Community Map

[cohadar]

As long as we are here, we might do some map development.
Does anyone have any ideas for a map type that would be suitable for development by a large group of people?

[2-P]

Lol this one never gets old.

[cohadar]

I know right.

But WHY does it never work?
Answer that and you will answer the first question.

[Rao Dao Zao]

Originally Posted by cohadar

But WHY does it never work?
Answer that and you will answer the first question.

Because we are all unique snowflakes that can never agree on anything.

[Kino]

^^Agreed

Ego and elitism probably play a major role.

[cohadar]

Ok than, now we just need a map plan that can handle ego and elitism.
Which means it has to be extremely flexible.
Any participant should be able to express his input on the map and in a such a way to avoid conflict with others.

We don't have a general idea for the whole thing yet, but lets break the problem into parts.

Lets take spell development as a problem segment.
In order for spell development to be distributed to multiple people following conditions need to be met.

1. we need to agree on tooltip format
2. we need to agree on some basic balance guidelines
3. we need to agree which libraries to use

The last item is potentially the major source of conflict.
We can solve it by this rule:
You can use any library you want as long as it does not have side effects!
This means no libraries that use UnitUserData or that have significant performance impact or can in any way negatively interact with other libraries or spells.

There are people here who worked on joint map projects.
Please describe your mayor sources of conflict you had so we can all learn from the experience.

PS: do we still have any terrainers here?

[Kyrbi0]

Part of the problem is, that the problem is (as XKCD put it) "nested fractally". You say we need to decide how spells are made, which boils down to 1. 2. 3. The problem with those is mainly 3, with whom mainly

You can use any library you want as long as it does not have side effects!
This means no libraries that use UnitUserData or that have significant performance impact or can in any way negatively interact with other libraries or spells.

is the problem.
However, who decide(d)s what is the problem in the first place? Or how much of it is a problem? Go back far enough, and is there anything that is basic, primal, or manifest?

