(Part 1, Part 2; Part 3)
An esport has many demands placed on it. There are three main categories, visual aspects, viewer aspects, and gameplay aspects. I will cover one of these aspects per post. No document on this subject can be exhaustive, this will not take you from total loss to rockin' boss. But we take a peak at the esport strength and weaknesses of several games, and demonstrate paths to improving esport game design.
Our aspect today it...
The first thing that comes to mind when talking about any sport, is watchability. Depending on the game you're making, this can mean many different things. The screen can be a mess, we have to identify what units and abilities belong to what player, and we have to have those cues for suspense. Also, if you plan engaging anyone besides the core audience, if you want people who have never played the game to watch and become engaged, abilities must visually represent their effect and units must visually represent the role.
1) Intuitive Representation
The psi storm is an excellent example of and ability done well. It just looks painful to stand in thing. It's also easy to identify which player belongs to and which player it’s hurting, because you don't throw lightning on your units.
On the other hand, when watching a DOTA2 game, I saw a character that was a sort of fairy like thing, with blue and wings. It gently roll the ball of light towards the enemy with a charming blue color. When that ball passed through the enemy, the enemy took damage, yet it seemed to just continue on its merry way. The first few times I saw the spell in the middle of the team fight, I had no idea what it did. There's no smash of impact, no recoil of the characters, no parts that indicates it would be painful to interact with. It looks more like a ball of magical Kool-Aid I'd like to drink.
Then, I learned that this little blue very thing could teleport to where the ball was. Now that is a cool ability! But nothing about the game visually communicated that. You can't understand it without it being done several times or being told. That might not be such a big problem when were only talking about 5 or 10 units with similar "you have to tell me" abilities, but when you're talking about a roster is deep as DOTA2, it severely complicates the viewing. I've watched about six hours of DOTA2 content, and I still have no clue what's happening half the time and team fights. The visual language is messy and inconsistent.
Imagine for a moment, instead of the blue fairy thing, the character was a Phoenix or fire sprite. It spits this fat fireball at the enemy, instead of a merry blue sphere. Now that thing looks like it will hurt when it hits you. You don't have to explain it to anyone, dodge the fireball, all mothers have told us this since we were young.
Now, not only does the spell looked painful when it makes contact, but being a Phoenix or fire sprite, we know that the character has an innate connection to the fire. It's not just a catapult throwing a fireball, this creature is born of fire. So, when the character teleports to the fireball's location, it makes a little more sense. Yeah, a Phoenix could appear out of the fireball, we think.
That champion is a really interesting champion, and I really enjoyed watching it be played once I understood what was going on. But by having such counter intuitive visual representation of abilities on so many characters, it creates a steep learning curve to understand what's happening in a match on the most basic level, alienating a larger audience.
2)Which Spells Belong to Whom
Another thing I find challenging when watching a DOTA2 or even some LoL team fight matches, is knowing which spells belong to whom. In Starcraft II, the psi storm is shredding the enemy army. In League of Legends and DOTA2, you're casting spells that are supposed to do damage right over your team in a team fight, and when everyone's doing that, it's hard to know what hurts and what doesn't. Instead of watching the action, I find myself watching the health bars, a much less visceral experience.
Recently, League of Legends has added a mode coloring spells the color of your team. This helps, but often the differentiation is a ring around the ability, a 2-D interface element that is the visual equivalent of sticking a flag on it. While it gets the job done, I have to look for that flag, it makes observing the battles a more intellectual calculating of which abilities belong to whom, rather than an intuitive one.
Let's look back at our blue fairy thing from DOTA2. So we turned it into a fire sprite, and now tosses fireballs. It helps that the fireball comes from the sprite and moves towards the enemy, but let's pretend for a moment that it also summoned a ring of fire, locking enemies in unless they want a healthy case of third-degree burns. Now, let's assume there's also a fire sprite on the enemy team. If they both throwdown fire rings and the team fight, we don't know which one is going to hurt the team were rooting for and which one is going to hurt our enemies. It's just a ring of fire, that's all we know. So we need to communicate more information. We could put a color-coded ring through the fire, but there is a better way.
Have one team themed with blue magic, and one team themed with red. We have hot, natural gas-like flames for our blue team's fire sprite, and classic red flames for the other. Now, when you see champions with blue markings walking through the red flames you know the cringe.
Envision this scene: our champion is caught in the enemy's blue ring of fire. She's low on health, out of teleports, dodging the enemy shots. She can't walk through the flames or she'll die. We scoot to the edge of our seat; will she pull it off! Well, we don't know. Unless we seen this ring used 20 times and already have a sense of how long it's going to last, we're not going to know how much longer she needs to survive. We don't see the goal getting closer, we simply have to take it on faith that it is. Not knowing that key piece of information sabotages the suspense of the moment. Watch 20 seconds of this:
http://www.youtube.com/watch?v=rFllSVjK5hY#t=4m20s
You can see just how close he is to killing the probe, and dying himself. If we didn't have the health bars to tell us it was that close, that would be a lot less exciting. So we need a way to communicate when our ring spell is going to end.
And fire has just the thing, when it starts the flames are thick and tall, but as it nears expiring, the flames die down. When they shrink far enough, they go out and, and our spell ends. Now we can see exactly what were waiting for.
