Geometry Wars vs. Assassin’s Creed: What do People REALLY Want for Next-Gen Gameplay?
Matt posted in Games, Playstation 3, The Industry, Xbox 360 on May 4th, 2006

 I love Geometry Wars. It seems everyone does, and for good reason: it’s really fun and looks very cool. Many people say that graphics don’t add to fun, but I think Geometry Wars seems to be a good counterpoint to that. I’d much rather play Geometry Wars: Retro Evolved than the original, even though it sounds like the original had more playtesting. (Stephen Cakebread, the developer, said that on the original, other employees at Bizarre Creations played it, but they weren’t allowed to on Retro Evolved because the first one ate up more time). I’m sure Geometry Wars: Retro Evolved is outselling Smash TV on Xbox Live Arcade even though the gameplay is quite similar. What is it that makes Geometry Wars so appealing? For one thing, it is a joy to look at. The first thing people say about it is how awesome it looks. Good graphics are appealing. It’s easy to try and say that “gameplay is the only thing that matters”, but most people are drawn to better graphics. Why is Panzer Dragoon Orta a better game than Panzer Dragoon? I love them both to death, but honestly despite all my Panzer Dragoon love, I really can’t tell what the heck the enemies are supposed to be in the Saturn incarnation. I’m not trying to say that graphics are the most important thing, just that they can increase your enjoyment of the experience. This is because good graphics can increase the immersiveness of the experience. It’s easier to get lost in a world that looks more realistic. It’s distracting when you play Panzer Dragoon and wonder what on earth the clump of 20 polygons is supposed to represent inside the Panzer Dragoon universe. In the end, it doesn’t matter though because the Panzer Dragoon universe sucks you in anyway. The post apocalyptic world struggling to regain control over the technology of the past that destroyed it is a very engrossing storyline. Fusing the fantasy element of a dragon with the sci-fi genetic engineering and combining that with a primitive society that rose from the ashes of the apocalypse of the previous generation is a fascinating world that I love to be immersed in. It’s quite possible that I’m the biggest Panzer Dragoon fan in the world and few people really appreciate the series, but maybe it’s just that few people were willing to look past the low polycount graphics to see the world behind it – the world the polygons represent.

With next gen, this seems to be solved. Now, you can have the high quality art that the can more accurately represent the world that game developers imagine. Now you can engulf the gamers in the universe of your imagination. Game development is becoming a powerful tool. You can do more than just entertain – you can give someone an experience. But, what problems are there with next gen games? Well, let’s look at Geometry Wars. It’s called “Retro Evolved” for a reason – it’s a step back to the olden days where games were simple – where anyone could pick up a joystick and play. This concept seems somewhat shunned. Games are becoming increasingly complex. There is a large barrier to entry for those who haven’t been keeping up. Nintendo is trying to buck the trend with the poorly named Wii, but they are potentially introducing confusing controlling schemes that vary for every game since the controller has a bunch of unique functionality. Only time will tell whether they will be successful at capturing the audience that complexity left behind.

Many people view next-gen as modeling reality, but if we view Geometry Wars: Retro Evolved as a success, maybe reality isn’t necessarily the direction we should be going. The best example of this is the GDC talk about Assassin’s Creed developed by UbiSoft Montreal. (If you are unfamiliar with this talk, read Gamasutra’s coverage. It’s quite good and will get you up to speed) With Assassin’s Creed, the creative director, Patrice Desilets, mantra is: “Why do we create rules for our characters instead of using the ones we have in the real world?” He goes on to describe that the player has a range of moves and the player has the freedom to use them at any point. It seems like they are attempting to apply the idea of “emergent gameplay” to the platforming genre.

 Firstly, let me explain why I think using the rules in the real world is a bad idea. The real world is pretty dull. Not only is the realm of what the human body can do physically restricting to game designers, it’s also not as exciting. Can you imagine playing a game where the player character can only jump a few feet forward? If you use actual gravity for your player character’s jump move, it will seem very floaty. If you don’t let the player change his path in the jump, it’ll be very difficult for the player to hit his/her target. If the game automatically snaps to what it thinks your target is, then not only will it be frustrating when it snaps you to something wrong, but it’ll remove all the challenge. All of a sudden, you are playing an old FMV game since all you have to do is press the button close to a ledge, and you’ve flawlessly and effortlessly reached your destination.

