Double Buffered

A Programmer’s View of Game Design, Development, and Culture

Archive for March, 2008

Why Games Aren’t Creative Anymore

Posted by Ben Zeigler on March 31, 2008

“Think outside the box” is one of those catchy phrases that management-types like to throw around so much. The idea is that to come up with new and creative ideas, one needs to throw away all preconceptions and start with a blank slate. Unless you’re solving puzzles in a great game like Professor Layton (where it’s needed to solve probably 50% of the puzzles), this advice is basically worthless. It’s like handing someone a crayon and saying “draw something”. Unless that person already had something in mind, you’re probably going to get some incredibly boring scribbles. That’s not creativity. The human brain does not generate new, interesting ideas from some magical font of inspiration, they come from other ideas.

Whenever we do a brainstorm session at work for some game system or other, we always start out by listing some Principles. These principles are the Box from within we start our work. It’s easy to start filling out the obvious ideas that are implied by the principles, so we do that. Then come the variations and clarifications, where each person adds in ideas that are reflective of their own personal experiences and ideas. Random personal anecdotes from one person are combined with the game that another person played last night, and a crazy idea that combines them comes from a third person. Eventually, we realize that 2 of the principles we came in with were flawed, and we remove them. By the end of it we have a bunch of stuff that is exploding out the original box, and enough ideas to start making a game. This is creativity, and this is exactly how every great rock band or classical composer works. It can be a collaborative process between individuals, or one within a single person’s brain, but it works the same.

But there’s no guarantee that thinking within a box will ever generate new, creative ideas. Sometimes, it just results in incredibly focused iteration on a core concept, and never changes direction. In fact, smart people working on a problem will solve it in this way unless they can’t. The key is some sort of disruption to the focused iteration. In rock bands of 20th century, this disruption often seemed to be drugs. For the classical composers, I imagine it was often the whim of their patrons. In my example above, it was the interaction between disparate individuals in a group. And for much of the history of video and computer games, that disruption has been Technology.

Early arcade games were interesting because of the technical limitations, not in spite of them. Space Invaders is about aliens because planes were too hard to render, and the aliens speed up as you kill them because the game had to compute fewer aliens. Mario was red and had a mustache because it provided better visual contrast. The proliferation of character stats and levelling in PC RPGs compensates for the failure to bond with your player character through more direct means. Inability to represent motivations artistically led designers to add more complicated and interesting behavior to non-player characters. Striving against technical limitations drove all of the innovation in the 3D space and kept things interesting through the PS2 generation.

So why are games less creative and innovative these days? Well, we’ve basically destroyed all of the technical limitations, and are now in the “focused iteration” part of the creative process. There’s not enough disruption in the industry to create the real creativity needed to move us up to the cultural level of movies or music. Going forward I see two promising routes: Indie Games, and integrating ideas from outside the industry. Indie Games drive creativity in the exact same way as music and movies, by being limited in terms of resources. When you have very little money, you come up with creative solutions to problems. Also, there are many ideas from other media that are yet to be effectively integrated into the medium of games, and there is much fertile ground there. But, we can’t rely on the cycle of technical improvement to disrupt us from our boring but satisfying process of optimizing for the faithful.

Posted in Game Design | Tagged: , , | 6 Comments »

Piracy, Customers, and Making Money

Posted by Ben Zeigler on March 20, 2008

Stardock Software developed and published one of my favorite games of the last few years, Galactic Civilizations 2. If you’re not a fan of totally awesome turn-based space conquest games (and if you aren’t you should be), you may have heard about their stance on DRM. Specifically, they’re against it and ship all their games without it. Starforce (a leading DRM provider) decided it would be nice to encourage people to pirate Galactic Civilizations 2 in order to send a message. Classy that. Anyway, Brad Wardell (CEO of Stardock) recently posted a great essay on Piracy.You should go read it now, it offers a really interesting perspective.

As Brad outlines, the PC Gaming industry’s insane focus on anti-piracy comes from a fundamental misunderstanding of the marketplace. The basic business plan of PC gaming is two phase: Get as many people as possible to want to play your game, and then get as many of them as possible to pay for it. The concept of “Conversion Rate” in the casual gaming space is an example of this sort of process. DRM makes perfect sense in this context, because it limits the number of people who can play your game without paying for it. Surely the increase in sales totally justifies paying some shady software company or large conglomerate high rates for DRM protection.