This... introduces another design challenge: how much damage does this thing do? A ring of flames, well, you know you want to avoid it, but doesn't exactly seem that powerful. Earlier we said this is a powerful spell. How do we communicate that? We can just make the fire taller and brighter, that would begin hiding the action and overwhelming the screen. And if the flame dies down toward the end, but still does the same damage, we’ve created something that looks like it should hurt less, but doesn’t. This is a prime example of visuals failing their task.
So we needed to do something else that really looks like it hurts if it's really going to do a lot of damage. Well, what has to do with fire that looks like it hurts? There's explosions, and catching on fire. So now, when an enemy player walks through the ring, they explode and catch on fire for a moment with the opposing color flame.
Perfect, everything you need to know about the spell is communicated visually. You want to avoid it, you can see how long it'll last, when you step through it you can see it hurts and how much, and you know which team it belongs to. All without any labels or any text reading.
1) Now hold on a moment, Starcraft II doesn't color their spells. They express the individual style of the race through whatever color suits them, and even when it's a mirror matchup the spells are not colored to differentiate.
Each game has different demands. In a MOBA, you can have 100 champions, you walk into your own damage-dealing spells, and there are a lot of spells going down once. It's a key part of what makes it so much fun.
In Starcraft II, we don't have 100 different styled units, we have 3 styled races, and the spells that hurt you're usally not walking into. Most of and engagement focuses around unit positioning and attacking. There are only a few spells to learn, and they are easy to understand. Starcraft’s visual languages potent, and so the coloring is not needed. I have never been lost watching a Starcraft match the way I've been lost watching DOTA2, even before I knew Starcraft.
2) Shouldn't that little faerie creature have its own color theme? By forcing certain color spells, you take all the color creativity and feel out of the character.
This is a legitimate concern, since people connect with color so strongly. But, if your character roster is deep enough, the benefits to spectator sport by having this consistency strongly outweigh the negatives. Color creativity can be expressed on the characters. Creativity can also be communicated through shapes.
For instance, a character who smashes the ground with his fist and a crystal spike pops up somewhere else, damaging enemies nearby and knocking them back. This feels fundamentally so much different than our fire sprite. This crystal can be colored according to the team color. Perhaps the crystal is a very bright version of the team color, glowing. Other characters can have sickly looking swamps appear with the team color, which look and feel completely different than the bright clean clothing crystals. Or throw spears with speedlines the color of their team. The individuality of the character's abilities can be expressed through brightness and shape, character model, and animation.
A side note: one thing that bothers me about MOBAs is that the importance of killing minions is not clearly communicated. Killing 15 minions gives you the same buying power is killing an enemy champion. I mean, that's a huge deal. A good way to visually communicate this is to have a "taking their life force" visualization, so that when the champion slays minion, light is drawn from the minion to the champion and he glows for a moment, as if adding the minions’ power to his collection.
Also, they don’t visually communicate when things get stronger. In SC2, the Marine receives a shield, the siege tank goes into siege mode, and almost all the champions in League of Legends remain the same. The only visual clue we see to them getting stronger is when aura items create glowing cryptic symbols beneath their feet. I honestly didn't realize what they meant until I played 50 matches, that's how poorly they do it instinctually communicating what's going on. Sure, if you know what all the symbols mean he can intellectually understand what's happening, but it's just not the same as seeing a Giant Cho'Gath, a monster champion that gets larger as he grows stronger, storming down the lane.
3) Distinctive Shapes
There's a thing in cartooning called the silhouette test designed to see how identifiable and unique your characters are. You take all the characters in your story, make them look like shadows, just black silhouettes, and place them side by side. If your average reader can identify every one of the characters, you've done a good job creating distinctive shapes.
We can apply a lot of that test to designing units. We can recognize the silhouette of every Starcraft II unit. This means when we only get a glimpse of the Starcraft unit of the corner of our eye, or the screen jumps back and forth between locations and we only see units for an instant, we have a much better chance of recognizing the unit, even though we only see it for split-second. This is especially important for any game for the units are moving at a very high rate of speed and teleporting around. DOTA2 also does this well, and the one think I've been able to do while watching it is keep track of the units.
I talk to someone on the Unity3d game developments forums who is interested in creating their own MOBA. Their plan to make it stand out from the competition was to have fun characters and more sophisticated graphics. The first one is a great plan, although I doubt it's enough to carry the game success. On the other hand, more sophisticated graphics, more realism, is actually a detriment to the MOBA genre.
Stylized graphics are great for fast-moving content. They work with our brain naturally sees the world, simplified shapes with most of the data being thrown away. Imagine you're running through the house looking for your keys, and you glance at a fruitfuit bowl. Many people could not tell you what fruit was in the bowl, and how many. All they saw was a bowl of fruit. How is that possible? Surely the light from five bananas had entered their eyes.
The brain simplifies the world in a million ways for faster processing. You probably can't even draw the back of your hand well, and don't know how many hairs are growing from today. Because that's not relevant to what you're trying to do. For an esport, a place where things are happening very fast and unfamiliar, only providing necessary detail is essential. Necessary detail includes things like really cool looking characters, and the beautiful map for them to fight in. What you don't want to do is make a bunch of cracks in the mud that the characters walk over, to the point where it's hard to see the characters from the cracks.
A place I think that MOBAs can take a step visually is through lighting effects, but that has to be very carefully balance with the needs of not giving your audience seizures.
For more information on stylization versus realism, watch:
http://www.penny-arcade.com/patv/episode/graphics-vs.-aesthetics
It’s really quite enjoyable and informative.