Units Advancing Asymptotically?

Discussion among members of the development team.

Moderator: Forum Moderators

Is this a good idea?

Yes!
15
71%
No!
6
29%
 
Total votes: 21

Dave
Founding Developer
Posts: 7071
Joined: August 17th, 2003, 5:07 am
Location: Seattle
Contact:

Units Advancing Asymptotically?

Post by Dave »

At the moment if a unit reaches its highest level, it can advance no further. Any experience gained is useless.

It would be possible to implement a system where a unit that has reached its highest level can still improve with more experience. The unit would need a certain amount of experience, perhaps 100, to 'advance' again, at which point the player could choose one from a number of 'skill areas' for the unit to improve - 'strength' (unit does more damage), 'swimming' (unit moves faster over water), 'speed' (unit moves faster) and so forth.

To advance again, the unit would need double that much experience - 200 points. The player would once again get a choice, minus the ability she chose last time for the unit.

The gains would all be relatively modest compared to advancing a level -- they'd probably be about as powerful as an extra trait -- and the doubling of experience required each time would ensure that it would be difficult for a unit to get more than a few of these skills.

I think it'd add to the feeling of character building though, especially as the user gets to choose which area to focus on.

David
freim
Retired Terrain Art Director
Posts: 1113
Joined: November 29th, 2003, 11:40 pm
Location: Norway

Post by freim »

Sounds good, it's never fun to reach a cap.
Dobob
Posts: 123
Joined: October 6th, 2003, 9:21 pm

Post by Dobob »

Something that could (emphasis on could) be easy to do :

Everytime one unit advance when it has reached is maximum level, it gain a new trait, up to a maximum of 2 more traits (so it'll have a total of 4). Of course it would need to give every unit a list of possible traits to prevents giving loyal or intelligent. And also you would need a lot more traits :roll: .
miyo
Posts: 2201
Joined: August 19th, 2003, 4:28 pm
Location: Finland

Post by miyo »

High level units are already deadly, and now we are making them even deadlier... I sense uber units appearing.

- Miyo
fmunoz
Founding Artist
Posts: 1469
Joined: August 17th, 2003, 10:04 am
Location: Spain
Contact:

Post by fmunoz »

It's not a goodidea .. it will remove the incentive of recruiting new units...
and it will encorage player to save&reload when their über-unit dies... now lossing a 3th level units is a big loss but if that unit had extra skills...
My vote is to just give "stars" to units every XXX (fixed or variable amount) points... just increase the hp max in a 5% and heal it double that number. After 2 or 3 starts keep just the healing.
freim
Retired Terrain Art Director
Posts: 1113
Joined: November 29th, 2003, 11:40 pm
Location: Norway

Post by freim »

miyo wrote:High level units are already deadly, and now we are making them even deadlier... I sense uber units appearing.

- Miyo
How about trimming down lvl3+ a little, maybe have more incremental upgrades?

Also instead of always improving on everything, a unit specializing on certain aspects of a unit type could become a little weaker in other. Thus a higly specialized unit would become very deadly when using it's primary traits, but more vulnerable in other areas.
Bazarov

Heresy

Post by Bazarov »

This may get rejected out of hand as heresy, but why not apply this concept to all units from the start? That is, when you recruit you get a simple 'Elven Recruit' and then you can add training packages to them, costing XP and possibly Gold. So you could train them in swordsmanship, or bowmanship, or horesmanship or magic, then later add other abilities (including 'attack rate' and 'damage' and 'endurance training' etc.)
Graphics could be assigned based on their capabilites, 'level' (if you need it) based on total XP earned.
Or, recruited as they are currently then progress from there, buying abilities when they gain significant XP.
Yes, this means custom units. Yes it means codifying costs for attack balancing (which you should already have). It could be implemented with the race concept.
It could also be done between scenarios, making the game more like an RPG (another thread).
It could also still use the established unit progression (i.e. instead of buying 'attack rate' for 12xp you could buy Avenger for 220xp) .
Lastly, it would be something new to the genre - I have never seen it attempted.
It would allow for a great deal of fun and replayability, assassin mages, Troll Rangers, Warriors of Light.
Circon
Posts: 1200
Joined: November 17th, 2003, 4:26 am
Location: Right behind Gwiti, coding

Post by Circon »

Heresy?

Bazarov, that's a brilliant idea!

I'll appropriate it, by saying that it HAS been used in other games, and then I'll build on it.

An Elven recruit gets 4-3 for both ranged and close combat. He needs 10 xp to advance, and can become Fighter, Archer, Scout(prelude to Outrider) or Leader.
Human recruits get 4-4 combat and 2-4 ranged. They can become Horsemen or Mages.
Dwarven recruits get 5-4 combat and 0 ranged. The can become Thunderers, Warriors or Lords at 10xp.

