Hero Design - The most important aspect of most maps.

I have seen several painstakingly crafted maps with great-looking terrain and interesting gameplay dynamics; yet the map is unpopular, and people get bored and quit within minutes. Much of the time this is due to poorly designed heroes. In many maps (AOS, arena, and others), the primary way the player interacts with the game is through their hero. Even a great map with great gameplay is worthless if the player's hero does not facilitate interesting and dynamic gameplay. Conversely, a poorly designed map with fun heroes will be fun enough to play for some time (until you get tired of using the hero skills and discovering strategies). This would seem to suggest that hero design trumps all other aspects in terms of importance and fun-factor for many types of maps.

Let's look at the design of a typical hero:
Skill1: A directly targetted attack on the enemy that deals some kind of damage. 10 second cooldown.
Skill2: Perhaps some kind of support, utility, or defensive skill. Situational use only.
Skill3: A passive skill.
Ultimate: Powerful spell with a long cooldown.

Let's look at what these skills translate to, gameplay-wise.

Skill1: Click it on the enemy every time the cooldown is finished.
Skill2: The more situational the skill, the less often it will be used. Either way, it will not be used as often as skill 1, generally.
Skill 3: Never used, and the effects are constant (and therefore with little dynamic effect on the gameplay) or random (oftentimes taking strategy out of the hands of the player).
Ultimate: Only used when the long cooldown is finished. The rest of the time this skill may as well not exist.

So what does this mean for the player? They have one simple ability that they use often, another situational skill that may as well not exist if the situation to take advantage of it does not exist at the moment, a skill that gives the player almost no interactive opportunities, and a skill that they can only use a handful of times throughout the entire game. Furthermore, mana costs may limit the usage of these skills even further.

Basically, most heroes spend 80% of their gameplay time autoattacking weak NPC controlled units. This is not fun. There is no reason that heroes should be designed with this type of limited gameplay in mind. There is room for a revolution of WC3 gameplay by redefining how heroes are played to make sure that the skills present dynamic opportunities, interact with one another in interesting and unexpected ways, and making sure that the hero can use their skills OFTEN. Why not have them pressing a button every other second? Why not give a hero 5 skills, 6 skills, or more, with built-in combinations and strategies? As a sample of this design philosophy I will present one of the heroes in my map, which is currently in the early stages of development.

Model: Blademaster
Special: The hero always has 100 max mana, with negative mana regeneration. Gains mana when striking enemies. (Think rage from WoW.)

Skill1: Critical Strike
Not your default critical strike. Gives an x% chance for y% damage bonus. Killing an enemy with a critical strike gives an "Executed!" message, which gives the Blademaster bonus rage, causes a roar graphic, and applies a temporary damage bonus that applies to autoattacks and abilities. The chance for critical strikes and the "executed" proc apply to all abilities.

Skill2: Bladefury
The Blademaster teleports to all enemies in an x degree frontal cone for y range and slashes them. This skill has multiple uses: It's extremely hard to target or damage the blademater when he's moving from the effects of this skill. It works as a sort of blink-strike against a single target. Its primary use is to damage a large crowd of units. 10 sec cooldown, 20 rage cost.

Skill3: Bladestorm
Not your default whirlwind skill. Damages, knocks back, and pushes all nearby units. The closer a unit is to the center of the Bladestorm, the more damage they take and the stronger the knockback. Optimal damage comes from moving directly toward a target while spinning, or trapping them against an obsticle. Yes, this skill causes a lot of tiny critical hit messages to pop up. 5 sec cooldown, 2 sec duration, 20 rage cost, magic immunity while active.

Skill4: Headsplitter
Primary single-target damage skill. Requires at least 50 rage, and sets rage to 0 when it is used. All rage above 50 is converted into bonus damage. This skill can do an extreme amount of damage if the hero has a damage bonus from killing, has extra rage, and gets a critical hit. That's why it's avoidable. It is a slow-motion matrix-type slash, which gives the enemy 1.2 seconds to run away. Best used in conjunction with a teammate's stunning ability. This can hit 2-3 enemies if they are all directly under the impact area of the sword. 5 second cooldown.

