The tenth annual Game Developers Conference
is in full swing in San Francisco, CA -- and yesterday included a panel by Rob Pardo, Executive Vice President of Game Design at Blizzard Entertainment. Pardo spoke about design philosophy and how Blizzard approaches it, sharing not only Blizzard's success stories, but where they failed along the way, and what they did to fix it. Blizzard's design philosophy follows some key elements:
: Before anything else, you want to concentrate the game on the fun. All aspects of the game -- the design, the mechanics of encounters, the quests and story are focused on making the game fun to play. Not only fun to play -- but fun to play for players, not developers. The challenge is to keep players jumping through the correct hoops, while making those hoops fun. Sometimes this involves making some changes -- for example, only night elf males could be druids in Warcraft III
, but for the sake of making the druid class, something that sounded like all kinds of fun, they had to be made accessible to both genders, and both sides. So the lore was adjusted so that females and tauren could both be druids -- otherwise they couldn't have introduced the class at all. And that wouldn't be any fun.
Easy to Learn, Difficult to Master
: The concept here is to keep game play simple in terms of mechanics and objectives, but design the game in a way that the challenges scale with the ability of play. Pardo stated that Blizzard is focusing more on designing for multiplayer games first now, rather than single player, so they design for the multiplayer aspects, giving games a lot of depth so that players won't get bored with it. He also said that WoW
is a pretty hardcore game, but the key is that it's accessible to a lot more gamers -- endgame content like raiding and arenas are a lot more hardcore than leveling itself.
He cited the Diablo II death penalty as a failure in this aspect -- the death penalty was simply 'you die, you lose half your gold', but this was easily circumvented by dropping your gold off in town and not accessing it unless you wanted to buy something, which inflated the economy to the point that gold was meaningless, leading to players bartering and trading items rather than just using gold. They took those failures into account with WoW, not only designing the death penalty as a 'tax' of sorts where you'd have to pay to repair your gear, but by introducing money sinks that would make you want to spend your hard-earned gold like fancy mounts. The auction house was developed so that the player economy would revolve around gold, rather than simple bartering.
What is the Fantasy?
In other words, what should the game look like -- Pardo talked a little bit about the UI system, and how they intended for it to be simple to use and intuitive. He said the UI system was something that he considered a failure -- not because it was bad necessarily. But from a development standpoint, if the majority of your player base is using addons to modify the existing UI, that's a clue that something wasn't quite right with the way the UI was originally designed.
Make Everything Overpowered:
Every unit, every class should feel unstoppable, overpowered and epic -- because it's just more fun that way. Pardo told a short story about Designer Island, an area that used to exist in game for designers to play with landscapes and NPCs. He said he was given two abilities on his bar that you don't see in game – a grow button, and a shrink button that would either grow or shrink the target by 10%. After bringing in Nefarian, he said he must have hit the grow button about twenty times before he finally stopped and said "That's the size we want him." Afterward he noted "I don't even know why they gave me the shrink spell, I've never used it to this day!"
Less is more when 'less' is concentrated into one simple, overpowered and fun class to play. Rather than having 27 different classes in WoW
, they took the best elements from units in Warcraft III
(Thunderclap from the Mountain King, Critical Strikes from the Blademaster, Shockwave from the Tauren Chieftain) and combined those into one 'super-concentrated cool class' with many fun abilities -- the warrior. Other classes were approached with this 'concentrating' concept in mind.
The interesting part was that he cited vehicles in Wrath
as a failure -- he said it was a fantastic concept originally designed for Wintergrasp only, but the concept was so cool that the designers went overboard with it. Soon they had quests, zones, and even instances that revolved around the concept of vehicular combat, and it caused the vehicle system to lose the wow factor that made it so unique and entertaining in the first place.
