Game Dev-astation

Outdating old content, and opposing viewpoints.

2009-06-21T06:49:04Z

Humans have an inherent need to justify spending resources on things. People want to know that the time, money, emotion, empathy, or any other personal resource they spent on something is not wasted, even when it is. A good example of this is our time and money investment in leisure activities. There are thousands of comics and baseball cards in circulation, but only a handful of them are worth more than the paper they're printed on. Only a handful of people care who has the world's highest score at pac-man or pinball. Some things can end up being somewhat worthwhile, such as being a professional Magic: The Gathering player, or a professional Starcraft player. These are, of course, few and far between. The ratio of grand master chess players to regular chess player is extremely small, just like not everyone can be Kobe Bryant or Tiger Woods.

What does this have to do with games? Well, in games that save your state, it starts becoming important. The entire genre of MMORPG is all based on ascribing value to something where there really inherently isn't any. I've never seen another specific genre of game take this to such an extreme, and I wanted to spend a bit of time analyzing it.

We've all heard the old adage "When I was your age..." from older folks. We may have been on the giving end as well. Most of the time, it's older folks telling younger folks how much harder they had it while they were experiencing it. This is because the older folks ascribed a certain value to it, and do not perceive their descendants and successors to be valuing it as much. It is much the same idea that a bottled soft drink used to cost a nickel, and now costs $1.00 at a vending machine. The value of a bottled soft drink really hasn't changed, but the perception is that it has. In reality, it's the value of the money that's changed more than anything else.

You might ask what brought this on. It actually spun from a discussion I was having earlier with a guildmate in world of warcraft. One of the latest things that Blizzard has announced with their latest WoW patch is that they will be providing players with the opportunity to get higher level gear with less time or effort investment. Specifically, it will allow them to do lower-difficulty content and skip the more difficult (yet still old and somewhat obsolete) content. There are two opposing viewpoints, and it inherently is about the value they ascribe to their time spent in the game.

The first one is from those who don't ascribe quite as much value in the time. This could be because they haven't actually spent that much time, so it isn't a big deal to them. It could be because they don't value the time spent quite so much (possibly because they know it's wasted anyway). They see it as a good thing. There are several benefits for the change, and these folks are quick to point them out:

More people get geared faster, meaning the high end raid guilds have more candidates to recruit from

People who don't have the option to join raid guilds have new goals to achieve

People who were not able to defeat old content due to gear constraints will now be able to see that content

The second group dislikes the change. They are the ones who place higher value the time they spent playing the game. They dislike the idea of somebody following them not paying the exact same dues they did to get to that position. They had to walk ten miles through the snow to get to school, uphill, both ways. It was hard for them, and they feel that their play experience is cheapened by their successors not having to step through the same fields and traps they did. They see some drawbacks to this change:

It makes a lot of old content much less useful. A lot of time and effort went into making that old content; making it totally optional devalues it

It devalues the accomplishments of the players who spent a lot of time and effort getting to that point. Earning a particularly powerful reward is a joy to a player. It doesn't feel good to see that same reward being earned with less effort.

My perspective is more toward the first. Unless you are striving with a very specific goal in mind (like going pro, and receiving that sort of recognition), the time you spend in is largely "wasted". I use quotes because it's the goal that matters. People play games for a variety of reasons. Those that play to have the best loot are likely to be more disappointed in this than others. Those who play to enjoy the game as they go will likely not be as affected. I wish more people would play for the second reason than the first; most games are about enjoying the journey more than the reward at the end.

--Rawr

Quickly: Brief history of Frost Presence (Death Knight)

2009-06-02T15:25:47Z

During Beta:

Frost Presence is +60% armor, -5% magic damage taken.

WoW 3.0.8:

Frost Presence is +80% armor, -15% magic damage taken

WoW 3.1.0:

Frost Presence is +80% armor, -5% overall damage taken

WoW 3.1.3:

Frost Presence is +60% armor, -5% overall damage taken

Conclusion:

The change from -5% magic damage to overall damage was sufficient from beta to now. The major reason for the original changes came from a few other things:

- Launch DK needed a lot more defense to be uncrittable. In the intervening time, DK gained Rune of the Stoneskin Gargoyle and Sigil of the Unfaltering Knight, and no longer need massive defense socketing. This allows them to skew itemization towards massive stamina pools, which get multiplied significantly.

- Launch DK needed either the 20% bonus armor or the 5% damage mitigation. In addition, Blade Barrier changed from 10% parry to 5% all damage reduced. This combines to effectively form 'defensive stance' for DKs (like the other tanks have), but they also had the extra bonus armor that provides more armor than any shield.

--Rawr

PvE Raid Content Design: Cooperative Content Deconstructed

2009-03-12T01:12:42Z

Ok, so a couple of things I've been thinking about lately have been cropping up and I wonder about them specifically. WoW v3.1 is coming around the corner soon, and that means a number of things: #1, a new raid dungeon. #2, it means that there's bound to be a tidal wave of complaints as new content arrives. Whenever people think about the raid game, the five most (potentially) dangerous words in game design are often uttered by an armchair designer: "Wouldn't it be cool if..." And it makes sense, since at the outset everyone has lots of ideas and many of them would indeed be cool. They think about them, and some people even flesh the ideas out on paper, or the internet, or whatever. And that's cool, it's fun for somebody to do. However, others seem to take that and think "I could be the next designer and get paid to do this stuff!". While this is true, it's actually less likely than you think... mostly because the difference between somebody who's been designing encounters for years knows a lot of details to think about that the regular armchair designer would never have thought of. Let's investigate what I mean.

At the core, PvE raid design is about cooperative gameplay. You have a bunch of people working together toward a common goal which is to beat the computer in some sort of scripted encounter. The encounter will likely follow a pattern, and the players must follow the appropriate dance steps to succeed. Once they succeed, they move on to the next encounter, typically with a few shiny new bits of loot or some other reward - more hit points, a new ability, etc.

Scripted Boss encounters in single player games are often used to teach a player the value and usage of a new skill they just learned in the game. There is the period of experimentation, where you try to figure out what the boss does, and then try to counter it. Perhaps it is a pattern the boss does, or a set of moves that must be avoided and the boss will randomly choose one to do. Take this video, for example.

In most Legend of Zelda games, Link uses the tool he obtained in just that dungeon to defeat the boss in some sort of interesting and spectacular fashion. It isn't tremendously difficult, but it is fun and impressive, and teaches the player things.

Now... what's this got to do with cooperative content? When designing cooperative content, you need to know what skills are available to players, and give them challenges using those abilities. Typically, players will have roles in a cooperative environment - it's a natural extension of society. You have people who specialize in order to create a more versatile group. You can see this in games like Team Fortress 2, where the game is cooperative and competitive at the same time - two teams eternally vying for a goal. Players are individually good at certain things. They specialize and want to continue to do those things to help the group along. Maybe that is spearheading an assault, maybe it's defensively protecting the flag, maybe it's keeping the spearhead alive.

Group cooperative content is built on a couple of basic principles: Individuals do what they are good at in the game, and their collective effort is rewarded. However, any aspiring designer should remember the all-too-important caveats to this goal. Let us take your extremely basic raid encounter. You have three basic roles:

Tank: The one who gets the boss to hit him and not other people. Built specifically to survive large amounts of punishment.
Healer: The one who keeps the tank alive. Built to keep the tank alive as long as possible.
Damage Dealer: The one who kills the boss. Built to deal as much damage as possible for as long as possible.

After X seconds of combat, provided the tank is able to avoid or mitigate enough damage, the healer is able to keep the tank alive, and the damage dealer is able to deal enough damage, the enemy will die and the team will be rewarded. This is the basic foundation of MMOG group play. However, people are finicky and do not enjoy doing the same things over and over in exactly the same way. So if you are in a dungeon full of encounters like this, variation on things to do is a good thing. Thus, we have many different types of raid designs. Most of these are kept in the same general principle - the tank keeps from dying and holds the attention of the monster, the healers keep the tank and others alive, and the rest of the people are damage dealers that hurt the monster until it dies or is otherwise defeated.

Throughout most raid encounters, things happen and players find themselves with other things to do besides their given role. This helps mix things up and keep them from becoming bored. Sometimes these tasks are simple, like "Don't stand in the large glowing circle on the floor". Sometimes these tasks are more difficult, like taking control of a dragon, or taking control of a ghost. This is where the first major pitfall of raid design commonly occurs:

If you choose to remove the player from play for any reason other than "the player made a mistake and must pay the consequences", you need to make it so that they have something else to do during that time. Doing his or her job should be the basic level of interaction a player has for the game. If you do not allow this, you've got a player who has been forcibly removed from playing the game, and it just plain isn't fun. If it's a mistake the player made, then not being able to continue playing for the rest of the encounter is a punishment, and is perceived as one. Accidentally stepping into the fire and getting burned results in death, now the player knows not to do that. But if the boss randomly selects somebody and says "Ok, you're dead now", the player can't help that and it is extremely frustrating. This also makes raid encounters less scalable, since removing a player from play in a smaller group hits much harder than removing one from a larger group. Players play the game to have fun; making inescapable death (or removal from play) a random occurrence is not an enjoyable experience.

This isn't to say you cannot do it. Game design is, after all, a complete abstraction that has no real boundaries aside from "what is fun" and "what isn't fun". As a designer, you want to minimize the time a player doesn't play your game while still sitting in front of the screen as much as possible. There are a number of WoW encounters that have death touches. All of them are fairly short fights, and only one of them does not provide the player who is sentenced to death something special to do (though does give 20 seconds to continue playing). Furthermore, each of these encounters are typically very short.

It's mostly not a question of necessarily killing the player. It could just be removing the player from the game for an extended period of time. Extended could mean just about anything, but the typical duration is from 10-20 seconds at most; longer than that and players' attention will tend to wander and boredom will ensue. It isn't fun to not play the game; this is very similar reasoning to why things like Polymorph and Roots don't last any longer than 10 seconds in pvp situations. Total encounter length should also be adjusted to fit this; if the players have a high chance of death, the encounter should be fairly quick (a number of minutes) so that the dead players are not stuck waiting for too long. If the game has battles that last past the 15 minute mark, then there must be some way to return players from inactive play to active play.

So the first one in a nutshell is "Let the player play, and try to avoid having them removed from play for reasons beyond their control." Let's move on to the next one.

Players like to understand things. They don't like to have to guess. If they lose, they want to know why. This is the whole part of learning that makes things fun. They want to have the ability to look back and see why they lost, in order to correct it the next time they try. This means that whenever a monster does something, it should have a lead up and a follow-through. Similar to animation, monsters need to telegraph their attacks to the player. This way, even if the player doesn't avoid it, he or she will see it coming and realize "Damn, I made a mistake and need to not do that next time." Sometimes, due to technical or scheduling limitations, designers must use things that aren't animation-driven, such as audio clues. In a MMOG, it is likely to see things like 'the monster shouts "I AM GOING TO USE ATTACK #5 NOW!"'. This clues the players in and teaches them what happens. Having to go back and try to piece together what happened is something best left for CSI, or whatever television murder mystery show is currently on. It frustrates, confuses and annoys players when monsters do something without giving players adequate warning.

And finally, the last one. Complexity does not necessarily equate to fun. It almost never does. Really super technical encounters tend to only be interesting to a small subset of players that enjoy really convoluted problems. Keep in mind that the goal is to entertain players more than it is to challenge them. Multiplayer games tend to not have an immediately available difficulty slider, so Blizzard's concept of "hard mode" and scalable difficulty is extremely valuable for this sort of situation. However, one must keep in mind how difficult things can get. Players at the bare minimum, should be doing what they normally do. This is the basic level of gameplay. If you are pushing them to meet some sort of minimum timer, you can expect them to spend more concentration on perfecting their skills than on paying attention to what is around them. If the encounter is more based staying alive, then you can expect players to pay more attention to that. You should never force a player to have to expend too much concentration on more than one or two things at once; that causes confusion and makes it much more difficult for players to succeed. This is in part why the Sartharion encounter with 3 drakes is so difficult. It isn't because the gear requirements are so high, but it requires concentration on multiple levels from many players in the group.