Skill5: Charge
Charges toward a unit at high speed. If you successfully reach the target you deal damage and gain bonus rage, as well as stunning the enemy for 1 second. If you collide with something else first, you gain nothing. Has a minimum range. 20 second cooldown.

Ultimate: Blood Rage
For the next 30 seconds, every kill the Blademaster makes causes him to become more enraged. He grows larger (he can become quite giant if he manages to kill 10 or 20 units during these 30 seconds), turns a deeper shade of red, runs more quickly, and gains a critical hit chance and critical hit damage modifier bonus for each kill. All of these bonuses disappear after the 30 seconds are over. Note that the increase in size also increases the range of his Bladestorm and Headsplitter abilities. 2.5 minute cooldown (2 mins from the time the ultimate ends).

Obviously, there are many issues with using heroes such as the one described above:
First and foremost, these types of skills require moderate to advanced JASS usage.
Second, they can cause users to lag severely if memory leaks, excessive graphics, or code inefficiency cause problems.
Third, they can be extremely difficult to balance (note the exponential scaling of the hero described above: his damage depends entirely on how fast he can kill, therefore +damage items will have an extreme effect on him).
Fourth, it takes a great deal of time and effort to create heroes as complex as the one described. Certainly, several less-complex heroes could have been created with the same amount of effort.
Fifth, using 6 skills requires you to create an alternate menu for heroes to select level-up skills.
Sixth, if the skills are not intuitive enough, such an extreme change from standard hero gameplay may confuse players.
Seventh, it takes a while to design a hero with this many skills, and plan ways that the skills can interact with one another to form interesting strategies.

However, consider the benefits of taking hero design to such an extreme:
First, very few on Battle.net have seen such gameplay before. It provides an instant hook and distinctiveness for a map.
Second, complex skills interact with one another in unpredictable ways. When the blademaster is charging toward a target and someone nails him with a knockback skill, causing him to charge in some crazy direction, or perhaps he gets knocked into the air by a skill and teleports back down with bladefury... these kinds of things are possible as a natural consequence of the skills interacting; not necessarily as special cases that need to be coded individually. It means even you as the map maker can be surprised by the strategies that emerge.
Third, it's fun! Even if my map is 1% complete and only has 3 heroes (two more that are similar in complexity to the ones listed above), I recieve many positive comments about how fun the map is. However, I know better; the map itself is pretty much crap right now. Just having a fun hero makes people believe that the entire map is fun.

Anyway, I would like to hear thoughts and opinions about this method of hero design; if it's a new concept to you or if it's already been done in x map; if it's the next evolution of WC3 gameplay or viewed as impractical for some reason, etc. Personally I am dedicated to making these kinds of heroes work, despite all of the additional problems and complexities that arise with them.

Well, all skills you wrote down are already existing in the one or other way.
A Hero shouldn't have too much skills, because many players won't use all -> the same problem with the 4 Levels.
But you're right the hero design is often poor, well, i built a Hero Defence some time ago and you have 1 Base Hero with nothing, then you have to set all points (20 stat points) and which hero you get depends on the balance.
Also the skills became much stronger, when you made a hero with a high int balance, etc., but it wasn't that popular, because many players didn't like this way of creating a hero (they made heros with terrible balance, well, not my fault)

You missed one key point.

KISS.

Your ideas are very difficult to pull off (detecting attacks is nigh on impossible accurately), and are too complicated.

Simplicity breading complexity is the winning formula, rather than complicated-ness, which doesn't work unless very intuitive and behind the scenes.

Name an example. You could only do extremely limited (both in terms of customizability and strategic potential) versions of these abilities using pre-existing skills.

How many skills do you think are "too many?" I can't possibly imagine the kind of person who can't handle using 6 skills, most likely with one being a rarely-used ultimate and one being a passive, meaning only 4 buttons to press with any kind of regularity. When you play WoW, do you think that having 4 or 6 buttons would be "too many?" No, that's way too few. I'm talking about getting out of the mindset that heroes should only have 1 or 2 buttons to press. It's quite boring.