Play, Don't Tell:
Players should be playing as much of the story as possible, and text, voiceovers and movies should be used to enhance the story as it moves along. He cited the death knight starting area as a success, the introduction of phasing allowing the zone itself to change around the player as the story was played out, with players having a very 'real' impact on the world around them. What was interesting was that he spoke about the tendency of players to simply skip over or skim quest text. He asks quest designers, "If you make a quest, and players don't read any of the quest text, would they have a basic understanding of the storyline?" and tells them to keep that in mind, adding quest text after the quest has been developed. Quest text shouldn't be necessary to understand the story -- it should be there to enhance the story that's already obviously playing out.
Make it a Bonus:
Players respond better to incentives than to punishment. That's a no-brainer, but Pardo had a couple of funny stories from the WoW
beta to back it up. He said that originally the Rest System in WoW
worked like this: You started out gaining 100% xp, but the longer you played, the more that percentage dropped, eventually falling to 50%. This was to discourage players from playing more than a few hours at a time. Beta players hated
this system -- so Pardo changed it by doubling the amount of xp required to reach maximum level in the game, starting players out with 200% xp gained, and slowly dropping it to 100% xp as they played. Same effect, same numbers, the only difference was the way the numbers were presented -- and people applauded the 'change'.
He also said you don't want to fight player psychology. In the original beta, when a player was inspecting you, you'd receive a notice about it. The thought of being inspected creeped players out, and they said they didn't want people to do that. So rather than remove the inspect system, they simply removed the message, and everyone was happy.
Control is King: Controls should be as responsive as possible. While players have clamored for different animations and effects, Pardo gave some very specific examples of why they simply wouldn't work. As it stands, when you summon a mount it simply appears beneath you in a puff of smoke -- the animation department suggested that it would be really cool if you'd actually call your mount and have it run to you so you could hop on it, going so far as to mock up the animation for it. But there was a downside to this -- it took several seconds for that animation to play out, and if say, a rogue jumped out to stun lock you, you probably didn't want to be stuck stunned and rapidly dying while watching your horse gallop up to meet you. So they settled with the puff of smoke we're all familiar with.
Tuning It Up:
Pardo noted that tuning is easy to do, but hard to do well -- that you have to keep in mind who you're tuning your game for. With World of Warcraft
, they succeeded in matching the level curve to the level of content, making it so that every player can solo all the way to max level if they want to, adding enough quests that it didn't feel necessary to grind along the way. Pardo also noted that there was a myth about reaching max level -- that players would simply quit the game once they reached the level cap. Blizzard took the stance of 'if the game is enough fun for someone to get to level 60, they'll want to play the game again' -- a stance that seems to be working remarkably well for them so far.
Avoid the Grand Reveal:
Blizzard encourages designers and developers to show their work to everyone else often and avoid the 'I can't show this to anyone until it's perfect' mentality, creating an environment where feedback is encouraged and it feels 'safe to fail.' He cited Silvermoon City as a failure in this aspect -- the city itself was designed in parts, and so difficult to weld together in game that they didn't really do it more than once or twice while developing it. This resulted in a really beautiful city that didn't feel like it was very well put together. Arathi Basin was noted as a success -- the original map was very simple, and the designers simply built up from there, resulting in a battleground map that was playable from day one.
Culture of Polish:
With Blizzard, polish isn't something that happens at the end, it's something that happens all the way through development. Pardo mentioned that it's an atmosphere of a team that is making their favorite game even better -- that people love the game they are working on. Once the game gets far enough along, 'strike teams' are brought in -- a cross section of developers from other teams, artists, programmer, designers, both new and experienced players, and they play test for feedback and a fresh perspective.
He pointed out that every voice matters -- and that when dealing with player feedback, it's a matter of keeping in mind that the more passionate your player base is about your game, the more you're doing something right. He said if players are complaining about the game, they try to look at it from the perspective of 'this person is just trying to make the game better.'
Pardo followed up with a short note -- don't ship your game until it's ready. Self-explanatory, but refreshing to see that Blizzard is the sort of company that would rather a game be complete than push it out the door half finished.
With over 11.5 million players in WoW
alone, it's clear Blizzard is doing something
right -- and the panel did an excellent job of shedding a little light on what that something is. Game developers take note -- this is the way you want to start. Check out the rest of the slides from the panel in the gallery below.