Designers should try to adhere to the old >KISS principle. Keeping things simple is very good. Simplicity is good; complexity for its own sake causes a lot of confusion. If an encounter cannot be explained to a playtester easily, chances are it is too complex.

Accessibility Part 3: Sometimes it can't be learned

2009-03-10T07:21:23Z

So very recently, I picked up and started playing Street Fighter 4 like about 2 million others. If you know me, you probably know that I follow the game fairly well. I really like it, because it has a bunch of stuff that I think is really good for games: A great intuitive control scheme, low barrier to entry, and a good way for skillful players to differentiate themselves in ways that aren't (usually) cheating. This got me thinking about games, and when I think about games I tend to come up with something to blog about. This is no different. When last we talked about accesibility, it was mostly about designing content. This time it's a bit different. The discussion is still about things like how much of a game can be played by people, but content isn't quite the right word to describe it. I'll explain after the jump.

When you look at games, you must realize that there are certain limitations that may be beyond our control. For example, somebody who is 5'1" tall has a disadvantage in basketball against somebody who is 7'2" tall. The 5'1" tall person cannot magically change height to make things more fair. Somebody who weighs 300 pounds cannot help outweighing somebody who weighs 150 pounds. These are physical traits that, for the most part, cannot really be overcome without severe physical training (if at all... can't really do much about height or reach). In most games, most physical characteristics aren't required anymore... are they? Or aren't they?

Sometimes there are. At the top level, there are professional players like SlayerS_BoXeR of Starcraft fame. These players can perform many hundred actions per minute, measured by movement, mouse clicks, etc. Most people can physically perform that many actions per minute, but the difficulty comes from controlling those actions. It is much like typing - you can achieve the equivalent of over 100 words per minute by mashing the keyboard as quickly as you can. However, the difference between mashing the keyboard and being able to accurate convey meaning is very different. Similarly for Starcraft players, the actions per minute are all part of a sequence of weighted decisions the players make as they play. There are other similar limitations. Some players just have faster reflexes. Some are able to intrinsically time things better than others.

So the question becomes... what does a game designer do to deal with these limitations? Are they acceptable? The initial response would be "No, you shouldn't have such high limitations involved", but that isn't quite accurate. It then depends on the arena of acceptability. Here's a perfectly reasonable example of a game genre where physical limitations are perfectly justified in allowing players to lose:

As the game progresses, it becomes harder and harder to whack the moles. The revenue model is based on pay-per-play, and you want the player to enjoy the game but not for too long. This structure is duplicated in most arcade games. Racing games require you to race better and better, fighting game opponents become successively difficult, and even a game like Tetris eventually raises the requirement so that it's less reasonable to do so.

The difference is that you typically don't want this kind of game that you just purchase. If I purchase a game, I want to feel like I can play the entire thing. Games are expensive, players don't want to feel like they have wasted $50 or $60 because their fingers can't press the button or wave the wiimote fast enough. Players who have a vested interest of $0.25-$0.50 don't care if the game gets progressively harder to the point it is unplayable. It's only a quarter or maybe 50 cents. Players who have a vested interest of $50 or more, in addition to the console or PC to play the game on put the developers on the hook much more. If the game doesn't deliver on a good level, players will be less likely to purchase other games from the developer.

There's typically two levels of gameplay in any game. The first level is getting the character on the screen to do what you want him/her/it to do when you want him/her/it to do it. That part's the physical skill part of it. This typically involves "Push this button at this time". The second part is actually making the more interesting choices of "What should I do in this situation?". Most of the time, there's a learning curve associated with both the former and the latter. Simplistic control schemes provide short learning curves at the start, and simplistic gameplay decisions provide short learning curves for the latter.

So what does this have to do with Street Fighter?

Street Fighter actually has a pretty difficult physical learning curve. It's one of the few games that has two parts to the learning curve, and that's what makes it weird - the second part is actually much tougher to pick up than the first. Street Fighter has levels of complexity in terms of controls. I've attempted to detail them for you here:

Moving the joystick

Attacking by pushing the buttons

Performing "special" moves that are simple joystick motions, followed by button

Performing more elaborate joystick motions followed by button(s)

Pressing buttons in a set sequence

Performing button sequences that require more elaborate timing see: Link

Doing all of the above in a time-intensive situation

Now, while I could probably write a good number of pages to explain the combo theory behind Street Fighter to you, the important thing to realize is how far down (numerically) most players get. The first two are very easy to pick up. The #3 takes a little bit longer, and #4 shows up in certain games under certain circumstances. The #5 shows up in different types of games, and #6 and #7 are the hardest to master. Most of the time, when a player picks up a Street Fighter game, he or she can do #1-3 easily. #4 through #7 are pretty much beyond them unless they are experienced in how the game plays. Part of Street Fighter's legacy is how old and established it is - most people have tried it at some point and understand the basic joystick motions for #3. Furthermore, the series essentially created the fighting game genre, and so you see many similar control schemes have been adapted for other fighting games (king of fighters, Samurai Showdown, World Heroes, etc.). The biggest problem, though, is that the jump from #3 to #7 is so large... and it changes the game rather significantly, because of how much more the game opens up once a player goes further down the list numerically.

When the game becomes more about out-thinking the opponent rather than trying to figure out how to throw a dragon punch, it becomes a lot more interesting and strategic. People don't enjoy chess because they figure out how the knights, rooks and bishops move. The point of the game isn't to move the pieces, but to move the pieces intelligently. In a fighting game, it's important to assess risks. Somebody who will hurt you for 15% of your life bar if you make a mistake is a lot less scary than somebody who can hurt you for 50%. It provides you as a player a much larger comfort zone.

In a single-player game, performing the moves should be fun and interesting too. You should feel rewarded for being able to do more difficult things, and they should be more spectacular. There it is probably ok to have more difficult moves and such as long as they are mostly not required to experience the majority of the game's content. I didn't need to unlock every single combo and upgrade every weapon in God of War to finish the game. The option was there, and if I really wanted to I could do so. However, in a competitive game, it should feel like the choices you make directly lead to the strategy.

If you take away anything from this, take this: If you want to make it a challenge to make the game do what you want, that's single-player territory. If you want to make the challenge pitting one player against another, you want to make the game do what the players want very easily. This provides the needed accessibility to the game that lets the players get to the fun part - the interesting decisions.

Lights! Camera! Bad Game!

2009-03-04T22:51:55Z

So... in today's installment, I am revisiting the old "Why do bad games get made?" riff, and taking aim at a very specific subset of games that tend to be bad: Movie tie-ins. Now... on the outside, it would look extremely promising to have a good game that ties in with a movie. After all, if the movie is good, you've got a huge customer base that's probably hungry for more of what the movie was about, and a game is a perfect way to get them to spend their money to get more. It's got built in marketing, it's got a lot of buzz, and people will buy them. It has a solid story to build on,

The answer, of course, actually lies in the age-old adage "You get out what you put in." Time is the enemy of movie titles, and that's specifically the issue with movie games. Specifically, think about how long it takes to put out a quality game. Of any of the games that are accepted as "good", they take a significant amount of time and resources to develop. A game like GTA4 had a team of over 1000 people and took over 2 years to develop. Call of Duty 4 took two years, both of the God of War games took 2-3 years, many games take a long time to develop. Movies, on the other hand, are actually much faster to develop. They don't need as much testing, because very little has to change. They go through scriptwriting, casting, pre-production. Then they film for a few months, then post-production for a month or two, and then they're done and wait for the right timing to release the film. It is not uncommon for an entire film to go from green light to done in under 8 months total.

The difficulty comes because you have schedules that are mismatched. An important thing to note about movies is that many of them have their release date set even before they are completed. Since they know what day they have to be in the theaters, they can effectively 'work backwards' and know how much time they need (typically, 1 year). This means that the game developers effectively have to build things like gameplay for a year without knowing things like:

The story

The characters

The locations

That makes things extremely difficult. Furthermore, it compresses trying to make the story, chracters and location fit the gameplay they have built into the year the film is in production. Even worse, these are the very things the moviegoers are expecting to see - experiencing the spectacular moments of the film through the game, by playing as the main character. Finally, it needs that polish phase that doesn't usually make it because the film is coming out on that day exactly, and so you absolutely must be on the shelves before then. The polish phase is critical for many games, and often has to be compressed in order to make the ship date.

These factors tend to cause the downfall of most movie games. The best most game developers get from the movie studios is vague descriptions of the scenes and some ideas. Since the studios aren't actually filming yet, that's the best they can even do. Trying to work around these constraints is very difficult; it's why so many movie games tend to be bad. If you don't take them into account the game will have little to do with the movie, but if you do take them into account, you're essentially shoehorned into an extremely short development time. I can only think of two major movie tie-in games that were decent.

The best movie-to-game adaptation that most people remember is probably Goldeneye for the N64. However, if you think back about it, what is it that made it a solid and fun game? It wasn't really the fact that it was a movie tie-in at all. If you removed the 007 license and renamed everything, it would still be a solid and fun game. It just so happened that the Bond license was attached to the game. The multiplayer specifically made Goldeneye fun. Not many people remember the single player, but the multiplayer levels were built well, the weapons were fun to use and it generally handled like a competent FPS on the N64.

The other movie game tie-in I can think of was actually Chronicles of Riddick: Escape from Butcher Bay. The funny thing is that this game received generally strong reviews while the film it tied in with was a stinker. The guys who built Escape from Butcher Bay did something very clever. They divorced themselves from the film completely, and made a prequel game instead. This allowed them to come up with their own story, and as long as the pieces of the end of the game tied in nicely with the beginnings of the film, all was well. This allowed them a free hand to design their own gameplay without most of the constraints of the film.

Something old, something new

2009-03-03T22:41:00Z

Moving along in the FAQ series, one of the questions I get pretty regularly are variants of this:

"Why do publishers keep making sequels?"

Alongside this:

"Why don't they come up with new games, instead of keep rehashing stuff?"

Also, recently this has come up:

"Will publishers keep taking risks on triple-a games like Mirror's Edge after seeing them flop?"

These are all facets of the same issue that crops up from time to time, and it is quite true. Part of this stems from what makes a game a game, and part of it comes from customer confidence. Let's discuss this further, shall we?

The first thing to understand is what makes a game a game. A game is a collection of systems put together in such a way that lets a player do some basic tasks over and over in a simulated environment with basic rules. Doing these tasks generally makes it fun for the player. You can break any game down into component parts this way, be it chess (moving pieces on a board), twister (placing limbs on randomly determined colored spots) or counter-strike (moving, aiming and shooting in a 3D world). There are other elements to games, such as story and pacing and such, but when you think about it, what makes it a game is the actual systems that interact with each other that the player controls. This is what separates a game from another forms of entertainment media: the ability to control things.

This is important in looking at why sequels work the way they do, and why game sequels are more prevalent than things like movie sequels or book sequels. When movie fans want a movie sequel, it is because they want to know what happens next in the story (or watch more spectacular action scenes, or whatever). When book fans want a book sequel, they want to know what happens next in the story. But when game fans want a game sequel, most of the time it isn't because they want to know what's next in the story - it's because they want to keep doing the things they were doing before, but in newer and more spectacular manner than before.

For example, people wanted a sequel to Katamari Damacy because they wanted to keep rolling stuff up into balls, and they were tired of doing it in the same levels as before. They wanted more ball rolling, more stuff to roll into balls, and they wanted more ways to roll balls up. They wanted to try rolling balls up with other players. They wanted cooperative rolling. They wanted ways to share the fun of rolling stuff up with friends. So Namco made sequels.

I am sure you can think of many game series that are similar to this. There are whole game dynasties that are focused on perpetuating the 'feel' of the original game in the series through the past quarter century, like Mario. From Super Mario Bros to Super Mario Galaxy, the same basic concepts are there - jump, run and explore. They've added new features as they went, but the core is still there. Similarly, look at Megaman. Jump, shoot, defeat bosses and gain new powers. Most game series are based upon a core collection of game systems, with some innovation in terms of new ways to do stuff that is somewhat different but not completely from the previous game. Very little of the time does the story actually continue; some game series maintain the core systems but reset the story each time (e.g. Final Fantasy). Players want sequels because they want to keep playing the game past the content that they were given on the disc.

