnoth Multiplayer.

In conclusion, I don't know if this is the best approach, and it could be that in-scenario recovery is better or that there's a different, better approach. But, I do feel that there is a problem and one that is probably best approached in some way, if resources allow. This might not be shared by everyone, but right now campaigns are needlessly frustrating for me and left alot to be desired. I have deemed them not worth my time at the moment, it's just not my cup of tea.

[Pentarctagon]

I like something about this general idea. But this really caught my attention. Can the AI handle recalls at all or is it technically crippled in this aspect? I mean is it just "the way we do it" that makes the AI-sides lack a recall list?

you can basically do what mp campaigns do except that the enemy side will start off as a human player and then is changed via a start event to an ai side - and then the ai will have a recall list as well.

[Hierbo]

Hi everyone! I'm a first time poster here, but I have read the forums extensively prior to this in a futile effort to avoid sounding like a dummy. My point covers a bit of ground, so please bear with me.

It seems to me that each character in the game actually represents a battalion of indeterminate size. While I have not seen such a thing expressly stated by a game designer, I have seen it stated by many others and not refuted. Also, it only makes sense that the fate of the kingdom of Wesnoth does not rest solely in the hands of 30 guys fighting 30 other guys.

Based on that notion, here is my point:

It seems to me that there is really no chance that a clash of 2 units consisting of numerous troops could result in absolutely no damage to a side, unless 1 side has literally no means of retaliation (such as archers attacking cavalry). I know, I know, I'm talking about the luck thing. Before you banish me for doing so, though, please hear me out. I like the luck in this game. I think it would be much less entertaining without luck. What I am saying is that I think that more plausible results would be achieved if a character's damage were spread out over more attacks, to allow less likelihood of non-feasible outcomes.

For example, if a unit had an attack of 5-6 instead of 10-3, it would be dramatically less likely that he would completely gore his opponent or completely miss him. My solution to this perceived problem is just a notional quick fix that I had thought up, and certainly not necessarily the best one.

I think the core of my point when boiled down is this: I see a problem with the damage output -- the wild random swings glare at me as a lapse in the tactical nature of this skirmish simulator. Damage exchange should follow more of a bell curve, in my opinion, in which the average damage output should be the most likely, just as it is by rolling a pair of dice (7 is most likely, with ever decreasing likelihood as numbers get away from 7). The outcomes of battles, though random, are really the result of countless random happenings, and are therefore much more likely to tend toward the average than the fringes.

I have not heard this particular point before and so I was compelled to post it. Thanks in advance for not raking me over the coals (yet)!