Of course, the idea may get rejected for other reasons, such as requiring complete rebuilding of Wesnoth due to major unit changes.
Dobob
Posts: 123
Joined: October 6th, 2003, 9:21 pm

Post by Dobob »

In reply to Bazarov,

Don't you think it would add too much micromanagement? Getting to choose many advancements every turn would distract from the main strategy.

In my opinion, the player should only have limited options, but have a lot of possibilities in how he uses them :
- What he can recruit;
- Where he moves his units;
- In what order he attacks;
- Which attack he uses;
- Which units does fight;
- Which of the 2 upgrade he takes;

All of these are simple, have big difference on gameplay (red mage != white mage) and offer a lot of possibilities. On the other hand, choosing between a +1 on melee attack or a +5 hp (just an example) wouldn't change a lot in how you'll use your champion. Those little bonuses should be kept with random traits, so the player just need to do with what he has.
Dave
Founding Developer
Posts: 7071
Joined: August 17th, 2003, 5:07 am
Location: Seattle
Contact:

Post by Dave »

Bazarov & Circon,

I think your ideas are good ideas, however I also do think they are beyond the scope of what BfW is intended to be.

I would like a game where units evolve continuously, instead of discretely - in big jumps - like is currently done in BfW, but I don't think BfW is, or will be, that game.

But who knows? I intend for BfW to be 'finished' one day, a day that will likely be long before my, and others here, passion for making and playing games has died. Perhaps then, once the team has 'proven' itself, we could tackle an even more ambitious project that includes such ideas.

David
accursed

Continuous unmanaged in level upgrades?

Post by accursed »

Dobob wrote:In reply to Bazarov,

Don't you think it would add too much micromanagement? Getting to choose many advancements every turn would distract from the main strategy.
Dave mentioned that he liked the idea of continuous as opposed to discrete leveling up, but indicated that it wasn't how BfW works as of now...

Just curious, what if the leveling up was continuous (with no choices, the benefits are applied to one of the appropriate traits for that unit randomly by the game) and then *between levels* when you recalled a unit that had advanced beyond it's current level, you could choose what new type of unit it would be... For example, take a mage:

During in level combat, it could gain say one point greater maximum hp (say by killing an enemy), but this could just as easily been a slighly more accurate attack. Throughout the level, cummulatively it may have advanced to the point where it met the threshold to promote to white or red mage, but this promotion does not yet occur *(moreover you may or may not even be able to tell when this occurs)*
Then when you went to recall that unit in the next level, you could see that has levelled up (when you go to choose it) and as you recall it, you determine the advancement path (eg. red or white mage).

Is this reasonable compromise?
Dave
Founding Developer
Posts: 7071
Joined: August 17th, 2003, 5:07 am
Location: Seattle
Contact:

Re: Continuous unmanaged in level upgrades?

Post by Dave »

accursed wrote:
Dave mentioned that he liked the idea of continuous as opposed to discrete leveling up, but indicated that it wasn't how BfW works as of now...
I think I should clarify this by saying that I like both. I think both ways of doing it are good, depending upon the emphasis and style the game is meant to have.

I also like games where you can build buildings and cities, but that doesn't mean that we want those features in Wesnoth.

BfW's advancement style was designed with a few things in mind: (1) simple; (2) fun; (3) a little player interaction. I think it has achieved this well.

David
Dobob
Posts: 123
Joined: October 6th, 2003, 9:21 pm

Post by Dobob »

To accursed :

If someone manage to find a nice way to do it, I'll have nothing against it. The way it's done shouldn't interfer much in the way a player act, like forcing him to use a lot the ranged attack of his units to increase its strength. That's why I prefer level based upgrade, but I don't mind if it's like it is now, or if it is made more continuous (example : get three mini-levels or stars and then get your real new promotion to another class and higher level).
miyo
Posts: 2201
Joined: August 19th, 2003, 4:28 pm
Location: Finland

Post by miyo »

Wesnoth should stay as it is. Keep it simple, so we can finish it, otherwise it will become other massive project that never finishes. Anyway, many of us really like the game, we say that it is addictive, even without all those complex features.

After our developers, contributors and forum members have proven worthy in the 'Battle of Developing Wesnoth' we can consider starting another project... re-using work in wesnoth, taking advantage of 'lessons learned while doing Wesnoth'...

- Miyo
Dave
Founding Developer
Posts: 7071
Joined: August 17th, 2003, 5:07 am
Location: Seattle
Contact:

Post by Dave »

miyo wrote:Wesnoth should stay as it is. Keep it simple, so we can finish it, otherwise it will become other massive project that never finishes.
Indeed. We don't want to bite off more than we can chew. Many projects seem to be executed using the following algorithm:

Code: Select all

while(bit_off_more_than_can_chew() == false) {
    ++features;
}
We don't want to fall into that trap.

David
Post Reply