So now that we've identified the core of what makes a game a game, and what makes a game series a game series, it is important to identify why publishers continue to make sequels. Publishers like Activision-Blizzard and Electronic Arts are in the business of making money. Intellectual Property isn't just what they are guarding, but it's the core game systems associated with that intellectual property. When you think about Tomb Raider, you think about Lara Croft... but you also think about doing acrobatic jumps and navigating death traps in ancient tombs. You would be more skeptical about a Tomb Raider kart racing game, or a Tomb Raider fighting game. The strength of the brand is rooted in the gameplay developed for that brand, and the trust of players that the gameplay continue in a similar vein as before. Thus, by creating more content with similar gameplay and a brand famous for it, they will earn the money of the players who enjoy that type of game.

Sequels will, oftentimes, eventually wear out their welcome. A lot of game series become "boring" because they don't do a good enough job of expanding the gameplay each time. However, a brand doesn't necessarily die because of this; it just gets set aside for a time in order to capitalize on factors like nostalgia. A perfect example would be the Turok game series. For a while, Acclaim put out several Turok titles from 1997 to 2002, and they weren't sufficiently evolved to be fun or different and players got sick of them. The Turok games went on hiatus for a few years, then eventually came back in 2008. If the development team can keep the gameplay systems interesting and fresh, they can continue a series nearly indefinitely and it will continue to make money (e.g. Zelda, Final Fantasy, Castlevania, etc.).

Games like Mirror's Edge and the like are attempts at creating new franchises. Every game has the potential to be an ongoing franchise, because you can simply reset the entire story and still make a sequel. As long as the gameplay systems are the same, you can continue to move forward. However, the trick is to come up with gameplay systems that entertain and draw enough support from players to buy the game. It's harder to come up with an entirely new game with enough gameplay to make it fun. Thus, you'll never see a publisher stop putting out new IP. It is within their best interests to continue to try to find new brands they can continue to build sequels for.

When it comes down to it, new games are a risk. A publisher has to risk millions of dollars on development for a new kind of game in order to create a new brand, and further millions to market it properly. The rewards for doing so are great - if you generate a good game with good publicity, chances are very good that the good will from players generated from that first game will spill over to the first sequel, the second sequel, and so on. Look at Devil May Cry, for example. The first game was such a treat that the second game (which was generally panned by most players and critics) still sold decently well, and the third and fourth games were also generally well received. Sequels are generally not as risky - they have a built-in fanbase that will be more likely to buy the game. When it comes down to it, the risk is the most important factor, and that directly ties to things like the economic situation. If EA is riding high and earnings are up, they are much more likely to take risks and bankroll more high profile new IP games like Mirror's Edge. Most games take around 2-3 years to complete. If you look at EA's financial data in the 2005-2006 range (when Mirror's Edge started development), they were doing pretty well. Their stock price ranged from 50-60 a share and they were doing pretty decently. Their share price stayed above 45 points until late august of 2008, when it tumbled. Now it's under $20 a share.

Generally, sequels are safer bets that are used to fund new game experiments. The new game experiments are built with the concept of creating new safe bets in mind. In good economic times, you'll expect to see more new games and riskier games. In bad ones, the publishers are more likely to stick to what earns money. This is why you'll likely never see new games entirely vanish, but you will see them sometimes start to dwindle in the face of a bad economy.

Fighting upstream

2009-02-16T23:47:15Z

So, as stated before, we've got iconic imagery, and iconic motion. When you combine the two, you have something that becomes instantly recognizable to players viewing it as "cool". Here's a good example of that:

Bayonetta is an upcoming 3rd person action game from Sega. Despite being a totally new IP and without much knowledge of how the game plays, just by watching the short video you get a good idea of who the main character is, and what her modus operandi is. She's essentially a sexy warrior who fires guns in every direction, reveling in her own strength and ability to kick ass. Her movements are lithe and graceful, she moves liberally in all 3 dimensions, and her combination of skin tight clothing and sexualized demeanor all contribute to this. Nobody would ever confuse her for a blushing bride or somebody who just wants to be normal. The confidence with which she speaks and moves is apparent, even in such a short amount of time.

But what happens when this sort of thing works against you?

The best example I can think of is probably a game that still managed to earn an average 96 metacritic score. Yes, that's right.

The main character of Metal Gear Solid was Solid Snake. He's a character that's been established as a badass. Visually, he looks tough, old and experienced. He has well-defined musculature, smokes cigarettes often, and his look is also something of a perpetual scowl. He's even got perpetual stubble.

In addition there are many other factors that add to this. His voice is low and gravelly. He carries himself extremely confidently. He is a highly decorated special agent who is famous for being inserted into dangerous situations with no gear except for his cigarettes, and still getting the job done. In essence, a macho example of a manly man's man.

In the first third or so of Metal Gear Solid 2: Sons of the Patriots (the Tanker incident) the player controls Solid Snake. Then the rug gets pulled out for part two (Big Shell incident).

The majority of the game is played as the new character Raiden. Raiden's character design is extremely different. He has long, flowing hair and a higher-pitched voice. Within the first few minutes of his character, his commanding officer has established that this fellow is new, fresh out of VR training and hasn't had any actual field experience. His voice is higher than Solid Snake's, and even some of the moves he can perform, while analogous to Snake's, look more delicate.

Much of the story is dedicated to Raiden discovering his own past and where he came from. Raiden had a similar mysterious past to Solid Snake. Raiden had the exact same bunch of abilities that Snake did (with slight cosmetic differences such as Snake diving and rolling, while Raiden would perform a vertical butterfly kick). But even without the story, just the way that the character is treated automatically sorts Solid Snake and Raiden into two separate categories of character, and that's a dissonant value that, overall, hurt the enjoyment of the players of the game. They expected Solid Snake, and got a cheap knockoff instead. And that made them unhappy. Sure, there were plenty of other complaints about the game, but the one that turned the majority of players off was that they were given a character that was not the same kind of iconic character that Solid Snake was, and therefore rebelled.

The issue isn't that Raiden is a bad character. Anyone who has a troubled past, but overcomes it in the face of adversity while becoming a hero really isn't all that bad. His story really isn't terrible, despite some of the issues with how it is handled in certain sections. If they had cut the rest of the game and called it the Metal Gear Solid Sidestory, it probably would have been treated pretty well (minus the whole WTF? ending). The problem is that they did this in the canonical Metal Gear Solid game, which had been preceded by several other metal gear games all of which were all about Solid Snake. By changing character icons so quickly, it jars the player from that comfortable place of make-believe back into holding the controller in front of a TV. Even worse, it actually antagonizes the player. It was especially bad with MGS2, because there was no warning to the player at all that he or she wouldn't be playing as Snake.

When you go to a restaurant and you order a medium rare steak, you'd be angry if you got a filet of sole, even if the sole was cooked very well. You were in the mood for steak and ordered steak... yet you got something else completely. It is extremely important to keep this in mind - the player is your customer and the game you are creating is the meal. Making the meal to the customer's order is as important as making the meal taste good.

Economy of Motion

2009-02-14T23:04:58Z

So, in the second of a series (probably of 3), we're discussing iconography. Specifically, what it means to make something recognizable. Last time, we went over how a picture is worth a thousand words. That isn't sufficient though; a picture is a picture. A large part of this is because of the whole Uncanny Valley effect. When a human sees a still image, that person will be more forgiving of certain abnormalities than seeing the object in motion. Most of this has to do with subconscious acceptance of flaws.

Setting the foundation here is what we're all about. Using something that is iconic is to help establish something for the viewer/player. As we discussed previously, somebody like E. Honda or Barbie can't exist in real life. However, they are acceptable in video game, movie and comic environments because of the various aspects of their design that are acceptable to us to some extent. We know that it's unrealistic, so we accept those flaws because it is unrealistic. It's already "not real" so mentally, so it's ok. The closer you get to being real, the more unsettling the small changes are. So, for a quick example:

This is a work-in-progress of a Sagat bust that some guys at Gameartisans.org are doing for a Street Fighter bust competition

Here is an example of a Zangief:

Obviously, Sagat looks more realistic than Zangief. When I presented these images to my friends, the immediate reaction was "sagat looks like he is made of rubber". Nobody says "Zangief looks like he's made of marshmallow".

When things start moving, it's even *more* apparent. We subconsciously understand how human motion works. That's why, when we start seeing it not working right, it looks creepy or weird.

Here is the original music video of "Thriller".

Now here is the version that was included as an Easter Egg in the Final Fantasy: Spirits Within DVD:

After watching the two, do you notice a few things? Aki Ross and her team seem much more "puppet-like" than Michael Jackson, despite Michael Jackson's intentions to pretend to not even be human. Even at the very beginning, when it is focused on Jackson and Ross, you should notice small mouth movements. When Jackson does it, it looks reasonable. He's making minute musculature movements around his face and mouth and tongue. Ross... doesn't quite do the same thing. It looks like she's trying but it isn't right. And since it's close enough to be right, but still missing those minute tiny things, it looks downright weird.

So pulling this all together and going back to iconography: When making a game, you need to not only understand what it takes to make a character acceptable to a player in terms of the immediate look, but also in terms of that character's motion. A character (or even an object) that doesn't move right will instantly break that sense of immersion and feel for the character, and catapult the player out of the "zone" of being "in" the game, and back to sitting in front of the TV with a controller in his or her hand. That's the enemy and the bad part. As a game developer, this should be the bad situation that you'd have to avoid as much as possible.

For characters, iconic motions come into play. Most motion is done in 3 parts: The wind-up, the hit and the follow-through. These three parts provide the feeling of depth to a motion (often in combat, but it can be applied to many other types of animations). From a very good blog post by Eric Williams, he explains how attack motions are important. This sort of thing is called Keyframing, and is even an important part of old style 2D animations. Keyframers would draw the special frames that captured the essence of the action that was happening, and then "in-betweeners" would draw the frames that linked the key frames together. See if you can find the spots you think are the lead-in, connection and follow-through spots for Zangief here:

This applies even for things like firing a gun or kicking a soccer ball. When you fire a gun, you first aim, then there's a slight jerk, then the muzzle flash, and then the recoil from firing kicks in. The aiming is your lead up, the muzzle flash lets you know that something happened, and the recoil is your follow-through.

Iconic motion is something that may or may not really be effective in real life, but is still recognized as being good because it looks cool, and is acceptable to players. When you provide the lead-in, connection and follow-through, you make your motion much more iconic and acceptable to players even on a subconscious level.

Realistic vs Not Realistic

2009-02-13T21:18:34Z

Apologies for the lack of updates. I'll try to be better about it.

A lot of people seem to think that there's really a sliding scale between realism and not realism. On one end, you'll see something like Gran Turismo, and on the other you'll see Mario Kart Wii, and you'll declare those two the opposite ends of the scale. But I don't really think that's how it goes. It's actually fully possible to just replace the cars, karts, sprites, models etc. with abstract shapes and such. Sometimes things get weird like that, but what it does is change the way you view things. What most people are thinking of isn't that things aren't realistic, but that they are iconic.

According to dictionary.com, iconic in the artistic sense means:

2. 	Art. (of statues, portraits, etc.) executed according to a convention or tradition.

There's a lot of excellent examples out there. One of the most famous examples is this person (and I use the term "person" loosely):

Barbie is an example of an icon. She's not really physically capable of being that size. From what I have gathered, if she existed she'd be 7'2" tall and have the measurements 39-23-33. That's not even close to realistic.

Let's compare some other iconic styles.

vs

Looking at this, you can pretty easily tell the differences. The top is a realistic looking E. Honda. The bottom is an iconic E. Honda. Things are exaggerated, like Honda's muscular structure. He doesn't have manboobs (probably not really possible for someone of his build), he has sculpted abdominal muscles despite having so much weight. The iconic Honda has very well-defined muscles, despite being so heavy. That only comes from a low bodyfat percentage, which is impossible for any sumo wrestler.

The thing is that most games don't go for realism. They don't want to, because it doesn't provide the kind of escape from reality many people like. The importance is to note what the icons actually *do*. In Honda's case, it's a combination of things to make the player visually understand what Honda is about:

Honda is a sumo wrestler

Sumo wrestlers are big and fat-looking

Honda is a strong guy who can fight

Strong people must have large, well-defined muscles which will indicate strength

Fat people don't look strong.