If you are the one choosing what kinds of skills are available, it's also your responsibility to create skills that are interesting and balanced in various combinations... one of the biggest challenges of a custom hero map.

It's only "nigh impossible" if you don't use JASS. In fact, Vexorian's HitDetect engine makes it extremely easy if you are at least moderately good with JASS.

I will be frank. I don't think you actually envisioned the gameplay as it would be using these skills. If you had, you would realize that the skills themselves are all extremely simple to use and intuitive. The complexity comes from their interactions and the nuances result from their knockback and movement components, exactly as you stated it ought to be. But you are right, it is senseless to add complexity for the sake of complexity; the skills themselves must be simple, and evolve in intuitive ways.

Complicated heroes with lots of spells works great in RPGs and other hero-based maps. It would overcomplicate the already intense microing in a melee game thou.
(note: didnt read the whole post so I dont know if you mentioned any special kind of game-type)

The only ways (as far as I am aware) to in any decent way detect attacks involve:

- Massive usage of damage events.
- Either giving every unit an on-attack ability, or trigger all spell damage.

Those aren't sacrifices I'm particularly willing to make, especially since it involves leakages (events cannot be destroyed).


I'll be frank:

ToB has got some of the best hero design. They did it without damage functions, etc. And they did it with mostly edited abilities with small changes. Yet they work together well, and fit together.



I'll tell you this:

You fight WC3 mechanics, you'll lose.

Yes, playing heroes of this type would be extremely inpractical if you needed to micromanage an army at the same time. That is why heroes are not designed this way in the melee game. In my map, spawned allied units are trigger-controlled.

The spells sound like they take little thought to use - they are all damaging spells which can be used to slaughter massive amounts of enemies. While they are cool (because killing massive amounts of enemies is always fun), it all seems very shallow. It looks like the behind the scenes work is overly complex while the part actually seen by the player is too simple. This kind of stuff is enjoyable for a short while, but after long amounts of play becomes repetitive. However it really depends on how your game is set up - it's hard to judge your abilities without knowing what kind of gameplay they'll be put in.

But back to your original point, of abilities in general being lacking, I agree. The main issue is the effort it requires to make large numbers of skills while still having decent variety in hero/class selection (because players want both!). Most people have been around a while, and seen all the usual spells. They want something new and interesting that not only looks and behaves cool but brings new concepts to the table, without being too complex.

Also, why are people so fixed into using mana and a cooldown? Because Warcraft 3 does it so that makes it the only way to do things? To me these things seem to slow the game down to a horrible pace - unless there are other things going on. Of course, mana and cooldowns have been proven to be effective, but to have these things it requires either other units to control, more abilities, or more strategy to make up for it. Just look at Warcraft 3, compared to Diablo 2, compared to WoW. Warcraft 3 has mana, cooldown, and only a few spells, but you have to control other units and execute strategy at the same time. In Diablo 2 you only control one unit, but mana regenerates very quickly, there is no cooldown on most spells, and you'll have a bit more choice of abilities than war3 (depending on how much you focus on one area). In WoW you have mana, cooldown, and you only control one unit, but you have a large number of abilities to use. In a genre like an RPG or AoS, something needs to be done to give the player more to do.

http://www.wc3campaigns.net/showthread.php?t=79545
Check here. This system uses on-attack abilities and allows you to compensate for it by including triggered orb effects. No leaks.

I agree that ToB has some nice heroes. It's the most fun AoS I've played, really. What methods they used to accomplish that are inconsequential to the player as long as they do not affect gameplay negatively or cause lag.

If you believe the skills I mentioned require you to "fight WC3 mechanics," you may be underestimating what the engine is capable of. An example of fighting the mechanics is trying to convert the game into an FPS. Even the best FPS mod will be constrained by limitations of the game engine, as that is not what this engine was designed for.

Hero combat is one of the main things this engine was created for. JASS perfectly facilitates these kinds of dynamic skills without having to make the gameplay awkward and without causing lag, as long as it's done with great care. This is why I mentioned it as one big disadvantage: it requires much more development time than usual.

