Pag on Games

The typical way to design game is to ramp up the difficulty progressively: the game starts easy and becomes progressively harder until the end of the game. But is that the best approach?

For some games it is. Some games test your skills and constantly push you to your limits to force you to get better. The games that gain the most from constantly increasing difficulty are those focused on the mastery of a simple skill. Guitar Hero would get boring if the level of challenge stayed the same throughout, for example.

Richer, denser games don’t need a constantly increasing difficulty nearly as much because what keeps the game fresh is the renewed content of the game, not the increased difficulty. In Mass Effect, for example, I found the last part of the game easier than many earlier parts — but that didn’t make it any less enjoyable. A cinematic game doesn’t need increasing difficulty to keep players interested, it’s the cinematic aspect that’s interesting — it can get more intense without being particularly more difficult.

It’s a bit counter-intuitive, but it seems casual games gain more from a constantly increasing difficulty level than typical hardcore games. Tetris has a much clearer increase in difficulty than Gears of War.

Richard Garfield is the game designers who created trading card games — he’s the mind behind Magic: the Gathering and many other traditional games. A few years ago he had a monthly column about game design in the magazine The Duelist. Ten of those articles are now online, and they’re worth the read. You can find them here.

2007 has been the year of casual games: Guitar Hero 3, Rock Band, Wii Sports, Peggle, etc. All casual titles that had a big impact on gaming last year. Another way of looking at them is that they’re all storyless — none of them have the strong narratives that some game developers told us were so essential to reaching mainstream audiences.

Is that a trend or is that just coincidence? Are games without stories making a comeback?

Xemu has posted detailed notes on many of DICE’s presentations:

• NPD
• Tools & Middleware
• Narrative Panel
• NanaOn-Sha
• Electronic Arts
• Character Design
• Massive Black
• ESA
• Activision
• Tetsuya Mizuguchi
• Shane Kim Q&A
• Blizzard

Some games are really good at keeping you hooked: you start playing and the next thing you know it’s 3am. How can you make games that have that kind of effect? A big part of it is making sure you don’t make the player turn off the game because he’s bored or frustrated, but there’s more to it than that.

Here are four tips to help you keep players hooked to your games:

• Give multiple objectives.  In a game like Civilizations you always have multiple goals at the same time: you’re trying to develop the city you just founded, you’re invading your neighbors and you’re developing new technologies relentlessly to be the first to get firearms. Because of that, every time you achieve an objective, you’re close to achieving another one too. So you keep playing until you reach that objective, but then you’re close to another goal, so you keep on playing — and so on until the small hours of the morning. By giving multiple simultaneous goals to players, you keep them constantly interested in reaching the next goal, so they keep playing.
• Eliminate Loading. Long loading time are a natural time to stop playing. You’ve just reached a new location, often reached a checkpoint, so it’s natural to stop playing at that point. The loading time also gives you the time to look at your watch and realize it’s time you did something else. If there’s no loading time, you don’t get that feeling of a pause so much: you’re instantly ready to keep on playing and you don’t have that time to think about the rest of your life that’s waiting for you.
• Change goals during levels rather than between them. That’s one of the things that was great about Deus Ex: you didn’t get your objectives at the start of levels, you got objectives in the middle of levels. You’d get in the middle of the secret base to talk with one guy, who would tell you to go find somebody else at some other place — but to reach that other guy you’d first have to leave the base you’re currently in. When you finally escaped that base and ended the level, you’d be halfway there to meeting the other guy so you might as well keep on playing… Giving objectives in the middle of levels is a simple way to get some of the advantages of the two previous points. First, you have two simultaneous goals: finish the current level and reach your official objective. Second, you diminish the problem with load times (if the game can’t avoid them) because it feels less natural to stop playing right in the middle of an objective.
• Restart seemlessly after failure. When you die in a game and have to replay the last bit of gameplay you failed in, you’d be excused to just stop playing and decide you’ll try again another day. Some games are more clever than that. In Grand Theft Auto, you don’t reload when you fail, you’re just immediately brought back to a nearby hospital or police station. Since you keep on playing, free to retry what your previous objective or do something else entirely, you’re less likely to just stop playing.

An experiment I’d like to try someday is to create a website to make a truly collaborative game design. The idea is to let the crowd of visitors to the site decide everything that’s going to be in the game.  At each step you first ask everyone to send in ideas and then have visitors vote on their favorite ones. The winner of the vote is kept and the design moves on to the next step.

For example, you would first ask visitors to vote for the genre of the game. Say “RPG” is selected, you then ask visitors to submit short “high concepts” for this RPG. Once you have enough high concepts, visitors can vote for their favorites. You keep the most popular concept and move on to the next step and so on. You could even do that with concept art, with visitors sending their drawings to be voted on as enemies, characters or environments.

I think it would be fascinating to see what comes out of a project like this. Since everything that go into the design is voted on by a big crowd, you automatically validate every concept — everything is focus grouped before it’s even designed in a sense. If you can get enough interest in this project, with a big community around it, you could probably show a publisher that there’s enough interest in the game to finance its actual development.

I don’t have the web development skills to make a website for this or the audience to get it started with bang, but if somebody has those things I’d love to help setting up a project like this!

Video games are the first form of art that lets the audience express itself — other forms of art are about the artist expressing herself. This is what makes games unique and so I strongly believe more efforts should be put into giving players ways to influence the games they play. Ideally, each player’s experience should be unique to that player.

Some people say that for games to be art, the creators must express themselves through their game. Some people claim that we must create games like movies, defining the experience precisely for the audience. I dislike those approaches because they deny the very thing that makes video games unique.

What is the most important goal of a game design doc? For me it’s getting the team to grok the game we’re building — to understand it at a fundamental level. A design doc with lots of details on everything won’t result in anything good if the team doesn’t grok the design, but a team who really groks the design will improve upon it and make a great game even if the design doc is lacking details.

What is the number one hurdle that prevents people from grokking your design? Not reading the design doc. Writing long, boring design documents makes people unwilling to read it, and unlikely to grok it.

That’s why I dislike long, extremely detailed design docs: people don’t read them. Design docs should be just complete enough that everyone groks what the game is about, no more, no less.

Last November I attended the Montreal International Games Summit, essentially a smaller version of the GDC. Among the many interesting talks, there was one called “How to Help Your Players Stop Saving All the Time” by Randy Smith, a designer working for EA LA. It was quite informative, going into great depth into the psychological impulse that makes players save their games often and how to limit that impulse so players don’t break their immersion by saving all the time.

I think Randy Smith missed the obvious: players should never save.

At least not manually. The act of manually saving — even if it’s just pressing F5 for a quick-save — breaks the fourth wall. It forces the player out of the game at least long enough for him to think “this is just a game, I’d better save to avoid any problems” — long enough to break the pace and tension of the situation.

Anyway, the designer knows much better than the player what’s coming ahead. I hate it when I forget to save for a while, then die because of some unexpected rise in difficulty, only to have to lose 15 minutes getting back to where I was. Since the designer knows what’s coming up, it makes sense if he’s the one to decide when to save instead of me.

I think automatic saving should be standard by now. It certainly has become more common place, but even major games like Mass Effect put checkpoints so far from each other that prudent players are forced out of the game regularly to save manually. Checkpoints should be common enough players don’t have to ever think about saving — even if they close the console at some random time during the level, the action should restart close to his last location.

Intelligent Artifice has a number of real game design documents that were used in actual production.