It's all going back to conveying information to the player as quickly as possible. Most players already have mental connections for things like 'big muscles = strong' and 'sumo wrestler = fat'. When you see Honda, you see large muscles and a large body wearing a sumo wrestler style getup. It's (nearly) instantly recognizable to anyone who is familiar with sumo wrestling. This decreases the amount of exposition a game needs to do in order to convey who the character is. This is because players don't want to have to sit through lengthy exposition sections. They want to play the game.

"A picture is worth a thousand words", but the right picture can be worth ten thousand.

In the coming posts, I'll also talk about other forms of iconography. Notably, the iconography of cameras, iconography of motion, and when iconography works against you (the designer).

The New Prince of Persia, innovating in ways that were done before.

2009-01-15T21:46:53Z

One of my friends sent me this video:

After watching it, I thought about it for a time, but I'm not sure I agree with him. I do think that there is some merit to what he says, but most of it isn't quite right.

So, let me establish what he's right about. Learning is fun. Trying new things, then getting feedback from the results can be fun, especially as the skills you learn allow you to get positive results in an entertaining way. That's often where the enjoyment from a game comes from.

What he's wrong about is the "typical" game he describes, and how it penalizes you for what you did or mistakes you made. See, his viewpoint toward this is somewhat narrow because many of the games out there have exactly followed the sort of thing he's talking about and they all existed far before Prince of Persia. Classically speaking, there are heaps upon tons of games that all essentially place you back to where you were, just about where you ended. They're called arcade games.

Back in the day, video games were developed for coin operated machines. They were specifically designed to catch your attention in the first few seconds, be simplistic to pick up, ramp up significantly in difficulty over the course of a few minutes, and finally provide you the marginal return - once you lose, you have the option of putting in another quarter to continue where you left off. These games were designed to eat your quarters. They are supposed to be hard after you play for a bit. One classic example of this is Street Fighter. The game is very simple - push buttons to attack, move joystick to move, and beat up the other guy (or girl). There are special moves and deeper gameplay for players to learn, but it isn't really necessary to play the game. It usually starts off extremely easy, and then the difficulty of AI opponents rapidly ramps up. It's a design to keep you putting in quarters, which is why the later AI fighters are much tougher than their predecessors.

We already discussed the controller issue a while back, so I will only mention it briefly here. The controller complexity is not really the barrier to learning to play, it's the accessibility of the game, and the design of such. Street Fighter has six buttons and eight directional inputs to learn, but it isn't exactly rocket science for new players to pick up.

Prince of Persia isn't the first game to change the way the failure option works. Their previous iterations had nearly the same thing - you could rewind time to erase your mistakes. The difference is the layering of the failures. You had a limited amount of time you could rewind, and after running out of the rewind resource, you'd have to start over at your last checkpoint. It makes failing in the game two-tiered, by providing a penalty for failure (but not an overly harsh one), and then a bigger penalty for continued failure. This is not a new concept; many other aspects of life follow it. In a war, you can lose a battle but still win the war. However, many battles lost will ultimately lose the war.

Tying this all together, somebody asked me whether I thought that the new direction Prince of Persia has will better bring more casual, non-gamers to the more core gaming market. I'm not sure, because many of the core gamers complained about Prince of Persia being too easy. I think that the thing that Prince of Persia teaches us is that it's ok for a game to hold your hand if you want it to. I'm all for the advocating of making things easier for folks who want them that way, but I do not think everyone should ride with training wheels or put up bumpers for bowling all the time. That's only going to alienate people more.

The other thing one must consider is pacing. Often times, a game's "normal" difficulty is intimately tied with the pace of the game. Regarding Quantum of Solace (on which I worked), one of the major complaints in the reviews was that it was too short. A large part of that was directly tied with the difficulty of the game. There are actually 16 different missions in QoS, which is very similar in number to the missions (and length!) in games like Call of Duty 4 and World at War. However when I first played through COD4 and WaW, I died a lot more often. The deaths and setbacks in particularly challenging (to a brand new player, mind you) areas add to the length of the game, since you have to play through it over again. If you blaze through and can kill everyone easily, the missions take a lot less time to complete. That was one of the problems with QoS. I am not sure the game was too short, but for its target audience I wholeheartedly believe it was too easy.

Selectable user difficulty is a good idea. If the player wishes the game to hold his or her hand, that's perfectly acceptable. I think that more games need to embrace the more casual players out there and give them an opportunity to experience and enjoy the more hardcore-oriented titles out there. However, it is also important that those companies do not alienate the existing people who buy their games right now. That's the problem I see with games like Prince of Persia. By not making it a choice, you make the core audience feel bored with how easy it is.

Finally, the other thing that needs to be done is that the game needs to just make tutorials better. Games should have tutorial sections, and the tutorials should be in such a way that does not patronize the player for not knowing. Typical game heroes are larger-than-life and powerful characters. They save the world, defeat villains and rescue the innocent from terrible fates. They should not be treated like rank amateurs or like they don't know anything by the game itself; it's disrespectful and breaks the immersive feel.

A perfect example of a game that gets it right would be Assassin's Creed. Whenever the player unlocked a new combat technique, the main character would be pulled aside by a trainer and teach the player how to do it. However, the player's character (Expert Assassin Altair) is the one doing the teaching in the game. He shows the novice assassins how to perform these expert combat techniques, and in doing so both reinforces his role as a badass and teaches the player a new technique.

Will design philosophy like the new Prince of Persia really help with expanding the audience for "serious" games? I think it would, but it would also need to provide that element as a conscious choice by the player, and most definitely not in a condescending way. For example, in Prince of Persia's case, rather than allowing Elika to always rescue you, instead engage her in some sort of story element that involves unlocking her power to increase her ability to help. A player can choose how strong the seal on her powers is, providing a choice - infinite saves, a handful of saves, or none at all? Then, by not expending her power (in the form of saving your ass from falling), provide the player some sort of in-game incentive to play better, like achievements or power ups.

Edit: Looks like Capcom is taking the tutorial route pretty seriously.

Tell me a story.

2009-01-13T20:24:00Z

Occasionally, people will ask how the story for a game gets created. Realistically, story is another priority thing like many game development things are. You probably won't expect much of a story in Madden 2009. There wasn't much of a story for Quake 1 (if memory serves correctly, it was some hastily written .txt file that came with it). So where does the story come from, and how does that happen?

The answer of course, lies with the type of the game.

When you look at game development as a whole, you have to think about them in terms of systems and not so much in terms of a story. The story typically comes later; if you can't get the combat, the exploration, and the leveling up working right, you won't have much of a fun game, story or not. Often these systems take precedence over the story, and certain story elements will be built to fit the game systems. If you look at Shadow of the Colossus, the story of the game mostly is extraneous. It's well-written, but it serves more as the grease to keep a player going; there's no real reason to care about the story unless you want to. Any player who picks up Shadow of the Colossus could easily just have fun killing the various colossi and not give a whit about the sleeping girl in the temple. I'm not attempting to downplay the story in Shadow either; it's terrific and the storytelling techniques they used were pretty damn good. I'm just saying that you could (nearly) just as easily remove the story and the game would still be really fun and really good. The story could easily have been added afterward and adjusted specifically adjusted for the sake of the gameplay, whereas the gameplay would not need to make any concessions for story whatsoever. In the case of a game like Shadow, the story acts as a smoothing element between the major chunks of gameplay. Other examples would be games like Portal and the Mega Man games.

Sometimes, the game systems are general enough that you can pretty much put them anywhere, like with the latest Tomb Raider games. Lara jumps, hops and swings through the various areas like south america, asia and the other various ruins she finds. However, they all appear to be built using the same tileset of ledges, poles and such so it looks different, but fundamentally could be texture swapped and everything would still be just fine from a gameplay standpoint. There's no feeling that things like ice affecting her movement, or some sort of puzzle that involves diverting the lava flow to melt through something. The feeling is that the environments are somewhat generic, so they do not interfere with the story. Games like Prince of Persia, Spider-Man, Call of Duty, and such are like this. Many MMORPGs also fall into this category, such as World of Warcraft.

Certain games are more story-oriented than others. RPGs, especially Japanese RPGs are typically very story-dependent. Since the story has such an impact on the game itself, many game systems are designed to further the story. However, you also have to realize that the story is written with a game in mind. You can't just turn Tolkien into a game without making a lot of changes for the sake of gameplay; there's a lot of dialogue and exposition in the Lord of the Rings which doesn't make for very compelling gameplay because it simply happens while the hobbits and company are walking. There needs to be stuff for the player to do that is considered exciting or at least interesting. For a typical RPG, this means that the player must take an active role. There shouldn't be a lot of sitting back; this is why most RPGs require the player to go about talking to people and initiating the encounters him or herself.

Typical RPG story progression happens when the player is going from place to place, talking to people, finding various objects and defeating certain key bad guys. This is an example of where the story drives the gameplay. In Persona 4, for example, the main foci of the game are the stories of the characters. Building social links with various characters you meet is a huge part of the game, and would be very very boring without the emotional attachment the player develops with the characters. This sort of game is completely dependent on the story of the game, both for justification and enjoyment. In most RPGs, fighting random battles, finding objects and interacting with various characters aren't enough fun by themselves to justify the game, so you provide the fun by making these activities matter to the player through the emotional involvement - by meeting, liking and growing attached to the characters and wanting to see what happens to them.

This isn't to say that it isn't possible to make systems like that lots of fun. If you look at a game like Diablo (1 or 2), the focus is on finding items and building a cool character. The story is mostly secondary; people don't keep playing it for hours upon hours because they want to see what happens next, they want their sword of asskickery and their studly armor of studliness.

So, coming back to the original premise, the story is part of the game to involve the player. It comes from writers who use it as a tool, be it to involve the player, to provide some level of emotional attachment, or even as caulk to smooth out the breaks between segments of gameplay. Sometimes the story is more or less separate from the gameplay, and could be left alone if one so desired, but is there for players who are interested in it. Sometimes it's extremely central to the game, and impossible to extricate without hurting the game significantly. But in all cases, it's about fitting the need of the game, and always built as such.

What's in a job: Designer

2009-01-08T05:23:11Z

Hey all.

Once again, I've come to answer a question I get a lot: What does a game designer really do, and how do you become one?

A designer, at the core, builds content for the game. The programmers provide the tools with which to create content. The artists provide the building blocks with which to create content. It is the designer that actually uses these assets to create the content of the game.

Game designers work within a world of constraints, while trying to push the envelope within the set of constraints as much as possible. At the early stages of a project, they help come up with the constraints for the game itself. These are the arbitrary rules of the game, such as "Mario can only jump 5 tiles horizontally from a standing position, and 7 tiles when he has a running start". Other constraints may be added for other reasons, such as "James Bond cannot attack an unarmed opponent who is not threatening him", or "Our engine cannot handle more than 5 enemies on screen at once without causing frame rate loss". These constraints function as the rules within which the designer must create. There are two main categories of designer: Level designers and System designers. This isn't to say there aren't folks who blend the two or do both, but the roles are somewhat different.

Most game designers are level designers. When people say a 'level', the term tends to mean a specific zone or area of the game. Being a level designer means that you typically are in charge of a specific level or zone of the game. You create the geography, and you place the NPCs in it. You script the specific encounters there, control the flow and pace, and generally make everything try to feel fun. A level designer typically works with a 3D modeling tool, such as Maya, 3D Studio Max, Hammer, or Radiant. This job can get pretty big, so you'll often see teams working on levels, with specialists building the world (placing props and little details, making it look like a real area) and other designer types creating the encounters within the area and making it flow properly. A level designer is most responsible for pacing and the feeling of a specific portion of the game.

Some designers are system designers. System designers are a little more specialized (and generalized at the same time); they focus on systems in the game, rather than specific levels. This means that they work on general things that can occur in multiple levels. There are several sub-categories of systems designers. For example, World of Warcraft's character classes each have three separate talent trees. System designers are responsible for how the classes (their systems) function. Another example would be combat designers. In a game like God of War, there are many different things that Kratos can do when fighting. Different weapons, combos and special moves must all be designed, and they fall into this category because they are often independent of the level or zone they are performed in. System designers tend to be a bit more of a senior role than level designers, because it requires a broader understanding of the game in general, and junior or entry-level guys typically don't have the breadth of experience to do so. Metal Gear Solid's Stealth system and Prince of Persia's jumping and climbing system is a third.