/philosphize.

~~~

I would enjoy working on things with others, but I am pretty "niche" both in the skills I have and the projects I like to pursue. Not sure I would have much to offer help in, aside from coming up ideas.

[cohadar]

The first definition is called an axiom and we accept it as truth by definition.

So I am going to start with some axioms.

1. The purpose of the project is to create a map together, all else is secondary.
2. Map should be simple in concept and it should be possible to make a playable version in less than a week.
3. Although it will start small it should have the ability to grow.
4. It should be multiplayer not singleplayer (random choice)
5. It should not be an AoS (seriously not)
6. It should not require full house to play (practical choice)
7. It should be able for people to contribute with less than an hour of work.

[Rao Dao Zao]

Originally Posted by cohadar

7. It should be able for people to contribute with less than an hour of work.

I think this is probably the key. Everything needs to be done in tiny self-contained units of work, as people on the internet are inherently unreliable. We all talk the talk, but most group projects fall apart because the helpers rapidly lose interest (because it's not their project).

[cohadar]

There could be a way to both make it possible to do meaningful work in less than an hour AND make people feel it is their project.

8. Map shall have heroes and heroes shall have random spells.

That way one hour of work could be creating one cool spell.
Creating a hero avatar would also take less than an hour since hero avatars would have no default spells. So it would just be choosing a model, giving it some proper hero names and possibly some art attachments.

Those artistically inclined can also submit their skins and models without the need to code spells to match the game.

And in order to avoid the total spell randomness, spells will have following tags:
melee/ranged/any
STR/AGI/INT/any
Human/Orc/NElf/Undead/any

That way you can limit which spells can go to which avatar.
examples:
Blink: any/INT/any
Marksmanship: ranged/AGI/any
Cleave: melee/STR/any
Raise Undead: any/INT/Undead

[Anitarf]

This is a good thread, not because I think the project will ever get done or even get off the ground but because I think there is potential for having some interesting discussion about how such a thing would best be done before ultimately not doing anything. :)

Originally Posted by cohadar

The first definition is called an axiom and we accept it as truth by definition.

Is that a royal "we"? :)

Anyway, I like the first axiom candidate, stating that in such a prominent place should help contributors keep their priorities and expectations real. The worst thing that can happen for such a project is for creative and personal differences to cause conflicts and discourage people from contributing. By reminding people of the fundamental nature of such a project, you might make them approach it more open to compromise.

I'm not entirely sure the second axiom candidate is necessary. The map obviously needs to be sufficiently non-specific to facilitate varied contributions; as such the base systems likely will be fairly simple. I don't think we necessarily have to impose a rule on how much time they should take to make, though; the time needed for development, whatever it may be, should be a consequence of above considerations, not an axiom assumed at the start.

Third, I agree. While content creep might not normally be a good thing, here it's kind of the point. Again, we're not talking about making a perfect game here, but a project where people can contribute for the fun of it, whatever and whenever they please.

Fourth, I kinda agree. You have multiple people making it, multiple people should be able to play it. Ideally, it should be playable in singleplayer as well, this strongly implies a coop game, which seems appropriate to me. It fits in spirit with the cooperative nature of the development as well as saves you the effort of fine-tuning the balance for competitive play. Obviously, a coop map can get away with far less fine-tuned balance. If there are any fans of competitive gameplay out there, they may argue against me on this one though.

I don't think why the map should automatically not be an AoS. Sure, my thought process above precludes that, but if that gets overruled by enough contributors asking for a competitive game, I don't see why an AoS shouldn't be on the table.

Agreed on point 6, as noted earlier my preferred option is a map that can be played by any number of players.

On point 7, I'm not entirely sure about it. The first thing I thought of when considering what such a project could be like was a minigame map: each user could contribute their own minigame(s), there'd be very little conflict between contributions as they'd be stand-alone games packaged into one map. I can see the advantage of making user contributions much smaller pieces of a much larger whole, it would certainly reduce the barrier to entry considerably. Making individual heroes and spells for some sort of generic hero map is pretty much your only choice in such a case, though; that's not necessarily bad, though, I have to keep in mind what I wrote in regard to axiom candidates one and three.

With regards to 8, random or shops? My first thought was skill shops but considering the potential for content creep it might be more fun for players to just get some random spells and start playing than having to navigate 10 different skill shops. Also, there's even less pressure to perfect the balance as weaker skills will still get played as much as stronger ones; on the other hand, this setup forces the coop gameplay even more, which as I noted earlier is fine by me so I'm kinda liking the random idea.

Not so keen on the classifications, some range/melee exceptions are fine, primary attribute limitations I like less, although they might make sense for really costly ultimates or obvious combat passives, racial limitations I like the least as limiting hero designers to just the four default races when even the WC3 melee game itself has neutral heroes seems kinda excessive (on the other hand, if you allow for more races, you end up fragmenting the skill pool which greatly reduces the number of possible skill combinations a player can get). If you want to differentiate the heroes in some way, you could always allow for an innate skill in the design.

[cohadar]

Regarding the map type.
==================================
Lets imagine for a moment that we have a bunch of heroes and ton of random spells already made.

What kind of games could you make with such combination?
It should obviously require little need for balancing.

Coop is not a good answer here because you need to balance against computer creeps.
AoS - not a chance, balance is everything with this game type.
Team vs Team hero survival (ENFO, Rabbits vs Sheep, ...) - again not good due to balance.
FFA arena? - again not good without at least some balance.

And we by definition cannot balance spells!!!

The answer is not in the game type but in the game mode.

And that mode is: MIRROR

Coop mirror mode - x heroes play against x bots that are mirrors of themselves.
AoS mirror mode - 2 teams with same random heroes on each side, it could work.
Team vs Team hero survival with same heroes - balanced by definition.
FFA arena? - everybody in each round gets the same random hero with same skills.
Team vs Team arena? - again with mirror mode, balanced by default.
etc...


Regarding the classification.
==================================
There must exist some spell/hero classification, otherwise every hero will feel like every other hero.
There will be no personality to them. So I am going to call a design shot here:

Only 4 default races are allowed. You can include heroes from other races but you must make them join one of the 4 big factions.
If and only if one non-default-4 race gets enough content: for example people make tons of contributions to Naga heroes and spells,
than that faction can break apart and form an independent race (in game mechanic terms)


Spell level problem.
==================================
We could just make this a comunity hero/spell content engine and let people make any game type from it they want.
(Althought we should definitely have a default game implementation)
What bothers me most now is the question: How many levels should spells have?
I have no idea how to answer this as to make it most generic for gameplay.

[Anitarf]

Mirror is boring though, asymmetric gameplay is far more interesting.

With regard to making heroes distinguishable, you ignored my suggestion of giving each hero an innate ability. This would make each individual hero different without imposing limits on the race of hero submissions, unlike your suggestion where two ranged intelligence heroes from the same race could still get exact same skills.

Balancing against computer creeps isn't such a big problem as you make it out to be. Giving the spells roughly similar power is not such a big problem and since they are randomly assigned, you will in most cases get an average spell power level. Some games may end up being a bit easier or harder but that's not a big deal, you can always implement gameplay mechanics that allow players to progress regardless of their power, just making the progress slower for weaker teams, which is a good gameplay mechanic anyway because even in a perfectly balanced game there's still imbalance in terms of player skill and such a mechanic makes the game more accessible.

I have another specific consideration: spell hotkeys. Should they be based on spell name or spell icon position? In case of the latter, should spells have a fixed slot or would contributors need to make multiple copies of each spell, one for each slot?

[cohadar]

Innate abilities on heroes are out of the question. Too much hassle.
Mirroring can be assimetric. (avatar swap)

Multiple copies of one spell will not work because of shared cooldown.

The solution for that is to group spells by hotkeys/icon positions.

Best system I came up so far is to group spells into categories by casting type:
* unit friend
* unit enemy
* area + point
* immediate + summon
* passive + autoattack

and assign a hotkey and icon position to spell based on that category.

[shadowvzs]

hehe i also started alot times making a map but dont really finished but my tip is make a open rpg style, i started on another forum, but it is normal jass, just use better villager animation model, dgui potion bar/health/mana bar.

but u can check my map, i know all of u better coder than me but maybe u will get few idea :)

http://www.hiveworkshop.com/forums/s...m-item-230191/

(anyway i guess that map never will be aproved anywhere coz takes too long to review the code if it isnt organised with libraries but u can try and check out in game )