If only that worked. Copy protection is always beaten, and fairly quickly. In the most important point in the essay, Brad argues that you should make games for people who will buy your game. The absolute number of people who play your game is important for developer ego and bragging to our relatives, but it doesn’t necessarily make you money. The solution Stardock has is to make games within a profitable genre. There may be fewer fans of turn-based space conquest games, but almost all of them are willing to pay for their games, as opposed to FPS or RTS fans. Stardock also makes an explicit effort to cater to the needs of their paying customers as opposed to potential users. This is the same reason that the subscription-based MMO model works. Our job in the MMO market is to serve our customer base and give them something valuable and unique for the money they give us. Our job is not to sell packaged goods to people who don’t need them. Some people will steal your game, but basically you’re better off just writing them off as a lost cause.

It’s not safe to totally ignore pirates, though. Pirates perform one important task relative to your game: they talk about it on internet forums and to their friends. Early adopters in the PC space are often pirates, and they can be effective for word of mouth advertising. There are other ways to get this kind of publicity (demos and free trials are just as good), but pissing off pirates will just make them angry and spiteful. Michael Fitch, head of the recently closed Iron Lore Entertainment, posted his own essay about piracy. It’s an interesting read and offers a counterpoint to Brad Wardell, because they basically totally disagree. One of the points Michael made struck me as absolutely insane though: the copy protection on Titan Quest caused random crashes on pirated copies and didn’t inform the pirates of why it crashed. Let me make this clear: This is the worst idea ever. Apparently pirates started talking about their crashes on forums and the game became known as unstable. Michael blames the pirates for this, but I’m going to have to say that I blame whoever mandated this decision (probably someone at the publisher) for killing their own word of mouth.

The last important point Brad makes is that all of their games purposefully target lower system requirements. World of Warcraft also does this very successfully, and I strongly believe that the days of targeting only high-end systems is now over. It turns out that people who buy expensive high end systems are early adopters who all know how to download cracked games from bittorrent sites. They’re also the kind of people who don’t tolerate the restrictions of DRM. And there aren’t that many of them. So if you’re targeting higher-end systems, you are targeting the small set of users who have those machines, have enough money left over to buy games, and who are willing to jump through hoops to buy instead of download your game.  This is totally obvious from a business standpoint, but it’s still hard to convince many in the game development community to buy in. The whole industry is still addicted to shiny new toys.

On a personal note, CD-based copy protection is basically the stupidest thing ever. You know what I do to every single game I purchase? I immediately download a no-CD crack and install it. I then return the CD to the packaging and never touch it again. Sometimes this fails, so I basically only download games from Steam these days.

In conclusion, trying to convert pirates into paying customers is basically always going to fail. Why would I buy and reinstall a game after I’ve tried it out using piracy? DRM is based on the theory of preventing people from being able to pirate in the first place, but this doesn’t work. Blaming pirates for the failure of your game is a waste of time, because the vast majority of them wouldn’t have bought your game anyway. The way to make money in the PC gaming industry is either to get as many initially paying customers as possible (by focusing on market segments with a lower rate of piracy and providing convenient download services like Steam), or by setting up a business model that is NOT about shipping packaged product. PC Games are not cans of Soup, and never will be.

UPDATE: I originally had a section about how it’s stupid that WoW has copy protection on it’s Cds but I appear to have confused my PC games. I should remember to check this stuff before blogging…

Posted in Game Development | Tagged: , , , , | 9 Comments »

Shacknews Gaming Quizzes

Posted by Ben Zeigler on March 16, 2008

ShackNews, which is a gaming news and community site, has been running a fun series of gaming quizzes. Once a week they put up an article that includes an interactive quiz, and you get graded each week. The most recent one gives you a bunch of concept art and you have to identify the game. This week’s quiz was fun but fairly easy, except for the spaceship one. I have no idea how you are supposed to identify concept art of space ships, because they all look exactly the same.

The first three quizzes are still up for taking, but you obviously won’t get credit for them. They deal with crate identification, RPG town music, and FPS sound effects. I can’t decide if I should be proud or horrified of my 4.17 GPA on the tests. Wait, nevermind, both.