The most important characteristic of a designer is actually not creativity. As the saying goes, "Ideas are a dime a dozen". The hallmark of an excellent designer is actually the ability to accurately convey his or her ideas through the written word with minimal explanation required. Designers are not artists, nor are they programmers. The programmers and artists are paid to create the tools and building blocks for the designers to create game content. However, without being accurately told what it is they need from the designers, they will not create what is needed. A well-written document should answer most/all of the questions a programmer or artist has about what it is they are creating. Of course, there will usually be a lot of questions, but the better design documents will limit the questions to being very easy to answer. An easy question to answer would be "How much time should the player have to press X?". A difficult question to answer is "What happens when the player presses X?".

This is why every designer job opening has the pre-requisite "MUST HAVE EXCELLENT WRITTEN AND VERBAL COMMUNICATION SKILLS". It's almost never "MUST HAVE EXCELLENT CREATIVE SKILLS". This also stems from the place most designers come from. Many designers (and producers, but that's a whole different blog post) have roots in QA and Test. This is because the required and core set of skills are very similar.

A game tester's job is also all about writing well. A tester finds bugs in the game, then has to accurately sum up:

1. How to accurately reproduce the bug
2. What sort of behavior the bug caused
3. (possibly) What should have happened instead.

Further, game testers very quickly become familiar with what is wrong with a game, because they see the flaws so often. It gives the more critically-minded among them the opportunity to see what goes wrong, and think about how to make it better.

I will warn you, though. Being a tester is probably the best way of becoming a designer, but is also pretty merciless and has a pretty low placement ratio. There are a LOT of testers out there (and there have to be), but it is, at the entry level, not highly skilled labor. The minimum requirements for being a tester are pretty much just the ability to play games and write reports. The best testers get promoted, and the others typically stay on a while and then move on (either to other careers, or up the QA ladder). It isn't highly paid work, and it doesn't get much in terms of benefits or job security.

There are, of course, other ways to become a designer. The real key is experience. Experience writing and conveying ideas, and experience in creating stuff for games. Right now, there is a wealth of free tools and things out there for game content creation. If you want to be a game designer, start doing it in your spare time. Create a mod. Create your own map. Make a Flash game. Make sure that you document what it is you do, so that people understand what you did and why you did it. Start building experience however you can, and it will set you aside from the other guys.

Ixobelle's raid design challenge part 2

2009-01-06T23:38:55Z

So, after glancing through the constraints on a raid encounter (previously posted), you're probably curious as to what it is I thought of. You might be thinking of your own raid encounters. That's cool, but this is my blog so here is my idea for a raid encounter.

I decided to go with a Hydra as the main encounter. It has many evocative traits that are commonly recognized by many people, and has a rich mythological background to draw inspiration from.

Specifically, I'm looking back at God of War, which had an excellent treatment of the Hydra, and can be translated (sort of) into a good wow raid encounter. It's got very evocative imagery too, and that helps establish a "cool" feeling.

So a quick little bit of backstory for it. Hydras are historically aquatic creatures. This is a giant hydra, one so big that it got stuck in its cave den. It's so enormous and strong that the denizens of its cavern worship it as a god. These cave denizens could be murlocs, could be naga, or could be any other aquatic race. The important thing is that the thing is established as being both big and powerful.

The second major aspect of a hydra is the fact that it regenerates, and as it does so it grows more heads. This means that you can start with this:

And quickly end up with this:

So, if this thing is so big and gigantic, how does one defeat it? Traditionally, the raid game of WoW's answer has been "Poke it in the ankles with knives until it keels over". But that doesn't work thematically. How does one reconcile it to a good raid encounter? Remember #B.

Using magical weak points has a long and storied history in games for years. It is at once both familiar and makes perfect sense. So I present to you the "win condition" for defeating the Hydra: Destroy its heart.

The actual area of the fight would look something like this:

The hydra is at the north end of the room, and there is a rather sheer cliff that the players cannot pass. Beyond the cliff is the Hydra's gigantic body. Its body would not be attackable; it's too big and your attacks would not harm it. Upon engaging it, the hydra would attack with a single head, like so:

A tank would start by tanking the hydra's head, and everyone focusing damage on it. Once the head is reduced to one HP, it will become 'stunned' for a certain duration of time (e.g. 30 seconds). When this happens, three things occur:

1. The heart of the hydra appears and becomes vulnerable
2. Adds spawn from both sides of the area (or just one side in smaller groups).
3. The Hydra head starts recovering health, and is immune to damage.

The heart is only vulnerable to ranged attacks; it is out of melee range. However, going back to #B, we take a little bit of love from Skadi the Ruthless: the adds will be of varying types (caster adds, melee adds) and have one (or more) Harpooners. Killing the harpooners would cause them to drop an interactable object (Harpoon), which can be used to fire at the heart, dealing a good amount of damage.

Scalable Difficulty Point - Achievement: Defeat the Hydra's heart without using harpoons
Scalable Difficulty Point - Achievement: Defeat the Hydra's heart only using harpoons

When the heart is vulnerable, the raid must spread its DPS - kill the adds, kill the heart. Of course, it starts off easy. This is where the hydra part kicks in. After the first stun wears off and its health is full again, the hydra roars and another hydra head enters the fray. Now the raid has two heads to deal with.

Defeating the "Lernaean" head puts it into its regenerative state. However, the Hydra will not expose its heart until all of its heads are (temporarily) defeated. The second head can be killed; it has no regenerative properties. Defeating both heads puts you back to the 'heart' phase. More adds spawn, the heart is vulnerable again. Optimally, a raid will defeat the non-regenerating head first, and then kill the regenerating one to maximize time on the heart.

Scalable Difficulty Point - Tuning: You can spawn more than one head each time. The saying goes "Cut off one head and two take its place".

After the second "heart" phase, a third head is added to the fight.

Scalable Difficulty Point - Tuning: The non-regenerating heads may require individual tanks to tank them, or a single tank can take more than one. In the 25-man version, perhaps you would need more tanks.

The fight will continue to oscillate between these two phases, adding more hydra heads each time. This would be a "soft enrage", wherein if a raid was unable to deal enough damage to control the adds and the heart fast enough, the raid would have to deal with an ever-increasing burden of Hydra heads each time.

Defeating the heart would cause the entire creature's body to shudder and die.

Now, to the extra flavor and optional bits we could add on to this to make it more interesting and compelling (in no particular order):

A. The Lernaean head has an aura that buffs the other heads near it, and must be tanked away from the other heads. Increase their armor, attack speed, or something.
B. Each head has a randomly selected breath ability. Poison, frost, fire, shadow, etc. The breaths can cause debuffs that are annoying to the raid (DoT, lowers damage dealt, lowers healing efficacy, slows cast time, etc.).
C. Adds may include healers to mend the broken heart.
D. Void Zone style situational awareness tests, such as rocks falling. Only occurs during the heads phase.
E. Defeating the non-regenerating heads causes some sort of area effect where they are handled. Perhaps providing a raid damage buff, or providing a void zone to escape.

Tunable points for difficulty:
A. Lernaean and non-Lernaean Head HP, Armor, Damage, etc. Can the extra heads be CCed?
B. Hydra head breath abilities. Can the debuffs be removed?
C. Amount of Heart HP and armor
D. Duration of each Heart phase.
E. Add HP/movement speed/armor/damage/etc. Can they be CCed?
F. Number of head spawns after each Heart phase
G. Usability of the Harpoons (Only for the heart? Can be used on the adds? Does it require a cast time? Can it be used on the heads themselves?)

Achievements:
A. Defeat the Hydra before/after a certain number of heads spawn
B. Defeat the Hydra with(out) using Harpoons
C. Reduce the Hydra's heart from 100 to 0 in a single phase
D. Defeat the Hydra without killing any of the adds (or perhaps only healer adds)

So that's my thoughts on creating a raid encounter design. The important thing here, of course, is that you tag all of the salient points, crib what works from other encounters, and package it all into a cohesive whole such that it makes sense thematically, and can be fun to play. I hope you enjoyed it. Comments and questions are always welcome.

Ixobelle's raid design challenge

2009-01-06T22:11:22Z

So I've become aware of Ixobelle's raid encounter design challenge. I've decided to throw my hat in the ring, because I found it interesting. I had some ideas in my head from before regarding how to make scalable encounters feel epic. I think I've gotten something pretty decent, and it will be coming soon. However, as I wrote this, it started to get kind of long so I've decided to break it up into two separate posts. This one will go into the normal sort of things to think about while designing a raid encounter, and the next one will be my design.

The first thing about designing anything is that you have to nail down your constraints. Designing within the constraints of a given system is very important - as a designer that creates content, you can't usually force the programmers to create new systems for you to add new functionality unless you can (usually) justify using it in many places, or you have enough clout. We'll assume that we're low on clout, so we have to work within the given framework.

So what constraints do we know about WoW: the raid content?

#1. It must be scalable in terms of the number of people needed. It should remain conceptually and thematically the same for both 10 and 25-man groups. If it can be done for a 5-man group, that's pretty great too.

Corollary: It should be scalable in terms of difficulty too. If you can take a harder path to victory, you should get commensurate rewards (achievements, extra bits of loot, etc.).

#2. It should have a time limit. Most encounters in wow now have a hard or soft time limit. Hard meaning that after a certain amount of time, the boss goes nuts and kills you for taking too long. Soft meaning that there is some sort of time factor that will eventually overwhelm you. For example, the stoneskin gargoyle monsters in Naxxramas will use an ability called Stoneform when they get low on health. If you cannot kill them before Stoneform finishes casting, they become immune to damage and regenerate most of their health back. This means that, unless your group can put out a certain amount of damage per second, you will never defeat them. Furthermore, they use a poison bolt volley that cannot be cleansed, and stacks up very high. The longer it takes to kill them, the more damage your group will take until the healers cannot keep up with the damage and eventually the group dies.

#3. It must limit the amount of randomness necessary. Any raid encounter should be repeatable with a high probability. If a group of players understand the fight, the encounter should not be difficult.

#4. A good raid encounter does not typically put too much responsibility on a very small number of people. There are encounters which become much more difficult when single players make mistakes. It usually isn't fun when 24 people suffer because of one person's mistakes.

#5. Mechanics used in the encounter should require minimal explanation. The KISS rule (Keep It Simple and Stupid) should be in full effect. Remember that complexity does not always equate to fun. You can make every raid member solve a differential equation to win an encounter, but that's not necessarily fun for many.

#6. Numerics can be tuned. Concepts cannot. Things like timing can be tuned and tweaked by adjusting factors like armor, health, damage, resistances, etc.

#7. Everyone in the group should have something to do that is interesting. There shouldn't be a feeling of "oh this is boring" to people.

Caveats to the constraints:

#A. Familiar is good. You want to provide a player some sort of good thematic basis for why the encounter behaves the way it does. This has a lot to do with theme. For example, Onyxia and Sartharion are fights with dragons inside of mountainous caverns with lava everywhere. They breathe fire, there is lava and dragonish eggs, it is all very thematic. With a dragon, you have certain things you can expect - fire breath, cleaving claws, and some sort of fiery theme. It would be very strange to find, for example, a giant walrus there.

#B. Imitation is the sincerest form of flattery. You do not need to reinvent the wheel each time. If it works at one place, you can use it for another. You should note that there are many mechanics reused across a variety of encounters. They are typically flavored slightly differently, but they are often similar in how they work. This also goes back to #A, because familiarity combined with consistency is a good way to reinforce that the player has learned something.

what the hell is wrong with people?

2009-01-06T18:05:17Z

What the hell? Seriously?

--Rawr

The myth of the ever-increasingly complex controller

2009-01-06T06:43:36Z

In today's post, we're talking about controllers and controller input. Somebody asked me a question about controllers, and why they keep getting more complicated. Further, will they continue to get even more complicated as time continues to pass?

Most folks probably grew up playing with one of these:

If you look at it, there are a total of eight total inputs. Up, Down, Left, Right, A, B, Select and Start. That's fairly finite, and the directional inputs are somewhat mutually exclusive (cannot push left + right at the same time, for example).