What I will say about it is that you may change you mind if you played as the hero, because you aren't considering the full implications of the abilities. The blademaster's skills are not simply used for damage: they are used for movement and defense as well. They all fit a specific theme that creates a distinct style of gameplay for this hero, but the skills are muiltipurpose in ways that aren't superficially obvious. For example: Charge is offensive and defensive. Charging away from an enemy can often save your live. Bladefury is extremely useful for teleporting beyond incoming spells or out of a shockwave that's shoving you. A well-timed whirlwind can block an incoming magic projectile; etc. And I can assure you that the "behind the scenes" aspect of every spell is only to provide what's necessary for a variety of strategic possibilities with a skill. Does it make the skill "too complicated" if you can deal more damage by whirlwinding someone into a wall? On the contrary; it adds no complexity, just a possible strategy that a player can choose to take advantage of.

However, I did not create this thread to continually defend myself against criticisms of a specific hero design; this thread is about a school of hero design. It seems that this departure from the norm may be so radical that people will not believe it's practical until they've experienced it.

Interesting debate

My two cents :

I think a crap really new we could add to heroes are element based spells...
Or an other attribute system (D2 like, or like the one that did Oger Lord)
Your blade master is great to kill enemies... but... I think ability aren't that complementaries ... Take for example the Pandaren : First ability is too damage, if you use it with Alcohol you'll get a bonus effect, it's hard to place because most of time, we do not get mana enough, the last one ability is here to have a bonus without wasting mana...

Here is an idea : when an enemy undergo the effect of an ability, she is wounded and lost 1 armor point (or more) they are stackable. Classic BM make hight damage, if your one would deal about 10 damage, we'll really need to weaken enemies this changes gameplay.

We need something different than "Attack and move till this point" wait to use ability... earn money... buy item... Most of time, this is made by the map, this doesn't come from heroes...

I though about 6+ active Spells, in a "slow" game it's ok to have many, but when you make a "fast" map with so many active spells some players will get troubels.


The Spells was nice and balanced, but many players didn't want to try a unknow combination, so they got bored and there were some pretty nice heros, but i doesn't play a role, because i couldn't restore the map when my hard drive crashed, but i think i'll start a new Hero Defence when i'm done with the other projects.

Yea, I would need to play your map or know more about the gameplay to truly criticize the spells. The way you're describing their use within the game sounds great actually ; ).


I find synergies that are blantantly obvious like this to be kind of dumb. They add a bit of strategy..but it's really not much.


I think 6 spells is a perfect amount, and I don't think a player can effectively use any more than that with consistency. You have many more than that in WoW, but you only use certain ones in rare cases - mostly you'd use the same 4-6 spells depending on what you were doing (grinding, farming, pvp, raiding, etc). 6 spells allows the player to become comfortable with them, and doesn't give him so much to remember that he can't use the spell quickly when needed. However, 6 spells also allows for decent variety and lots of strategy.

I thought you could only have 5 hero abilities? a lot of normal abilities but only 5 hero abilities... I probably missed something didn't I?

And JASS abilities aren't always the best. Sure it allows for more control and ability diversity but still, I prefer dealing with regular abilities in the World editor, but switch things around a lot.

BTW, how does Blizzard make abilities? not just taking a model, I'm talking about starting from scratch and deciding "this ability will make a missile that breaks into more missiles then breaks again. it will deal damage and slow enemy units down"? I was just wondering.

Hardcoding them within the engine, and giving a few customizable fields.

No, you're right... you can't have more than 5 hero skills unless they are added through a trigger, but you can't level up the added ones. That's why you either have to have one unlevelled skill (perhaps it can use triggers to scale automatically with the hero's level), or include it in your custom level-up system to make it levellable.

I don't see how anything that alows more control and diversity isn't best. Sometimes it can be easier to use normal abilities if you don't have to customize them. But once you develop several abilities in JASS you can start using them as templates for new ones, and it doesn't take very long. Also, dealing spell damage without triggering it means you can't modify it in-game without replacing the skill or changing its level. I like to be able to modify spell damage on the fly, like for items that increase spell damage.