Posted in Game Culture | Tagged: , , , | Comments Off on Shacknews Gaming Quizzes

Character Classes and Skills: Why They Suck

Posted by Ben Zeigler on March 13, 2008

“Classes vs. Skills” is a very old debate in the MMO and general RPG business. It gets trawled out every year or so (here’s a random slashdot post with a bunch of blog links from last year), and it’s always the same argument. I went to a GDC roundtable on the subject last year, and it boiled down to half the room really liking Everquest, and the other half really like Ultima Online. It was bizarre, anachronistic, and sorta worthless, which is why I’ve forgotten most of it. If you were there, I’m probably not specifically blaming you. Because there were a bunch of people there. And you’re probably a nice person.

The problem with the whole debate in my mind is that a Class-based system is a mishmash of unrelated concepts that is only unified due to historical reasons (D&D, mostly). Likewise, a Skill-based system always seems to imply a certain type of advancement as well as character customization, at least within the MMO field (I blame UO and Elder Scrolls). Why are the only choices for character differentiation these two random amalgamations (or more rarely some sort of vaguely linear hybrid between disparate concepts)? I blame MUDs for no particularly good reason.

So, what are the different components of a Character Class?

  1. Team Role: This is probably one of the most important ones. Are you a Tank? A Healer? A Mezzer? Knowing team role allows players to create Pick Up Groups in an efficient manner, and manage team tactics. I personally feel that it should always be easy to figure out a player’s preferred team role.
  2. Soloability: In general, classes are either optimized for soloing or for grouping. Good design can alleviate this, but some classes will always be better
  3. Combat Style: This refers to different ways of activating abilities, setting up combos, and the general micro-management of your character. Mages and Rogues both operate as damage-dealing classes in WoW, but have dramatically different play styles
  4. Non-Combat Abilities: For thematic reasons, rogues always open chests and steal things. Clerics decrease party down time. Warriors sit around and make racist jokes about kobolds.
  5. Advancement: Having a class always seems to imply discrete class levels and ability improvements that come in large chunks at arbitrary intervals.
  6. Character Theme: Paladins always save the innocent and aren’t allowed to make complicated moral decisions for fear of losing their combat abilities.

How does a Skill-based system mitigate the problem of grouping together these elements? It mostly makes them worse. Team role is lost completely, because there is no way to summarize team role and min-maxing for soloing often requires you to pick and choose only the “good” skills. Combat styles end up all being the same because when players are presented with overwhelming options, they just pick the easiest and most boring one (witness City of Heroes builds). Character advancement sucks because instead of being grouped into overly-large chunks, they’re delivered in overly-small chunks. Soloability sucks more because players are extremely tempted to fully optimize for groups or solo. Things aren’t tied together any more, but variety suffers because there are more but less interesting choices.

What I want out of a system is the ability to mix and match different discrete aspects of a character. I want 5-dimensional character development, instead of 2-dimensional (class + talent tree) or 25-dimensional (skills systems). CoH almost got there, but it was largely accidental and didn’t go far enough. “Defenders” had different team roles and play styles, but players still thought they were all healers because the archetype was too close to a traditional healer class.

I want a system where I pick my play style, my team role, and a set of out-of-combat skills. I want my character to be exactly as good at soloing as he is at grouping, and not feel like I’m gimped in either. I want to be able to quickly build a group of complementary team roles, and not worry about a player being gimped because they really wanted to use the psychic damage set. I want these to be organized into functional groups, so I can describe my character with fewer than 100 characters. I want my Raging Nuke (vs sustained damage) weapon-using guy who gets to agility-tank. I want a Melee Mage that has a Heal Other. Screw D&D and Ultima Online, lets make a system where every player makes fulfulling, interesting choices about their non-broken character.

Posted in MMO Design | Tagged: , , | 5 Comments »

SQL Considered Harmful: Presentation Slides

Posted by Ben Zeigler on March 7, 2008

As I mentioned last week, Shannon Posniewski from Cryptic gave a presentation at GDC about SQL Databases and MMO’s. Well, today the Official Presentation Slides were posted on our corporate site. Go check them out, they’re in PowerPoint-inside-zip form. Make sure to scour the presentation for important company secrets, such as our exciting variable naming conventions.

Oh, and here’s the official title of the presentation, preserved for awesomeness:

Current Relational DBMSs are Inappropriate for Low-Latency, Controlled Working-Set Databases
and what we did trying to circumvent that
and how that didn’t work out very well
and what we’re doing now

Posted in Game Development | Tagged: , , , , , | 3 Comments »