Controllers have evolved significantly over time. It's fairly normal to see how they could have. Moving from the NES, we went to the SNES:

The number of inputs has increased. Now we have two additional face buttons (X, Y) and two shoulder buttons (L, R). Our total number of inputs has risen from 8 to 12. 50% more things to keep track of.

This further evolved into the Playstation controller, followed by the Dual Shock.

They add another pair of shoulder buttons, and now add left and right analog sticks. There are also two 'hidden' buttons... you can actually press the right or left analog stick in, which functions as 'another' button. This raises the total number of inputs from 12 to 18 (adds two sticks, two stick-press-in-buttons, two shoulder buttons).

Another variant of this is the X-Box 360 controller, which has the same number of inputs, but puts the left side analog sticks in different places.

And finally we have the Wiimote. The wiimote is a bit odd to try to classify. You have 11 face buttons on a wiimote (A, B, +, -, home, 1, 2, and 4 directional pad directions) in addition to motion sensing.

Now, looking at this it sure seems like things are just getting more and more complicated. And to some extent, they are. However, approaching this from a design perspective, you have to key in on a few things:

#1. Many of these inputs are mutually exclusive.
#2. Humans can generally only quickly grasp a certain number of things (short term memory) well. Trying to push more than that too quickly often leads to confusion and the inability to remember.
#3. The ultimate goal of a video game is to convey an experience.

Let me break this down a bit more.

#1. Many of these inputs are mutually exclusive.

If you're using a Dual Shock controller, you're likely to be using either the directional pad *or* the left analog stick. In order to use both, you'd either need two left thumbs, or to take your right hand off of the right side of the controller. Neither of these is particularly comfortable. For this reason, control schemes involving extended use of both of these pseudo-mutually-exclusive inputs heavily emphasize only one of them. The game is designed such that you can do both at once, but your actual necessity to do them both at once is very low, to the point that is nearly nonexistent.

Example: Mario Kart Wii allows you to play with a Gamecube Controller, rather than the Wii Wheel. In order to perform a wheelie or a trick when you do a jump, you must press 'up' on the directional pad. However, most of the time your left thumb will be on the left analog stick for steering. The only time you don't really need to steer is when you are going on a straightaway (when you would use the wheelie), or when you are going into some sort of uncontrolled area (like a jump, where you can do a trick quickly, then put your thumb back on the stick).

#2. Human Short Term Memory is not very good

Most people tend to start forgetting things they put into short term memory at around the 7th thing. This is why most normal phone numbers are 7 digits. The numbers go up to 10 at a time (which is why area codes are only 3 digits), but those last few tend to be pretty tough to recall when it comes to short term memory. This means that if a designer wants his or her game to be fun and easy to pick up, the control scheme has to fit comfortably into short term memory. For example, think about the inputs when given a driving simulator: Throttle, Brake, Steer. That's it. Two inputs and rotational control. How about a normal First Person Shooter game? Aim, Move, Fire, Secondary Fire (e.g. Grenade). Those are generally how almost every first person shooter in the world can be played. Some add extra bells and whistles, but they aren't necessary to immediately get a 'feel' for the game.

Built-in mnemonic devices also help a bunch. Take Street Fighter, for example. There are seven major inputs (joystick + 6 buttons). Could you imagine if the buttons were punch, kick, throw, block, jump, and duck? It would have been a controlling nightmare. The Street Fighter control scheme is terrific, because you have punches on top row and kicks on the bottom. As you go left, you get weaker, faster attacks. Right side is stronger, slower attacks. The defensive technique is just holding the joystick directly away from your opponent, which is the completely natural response for defending - you want to get away. These built-in mnemonic devices are easy to pick up, and make it a lot easier to grasp the controls immediately.

#3. The ultimate goal of a video game is to convey an experience to the player.

This part deals with the 'future' of game controls. The purpose of any controller is to give the player instant access to whatever it is he or she needs to do in order to play the game.

Sometimes, the experience is something that the player can physically model, such as driving a race car. With controls like this, the analog sticks went a lot further than the old digital pads because you could press a stick all the way, or you could gently nudge it to differentiate your controls - much like you would slightly rotate the steering wheel for a gentle turn, or rotate it a lot for a more sharp one. The natural evolution of controls like these are to simulate the experience as much as possible, and provide as much immersion as one can, without overly complicating things. Take a look at Rock Band and Guitar Hero for excellent examples of this.

Other times, the experience is something that the player cannot physically do, such as expert martial arts techniques, magical spells or firing weapons of mass destruction. The goal is still the same though; you have to provide the player an avenue of input and control to make their avatar on screen still *feel* like them. The player needs to feel a sense of ownership - feel like it is him or her performing those actions and not just watching from afar as a computer controlled character does it. Since there is no real easy way for a controller to accurately convey *everything* a ninja can do, you have to break it down into the things you can expect a ninja to do *most* of the time. Typically this involves something like move, kick, jump, weapon attack (or ranged attack) and maybe use some sort of selectable ninja magic attack. This covers most of the general things a player would expect a ninja to be able to do.

Generally, most controllers get more complicated just because different games will tend to push the envelope when it comes to how the control scheme directly affects the feeling of immersion. The developers of the console try to cover as many bases as they can, but you've already got to think that they are pushing the number of buttons a player has to remember. For the future of the controller, I expect that the number of buttons won't significantly increase, but the methods of control (to facilitate immersion) will continue to become more broad. We have a lot of buttons to do things with, but what controllers are lacking nowadays is the immersion.

--Rawr

Serving up DLC

2009-01-03T01:55:16Z

One of the biggest effects of the advent of online console gaming has been Downloadable Content (DLC). Once limited to PCs only, DLC is now available to console gaming. Furthermore, console gaming is a lot *bigger* than PC gaming, so it's starting to make quite a buzz. There are two major things about DLC that I get questions about a lot, so I'll try to outline the first today, and the second at a later date.

Here's the question I get the most.

Why do I have to pay for DLC?

I get it. The economy is a shambles, and you're less willing to part with your hard earned cash than you may have been in the past. Furthermore, it jars you because the same game on PC offers the same downloadable map pack, costume pack, or what have you for free. But the one you own, on your console, requires a price tag of $8.99, payable through whatever applicable currency. Where do these crazy publishers get off charging you money for something that should be (and sort of is) *free*?

Here's the thing. There's no such thing as a free lunch. You know it, and I know it. The "free" map packs for the PC version aren't free. When was the lat time you downloaded something from the web without ad banners everywhere? These online distribution sites earn money via their advertising revenue. Further, they don't have as much in terms of overhead costs associated with them. Sure they have to pay for bandwidth, and that's still the major overhead for any sort of digital data distribution scheme. However, there's several key differences between DLC on (for example) X-box Live and the hairy scary internet.

Remember that XBL is not the same as the internet. XBL is a service, a sort of subset of the internet. Both you (the adoring public) and publishers like Activision, EA, Eidos, THQ, etc. pay Microsoft to use that service. Part of that is less advertising to annoy you, but the other part of it is that Microsoft accepts the overhead costs of keeping the system running, tracking things like buddy lists, the ability to join games, etc. and in return the games that are playable on XBL give them some money back either as a portion of the game sale price, or some other sort of contractual agreement. This is how console makers earn money even when they are still losing money on each console they make. They get a portion of the proceeds of every game sold for their systems. Part of those proceeds can also go to paying the company for online support (or there could be some other contract worked out as well).

So we've established that XBL is not free for anybody involved. We also know that hosting stuff on the internet for downloading is not free either (bandwidth costs at minimum). All these add up to a bill somewhere. If you get a DLC for free someplace, that just means the bill is being paid by someone else. Furthermore, if there are other rights-holders involved, those guys have their hats in hand as well. The most famous examples I can think of are Rock Band and Guitar Hero music packs. If you've got a musical group, they (or often more accurately their publishers) own the rights to their music. They need (and deserve) to be paid for their work. All of these result in costs to the publisher who is putting out the DLC.

Somewhere out there are financial analysts whose sole purpose is to figure out just how much they can make from this sort of thing. If you price the DLC at $5, how many packs are you estimating will be sold? If you price it at $8, how many? And they also have to consider other issues, like whether it will be better to take a loss on it, but use it as advertising for the coming Guitar Hero 5: Galactic Tour (or whatever). Much of the time, DLC starts off at a certain price, then the price goes down as time passes, and it eventually ends up being (nearly) free.

And I'll just point out that I haven't even talked about how the content itself costs money to develop!

So let's bring it back to the original question. Why do I have to pay for DLC? You pay for DLC because DLC isn't free. It always costs somebody something, and for online distribution channels like XBL and PSN, the costs are often passed on to you simply because the publisher has decided (through the assurances of the analysts) that you're willing to help foot the bill. Those costs can get defrayed by some things like advertising revenue, but for the most part it's because there's a whole lot of people who want to make money getting the content from the publisher to you.

--Rawr

Coming soon...

2009-01-02T01:36:14Z

In the spirit of the new year, I've decided to write more. Each of the posts with the "faq" tag are questions I've received about the game industry at some point or other. I've started collecting a list of topics to write about, and they will appear every couple of days here.

I'm not going to claim 100% accuracy, but these are all reasonable conjectures and experiences I've had from what knowledge I have of the industry. Your mileage may vary, standard disclaimers apply.

--Rawr

QoS E3 trailer

2008-07-15T18:17:11Z

--Rawr

Throwing stuff at the wall and seeing what sticks.

2008-07-12T18:34:49Z

Two posts in one day... amazing.

It looks like Flagship Studios is going under. That's pretty disappointing, considering who they are... that was Bill Roper's gig. However, I'm looking at a bunch of these "children of the Blizz" studios and noticing some trends.

Castaway Entertainment ate it about 3 months ago. Flagship just bit it recently. Arena.net continues to push the pseudo-MMO with their guild wars franchise... which, while not bad isn't anywhere near the resounding success of previous Blizzard titles. Not much news out of Red 5, who had a good amount of funding. What does this mean? Why are these developers floundering?

The more I think about it, the more it comes down to Blizzard's particular style of game development, and how it doesn't work very well for anyone who isn't Blizzard. Blizzard is very undisciplined with their design. They make sweeping changes, they're never on time, and they often have to spend huge amounts of time prototyping and trying out various features before realizing that the idea sucks and should be scrapped. However, they have proven to be very good at figuring out when something is fun, and polishing it to a fine sheen once they've actually found it. The problem lies in finding it; that's where all the time goes.

I was one of the lucky handful to take part in the original starcraft beta, way back in the day. That beta process took several months... if I remember correctly, it was toward the end of 1997 that it happened. Back then, Starcraft was a far, far different game than it is now. Archons could mind control. Templars, Queens, Defilers and Science Vessels had attacks (Queens even had the glaive wurm attack). Mutalisks had a short-range 20-damage acid spore attack. Wraiths could not attack ground units unless they researched an upgrade at the Control Tower. Scouts used to do 15 damage to ground units instead of 8. And so on and so forth. That was a long time ago, and the game has changed significantly since then. If you look back at Warcraft 3, Diablo 2 or World of Warcraft, you'll notice similar sweeping changes in their histories too.

So what's this got to do with the other studios kicking the bucket?

Blizzard is given a lot of leeway when they make games, because the game ends up being very good when it finally *is* finished. They sell a lot, and have a good, strong brand name. This is because the Blizzard executives are actually very good at their jobs, and protect their brands with near religious zealotry. Whenever Blizzard puts a game out, they can pretty much guarantee several million sales. The Blizzard execs are that shield between the developers and the higher ups that want their product NOW.

The problem is that Blizzard's developers are not disciplined. Making changes willy-nilly throughout the entire development is pretty nerve wracking, since it puts things into a state of constant flux. When I say "disciplined", I mean things like finalizing certain basic core functionality, in order to build areas around it. For example, think about a game like Super Mario Bros. The levels are built with a few basic systems. Mario can make a maximum running horizontal jump of X units, and a standing horizontal jump of Y units. Mario must run at least Z units before making the full length jump. Then, based on these assumptions, levels are laid out. No jumpable gap can be wider than Y units, unless it has at least Z units of run space before the jump begins. Then the jumpable gap can be extended to X units.

Now suppose that later on, the designers decide that Mario needs some sort of special at-will power, where he can crouch for two seconds to charge up, and then make a horizontal jump of X+5 units (further). All of the previously difficult jumps just got trivialized, because rather than requiring players to run and jump and barely make it, they can just crouch and jump across with no timing or skill requirements. All of the work spend balancing those levels gets a monkey wrench thrown in, because of late additions to the game that weren't there when the level was being built.

In some cases, this can have weird results on old content. World of Warcraft has a ton of issues stemming from this... legacy items have some unintended effects on new content, or new stuff makes old stuff completely trivial. This sort of thing often invalidates a lot of work (and thus time) spent on building things. It isn't fun to spend five months building a square house, only for the architect to decide he wants a round one instead.

You *can* make a good game this way though. Blizzard's proof of this. It's the reason why they are always so late on their release dates. However, not many studios out there have the kind of clout that Blizzard does. Blizzard can do this because they can afford to. They know that they've earned a gazillion dollars in the past, and that their owners will extend them the courtesy they need, even if it means another two years of development and several more million dollars in funding.

Picture a studio that doesn't have that kind of clout. If they go with a similar strategy, they end up with a bunch of stuff that doesn't get ironed out. However, if they ask for an extension, often the publisher will not grant it... or grant it, but not give them any more money, because the publisher has a contract that says they will give $X million for a game by Y date. They are faced with a hard choice - ship the game in the unpolished state, or face possible starvation to make a better game. Most studios choose the former, because people like to eat, and because good game does not always equate to great sales.

There's an old concept floating around, showing that the earlier in a project you change something, the less expensive it is. Conversely, the later in the project it is, the more expensive it becomes. That illustrates this concept completely. I fully believe that if more studios had the leeway that Blizzard does when it comes to development, we'd have a lot more good games. But that's mostly a pipe dream, because people need to eat, and games need to sell for that to happen. You're not going to earn as much money from a movie licensed game that isn't released with the movie. You're not going to earn as much money from a football game that isn't released at the beginning of football season. People aren't going to keep giving you money unless you can convince them that it will make more if they keep giving you money, and you work on it a little longer. That's just how it is.

--Rawr

2 years of work on this...

2008-07-12T17:11:53Z

--Rawr

It's What? AGAIN?!

2008-06-09T20:41:35Z

So... one of the things I've noticed lately is that there's a rise in the buzz about Call of Duty 5, now that some details have been released. The first thing to note is that it is, once again, set back in World War 2. Again. The second thing to note is that there are a lot of players who are unhappy with that, because at least 10 million people all liked Call of Duty 4 enough to buy it. That's a hard act to follow for certain, and now people are getting up in arms and angry about it being a WW2 game.

What a lot of folks just don't seem to understand is where the studio begins and where the publisher ends. Let's look at a couple of things, in order to better understand the situation.

The first thing to note is that most games take around 2 years to develop. It takes time to make a good product. You need time to design the features. Time to iterate to see if the ideas are fun enough. Time to polish. Time to build content around those features. Grand Theft Auto 4 had a team of over 1000 people, was in development for 3 and a half years, and cost around $100 million to make. That's not chump change, folks. Specifically with Call of Duty 5, you have to look at the timeline. They are shooting for a release this year (late 2008). This means that, working backward, Call of Duty 5 was started late 2006.

Now... when did Call of Duty 4 ship? November 12th, 2007. As much fun as modern warfare is (and don't get me wrong, it is an excellent game), Call of Duty 4 at the time was still in development. Activision did not know how well it was going to do. They made a choice at the inception of COD5, and decided to go with 'old reliable' - a WW2 setting.

Now, it's easy to look back with hindsight and say 'well, they should have followed it anyway'. And maybe that would have been the right choice. However, I would like to point out that it's just the way the publisher/development studio dynamic works. There are usually two different approaches out there. The PRO and CON are from the dev studio's perspective.

#1. Independent Dev Studio is working on a game, pitches it to several publishers and tries to get them to bankroll the project.

PRO: You get to make the game YOU want. Artistic integrity!
CON: You run the risk of being unable to find a publisher. No publisher, no money.

#2. Publisher has an idea for a game they want made, and offer the project to dev studios. Dev studio accepts the contract, and makes a game based on the publisher's idea.

PRO: Steady Income! Dev studios don't make games if they have no money. I like eating, and would prefer to be able to keep doing it.
CON: Sometimes the idea is kind of lame, and you're forced to work on something you don't quite like.

This isn't specific to Activision, or EA, or anyone. This is business, and how stuff gets done. If I ran a dev studio and EA came up to me and said "Hey, we'll give you $30 million if you make us a game about racing dolphins across rainbows", then I'd make a game about racing dolphins across rainbows. I'd have more creative control, of course. Third person vs first person? Different types of dolphins? Do some dolphins swim faster, or some corner better? Can the dolphins attack each other? Can we distract enemy dolphins with rainbow fish? etc. etc.

These would all be within my power as studio head. However, I've got the contract with the publisher still, which says stuff like "We want at least 5 different dolphins" and "You are not allowed to kill anyone at all". The publisher will also want to periodically check on our progress, to make sure that their money is being well spent. They don't want to come by and see some crappy 2D flash game while the dev team plays world of warcraft and sips hawaiian drinks.

In reality, sometimes you don't get a choice about what the game is going to be about. The game I'm working on right now, for example, is a movie-based game. We're trying to make the game as good as possible, but we're limited by the rules set forth by the license-holder. They say 'The main character must be this way' and he must. For example, our main character can never kill an unarmed person for no reason. He can never shoot first without probable cause. He has to be a hero, and can never be shown in a villainous light. The license holders reserve the right to yank it if they see something they don't like.

Bringing it all back... I'm not in disagreement that WW2 games are probably old hat by now. However, sometimes it's the publisher that says "This is how it's going to be", and then that's all you can do about it. The only thing to do in that situation is to make the best game that you can given your restrictions, and hope people judge the game on those merits rather than circumstances beyond their control.

--Rawr

(Mostly) Fixing arena point selling.

2008-04-24T18:24:00Z

I got into a discussion this morning and talked a bit about what's currently causing the arena system to be exploited. I came up with a fairly simple solution, something that I think would solve most of the problems with the transformation of arena ability into commodity.

Right now, across the various battlegroups, you've got a number of high-end arena players. Each of these people enjoys the competition, plays hard and enjoys the game. Some of them are better than others, but they all love it when a new season begins because they can duke it out amongst themselves, and get new gear for it.

It's a good time for them to be around, because it's what PvP players enjoy most... playing hard, fighting intelligent foes, and getting rewarded for it. However, after the party starts to wind down, things kind of get messy. Here's why:

There are 5 pieces of any given Arena player's equipment set. The total cost of this is 8250 arena points. In addition, there are weapons. To fill out all slots, you need:

Weapon(s): 3750 points for melee, 4250 for caster, 1000 for hunter
Ranged slot: 1000 for anyone but Hunter, 3750 for Hunter

This means that for an entire season's gear set, players will need between 13,000 and 13,500 arena points. Herein lies the problem. Now, let's consider the length of an arena season. Season 2 began on June 19th, 2007 in North America. Season 3 began on November 27th, 2007 in North America. It is currently April 24th, 2008 and there is no Season 4 yet. Season 2 ran for 161 days, 23 weeks. If the timetable is accurate, Season 4 will begin May 6th (23 weeks).

What does the length of time have to do with the problems with the Arena? The answer is pretty simple. The Arena calculator will tell you exactly how many points per week you can expect from the arena, given a specific rating. Let's say that, for example, you are part of a high-end team that has 2100 rating. If I am on a 2v2 team with 2100 rating, I earn 892 points per week. On a 3v3 team, I earn 1033 points per week. And on a 5v5 team, I earn 1174 points per week. Furthermore, there is point carry-over from the previous season. When season 2 began, most people who were ranked 2000 or higher already had 5,000 points saved.

13,500 - 5,000 = 8,500 points left to spend to get all available arena gear. On a 2v2 team, that would take 10 weeks to obtain. On a 5v5 team, that would take 8 weeks to obtain. Notice something here? The math trails off, because there is no further need of spending arena points. 10 weeks into the season (less than halfway through), your high-end arena player has nothing left to spent points on unless he or she wants to try offset or random extra gear for the fun of it. Let's say that they wish to save 5,000 points for the next season as well. That's still only 16 weeks spent to accumulate enough points for a full set, weapon(s) and ranged slot. You still have 7 weeks of literally wasted points. Furthermore, it is highly unlikely that those in this bracket (2100) would have a shot at being the top 0.5% that qualify for the title and special mount for "winning" the season. On the Bloodlust Battlegroup, a 2100 rated team is ranked #76 for the entire battlegroup in 5v5. For 2v2, a 2100 team would be ranked 198th.

So what do you do when you have an extra 7-12 weeks to do nothing?

You sell your time. People will pay heavy amounts of gold for arena rewards, especially if they are incapable of obtaining such rewards themselves. Gold is easy to obtain, it is fully possible to do a number of daily quests and receive 200, 300 gold in a single day. Given those numbers, can it really be that difficult for somebody to sell a week's worth of arena points (say, a good number like 700 points) for 1,000 gold or more?

Ok Rawr, I get it. The arena players are bored, so they spend the time to enrich themselves. So what though?

The problem is that these point seller players are heavily geared, heavily experienced, and will just plain roll over anyone they run into in lower brackets. A new team, or an average team, or a bad team will just roll over and die. The whole point of the ELO rating system is that players will advance in opponent difficulty until they reach an equilibrium point and stay there (approximately 50% win/loss ratio). Thus, people who stay in the 1500 bracket should win approximately half their games against other 1500-rated teams, and lose the other half, keeping them at this rough equilibrium point. In the end, what happens is that the low end bracket (1500-1800) is dominated by these players who cut a swath through all of the teams they face, because though their *current* team's rating is only 1500, their *actual* rating is 2100. This means that where a team *should* be seeing roughly equally geared and equally skilled teams, they don't. They instead get trampled by much better teams. It's like people who are ranked as chess masters entering a tournament for amateurs.

Fine, so they make the lowbie team's lives miserable. How do you fix it?

There will always be point selling, there's not really any way to get around it. Where there is demand, there will always be a few who will supply it. The important thing is that right now, there's nothing else for these arena players to do. They've got nearly two months of nothing to do. So how do you fix boredom?

First, you change the way points are earned. Instead of giving people the highest of the available points, you give them points based on what teams they are a part of. Since 2v2 gets 76% of the points of a 5v5, and 3v3 gets 88%, you have to rebalance and redistribute. You give a standard arena point distribution formula, and calculate an 'effective overall rating'. [ (5v5Rating * 100) + (2v2Rating * 76) + (3v3Rating * 88)] / [(100 * On5v5Team?) + (76 * On2v2Team?) + (88 * On3v3Team?)] = Overall Rating. Overall Rating would be calculated based on a weighted average of all teams a player participates in. If you are good at 5v5, but not at 2v2 or 3v3, your 2v2 and 3v3 team ratings would drag down your overall point totals. This change would address the problems with selling points on your 2v2, while keeping your 5v5 around for earning points.

Then you give them an outlet for the extra arena points they are earning, so that they have a reason to continue playing and earning points for themselves. To do this, one needs to give them rewards that they would be interested in obtaining. The easiest:

Give them raid gear. Season X should make Tier X+2 gear available. Thus, Season 4 makes Tier 6 gear available. Season 3 makes Tier 5 gear available. Season 2 makes Tier 4 gear available. Put rating requirements on the purchase of raid gear equal to that on the arena gear. This does two things: it encourages players to experience more of the game, and provides opportunities for them to 'catch up' without having to go through the nasty business of hopping from guild to guild until they get to one on the leading edge.

Furthermore, other types of consumable items should be available. Epic quality gems should be available for purchase. Armor kits, enchantment kits, etc. should all be available for arena points. Nethercleft leg armor can be made or it can just be straight-out bought. Finally, you need to provide fluff items like non-combat pets, special looking mounts, etc. that cost a lot and are only cosmetic... but it would be a status symbol. People would buy them.

And that's how you fix it. Make it so that the high end player can't actually buy *everything* for the season 2/3 of the way through it (or even all of the way through it), and you've got something for players to strive for and earn, which will curb the point sales.

People will still sell points, of course. However, it comes as an opportunity cost to them - by selling arena points, they will lose out on their own maximum point earning potential, and thus lose out on the rewards they are aiming to get.

--Rawr

Foundations, and why they are important

2008-04-22T23:15:25Z

So once again, I am reading what looks like yet another hardcore/casual player debate and I had an idea for a MMO Project. Specifically, the rollout of content for MMO players. The biggest problem from the beginning of the Burning Crusade was that, despite his name being everywhere, and his voice being everywhere, and his likeness being plastered on every single advertisement for the Burning Crusade, the quintessential Burning Crusade character (Illidan Stormrage, naturally) very rarely actually showed up to deal with the player. You sort of dealt with him through a couple of quests here and there, but he almost never really made his presence known. At the end, he's just a long and involved boss battle in the black temple for 25 hard core raiders. Keep in mind that, according to the current data at wowjutsu, 6.41% of all ranked guilds (EU + North America) have killed Illidan Stormrage. That's an awfully small number of people compared to the total they have playing. They've already stated in interviews that they wish to make the NEXT big lore figure more accessible to the general player.

That got me thinking.... (a dangerous pasttime, I know)

In this case, I will present: Rawrasaur's version of Wrath of the Lich King.

The Problem: People consume more than can be made

The biggest problem with this is that rolling out content is hard. Players consume it too fast, designers/artists/programmers cannot create it fast enough, things get out of hand. To illustrate a point, less than 48 hours after the initial launch of the Burning Crusade, some player was already maximum level. When the expansion pack was rolled out in China, it took chinese guilds less than 7 weeks to finish every last bit of raid content available at the time. This includes actually leveling from 60 to 70 as well. People consume content way faster than people can make it.

So what can be done? Remember this is about foundations, not just a one-solution-fixes-all.

The First Pillar: Artificial Real Time Delays

This has been done a couple of times before. The 25-man raid instance introduced in patch 2.4 is initially only half-open. The first three bosses (of six) are available to be slain, but the fourth, fifth and sixth will only be released after a few weeks of real time have passed. Similarly, the deployment of content on Sunwell Isle is on a time-released basis... more specifically, it is tied to doing daily quests.

Here's how it works. Every day, players can go and do quests on Sunwell Isle to make progress toward certain goals. The server tallies the total number of quests completed, and when the tally passes a certain number, it moves to the next phase of the area, unlocking new quests, new areas, new vendors and the like. If players want to grind reputation with the faction faster, they can do so by killing monsters, or going through an instance or whatever. They can do it if they so desire. However, this will not necessarily make the new content come that much faster, since it is still on a per-day update at most. By spacing out the content this way, you artificially increase its longevity.

There is a key element to this though, since we had the same system in Silithus and it did not do as well as the Sunwell Isle rollout.

The Second Pillar: Delays should be discrete

The opening of the Ahn'Qiraj gates was similar. Everyone does quests, and eventually the gates would open. The problem with this was that there was no real sense of progress. There was no expansion, there was no feeling that you were anywhere near being closer to your goal at the start than at the end, except for an NPC that says "We have collected 97% of the resource we need".

By giving smaller rewards and spacing them out as smaller "milestones" of progress, you encourage players to keep going. Silithus did not do this. Sunwell Isle did. By giving a preview of the carrot, and saying "Hey Players, go do this and you can get the carrot", you give players a discrete goal to work towards. And guess what? They'll probably go do other quests while they are there to line their pocketbooks and further any other goals currently in the area, since it is convenient to do so. Players want feedback. They want to feel like they are making progress. By giving them intermediate advancements, rather than a binary off->on situation, you give them a sense that they aren't sisyphus pushing the boulder up the hill. Give them small rewards and conveniences (this flightmaster needs help gathering wombat hairs to set up his shop. That innkeeper needs you to chase away the angry dwarf molemen in the area to set up her Inn. etc. etc.) to work toward, and they will do so.

Furthermore, to drive the point home, Arthas should appear during some of these milestones. Perhaps he is furious that the puny Alliance and Horde have taken another of his outposts. He shows up when this happens, leading a force of undead warriors and does battle with the defenders of the area for a while; the players should feel free to join in the battle and drive him back. When he is weakened sufficiently, he will retreat, and swear vengeance. You drive the point home that Arthas is a very real and very powerful enemy, and that he does not take kindly to the players encroaching on his territory.

The summary of the First and Second Pillars

The key to all this is to tie the steady roll-out of content in to the storyline. The perfect example would be that the Scourge armies are battling the Alliance forces, and you have to actively take back zones from them. Each zone you occupy and take back literally weakens the power of the Lich King, because the players are literally taking away his resources. By pushing forward like this, you give the players the feeling of advancing their cause through their own work, while still enjoying the game and not feeling forced to have to poopsock in order to get it done. This should, of course, extend to the 5-man dungeons with daily quests to take back these macguffins to weaken the lich king or strengthen the allied/horde forces.

This wouldn't need to be done for the leveling up areas, but would be good to sprinkle a couple of the advancement quests through these areas anyway. The majority of the sort of time-released advancement content would have to be in the zones designed for maximum level players.

The Third Pillar: The Hardcore should have something to do

One of the good ideas that came from the Gates of Ahn`Qiraj opening was the idea that the raiders had their own quest line to accomplish, which involved a lot of work on their part. It was pretty hardcore in its requirements, but that's ok because it was directly targeted at the hardcore, and certainly not required for everyone to do it... only a handful of people on each server. Better yet, the reward it granted is a perfect example -- bragging rights and something cool to look at, but not really anything that would affect gameplay.

So what does this have to do with Wrath of the Lich King?

Let's say there are four different raid instances that ship with Wrath of the Lich King. We'll call them A, B, C and D. Rather than making them all available initially, we'll go with the time-released method again. The leaders of A, B, C and D are all lieutenants to the Lich King himself... important and powerful figures, with strong minions and each guarding a specific item of power that Arthas obtains his powers from.

Initially, only perhaps A and B are available to raid, as the first tier of raiding. After A and B have been slain on a server, an internal timer begins. A number of weeks later, C becomes available to raid. C will be the next step up in raid progression (a higher tier), and when C is completed, another internal timer begins. A number of weeks after that, D becomes available to raid.

Arthas should definitely appear in these raids as well, as encounters where he is driven back (but not defeated). He should be more of a supporting role, helping a specific fight, summoning an enemy for the raiders to deal with, or perhaps even a short battle with him appearing to empower his lieutenant for a time, and then leaving. This should specifically focus on the macguffins of power that each of his lieutenants are guarding, because this is, indeed, the point. The raiders are taking out the special items of power and slaying Arthas' most powerful lieutenants.

I would also be encouraging raids E, F, G and H being 10-man encounters, with similar rules.

The Fourth Pillar: Tying it all together

A number of weeks after the final raid encounter is defeated and the final area claimed from the Scourge, there should be a server-wide event and the opening of a final instance. The Scourge has been beaten back, and the lands reclaimed. The Lich King has been weakened through losing his lieutenants, and it is down to a final battle between the players and the Lich King himself.

And now the players have a choice to enter the throne room... a large raid encounter, a small raid encounter, or a 5-man encounter. The raid encounters allows for a more esoteric and wacky fight, the small-raid encounter is a bit less esoteric and wacky, and the 5-man is definitely much more straight forward. However, you fight the Lich King in all three (depending on which way you want to go, and based on your playstyle preference). Perhaps you have some more legendary heroes showing up to assist you in the smaller groups. I don't know. Regardless, it is an opportunity for all players of all walks to test their mettle against the Lich King, and for each and every player to look back and say "Yeah, I really felt like I was a part of that, and I got to face the lich king and win."

The important thing is that the entire expansion be built with this model in mind. This isn't something one can shoehorn in after the fact with content patches. One has to plan out a reasonable time table for the entire expansion to be rolled out with this model from start to finish, including a rough schedule for how long it will take to finish, and when the next expansion should hit. Because, as we all know, players are voracious when it comes to consuming content...

That's how I'd build Wrath of the Lich King.

--Rawr

Lifetap nerf and what it really means.

2008-03-03T22:21:58Z

So I've been told that there will be a lifetap nerf in the next wow patch, and a lot of people are up in arms about it. There's really not that much to get up in arms about; it's an extremely silly thing to lose sleep over.

It's actually pretty damn simple, and I can break it down into 4 succinct categories of effects it will have.

Here we go.

Issue #1: It will make life tap more dangerous in pvp

By tapping for more HP, warlocks will have to be more careful about when and where they life tap in ArenaVP. This is an intended change.

Issue #2: It will make life tap a little less efficient when it comes to HP to mana ratio, putting some more strain on healers in group play.

Healers will have to heal warlocks for tapping than they used to. This is considered a bad thing. However, the thing that most warlocks-who-are-swearing-up-and-down-that-this-is-the-end-of-everything don't understand is that this has a negligible effect on most geared healers. At any level of gear where the warlock and the healer are at parity (about the same level of gear), a warlock will not be able to out-lifetap the healer's efficiency with Heal over Time spells without purposely trying to do so. A Rejuvenation or Renew spell will currently provide sufficient mana for a warlock to continue damage dealing for its entire duration, and then some.

Let's take an example destruction warlock with +1300 spell damage is spamming Shadow Bolt (420 mana, talented for 2.5 sec cast) for the duration of a Rejuvenation (12 sec, 4 ticks of ~800 hp healed per tick) from a healer with ~1800 +heal.

For those 12 seconds, you will get 4.8 shadow bolts off (2016 mana spent). If you perfectly tapped for free every single time (with no GCD whatsoever), you'd be healed for approximately 3200 hp in that amount of time, which is about 58% overheal (wasted healing).

With the new change, let's put the warlock at ~10500 HP and 9500 mana with buffs. With a life tap of 20% to 20%, you would spend 2100 health for 1900 mana, or ~1.1 health to mana ratio. Shooting those same 4.8 shadow bolts (still 2016 mana spent) would mean you'd need total healing of 2016 * 1.1 = 2217.6 health regained. You're still healed for 3200 hp over time by the same rejuvenation, which means that the overheal has only been reduced to 44%.

There are only two brands of warlock who fall outside of this model: ArenaVP warlocks, and tanking warlocks (in selected raid fights). The difference here is that ArenaVP warlocks have been covered already (issue #1), and tanking warlocks shouldn't care because most of the heals on them are overheals anyway.

Issue #3: It will make some warlocks have to tap more

So the healing required issue is mostly worthless at this point. Now you look at the other stat that affects group play: total mana regained per unit time.

Currently, our example warlock will have +1300 damage, which translates into a maximum rank life tap of 580 + (0.8 * damageTotal) = 1620. Thus, each lifetap will recover 1620 mana per 1.5 seconds (global cooldown, or GCD), or is roughly the equivalent of 5400 mana per 5 seconds. Given our sample grouped warlock statistics (10500 hp, 9500 mana), a 20% tap would mean 1900 mana per tap on GCD. This means the mana recovery goes up to 6333 mana per 5 seconds.

This is actually a buff to warlocks who don't care about the healing efficiency (which should be most of them), because it produces more mana per tap, which means they spend less time tapping and more time shooting.

The only time this would not be a buff is if warlocks somehow managed to stack their +damage gear massively independently of their other stats. A warlock with +1300 damage would require a mana pool of 8100 or less to get less mana per tap than they currently do. This sort of thing comes from stacking green items of Shadow Wrath, and would easily be solved by just getting better gear.

Issue #4: Effect on Soloing Warlocks

The final issue is that this will have an adverse effect on life tapping solo (leveling) warlocks. However, this isn't really as big an issue about this as it seems to be. Firstly, affliction warlocks won't be affected as much because they will still be using drain life primarily to tank their mobs and they will still be using Dark Pact to restore their mana instead of life tap. Demonology spec warlocks will likewise be less affected, because they mostly depend on the power of their pet while leveling. The only really negatively affected warlocks will be Destruction, but destruction was never a solo spec to begin with. All in all, you have the two solo specs mostly unaffected, and the third spec getting a bit more downtime when it was already bad to solo with from the start.

Honestly, after the entire analysis of the effects of the Great Life Tap Nerf of 08, it boggles my mind that people make such a big deal about it.

--Rawr