Now, as both a game developer and a gamer, I have mixed feelings about all the freedom they are promising. As a game developer, the more freedom you give a player, the harder it is to balance the game. Warren Spector talked at my alma mater (University of Michigan) about what a nightmare it was to test Deus Ex and all the interesting things people found they could do. It was possible to stand on mines or something like that, and I think there was the potential to jump out of the world. It’s always bad if gamers can find ways to cheat and skip portions of the game or mess up the state of the game, and the more freedom you give them, the more tools they have to do this. But, in a free world, no longer to gamers have to hone their skills of guessing what the designer was thinking like in old adventure games. So, I’m certainly not arguing that freedom is always bad.

I like the fact that we are giving gamers freedom, but as a gamer, I’m scared of the actual execution. Maybe people expect the opposite reactions. As a game developer, giving people freedom is always scary because as described earlier, they can break the game. But, as a gamer it’s scary too. For example, let’s look at one of my very favorite games, Sucker Punch’s Sly 2: Band of Thieves. Sly 2 has a very similar style of gameplay to Assassin’s Guild. But, in Sly 2, they took the opposite approach. There is pretty much only one way to solve the platforming puzzles, but they make it obvious what move you have to use. Although the sparkles sometimes ruined the immersiveness of the game, it was VERY CLEAR what movies you could do where, and that worked to their advantage because there was no frustrating guesswork. The scenario I’m imagining is this: I’m playing Assassin’s Creed, and I get to a point where when the designers played it, ran across the wall, jumped off it, swung on a pole above and caught the ledge. That was how they solved the particular puzzle. I’m sure some testers found plenty of “emergent” ways to solve it, but testers are quite good at figuring out all sorts of mind-boggling solutions to play through a game (oftentimes exploiting bugs). When I play it, I try to do a wall jump off of the wall to land on the top of the pole. I try that over and over again because I’m convinced I can make it this way. After failing many times, I try to think about what other moves I can use in this situation. I completely forgot about the wall run because it’s been so long since I’ve used it. I manage to make it to the pole 1 out of every 10 times, but the ledge is really small (10 cm, the minimum length of a ledge to grab as mentioned at GDC), so I don’t realize that I can grab on to it. (On a standard resolution TV, that may be hard to distinguish from normal mapping from some camera angles) Instead I go for a different higher platform, but I can’t make it. Eventually, I get frustrated and stop playing or look at a walkthrough.

What went wrong? Didn’t they adopt the next-gen philosophy of creating an emergent world that featured multiple potential solutions to the problem? Well, sort of. It is very difficult to create a setup that features multiple solutions of equivalent difficulty to the same problem, especially for platforming. Although there was some other combination that could have solved the problems, all the approaches I took were doomed to failure because I had trouble recognizing what solutions would work within the constraints of a system. Sure, you can argue with enough playtesting you can figure out all the different ways that people are going to approach it. But, normally when faced with a problem in a playtest where people have trouble doing something, the easiest (which is often the best) thing to do is to change the design in order to make the “correct” solution more obvious. That then defeats the purpose of having all these moves which can be used anywhere. Honestly, as a gamer, I’m leaning towards a more controlled but tuned experience that Sly 2 provides. That’s really the tradeoff: the more control you give the player, the less you can tune it because it has to support all the different ways a player can try to solve it. In the end, does the player gain? You always run into balancing problems when you have too many ways to solve a problem. In this example, if I want to support 3 different solutions to the platforming puzzle, I have to make sure that all will work. The problem is that moving the platforms around until they work for all 3 solutions will probably make one solution much easier, thus removing the challenge of the game. If I’m making 3 solutions work, does that take the emergent aspect of the game away? Do I want to have no correct solutions – just situations and with enough play testing hope that every player can work out their solution through it? If that’s the case, how do you design it at all?
In my experience, emergent systems are fun – for a while. I remember getting a thrill out of just running around buildings and jumping on stuff in Sly 2 – for about 20 minutes. Then, I stuck to solving the actual puzzles in the game, all with a pretty obvious “correct” answer. I played Fable all the way through to the end. I was going to have some fun with the emergent systems. I wanted to kill everyone in every town, buy their houses, rent them out, and become the ultimate slumlord of Albion. The only problem? I got bored after a few minutes. I took over one city and didn’t bother to branch out my murderous rampage. Maybe I just don’t appreciate emergent systems like other people do. Many people had fun just driving around and crashing in GTA. I’ve steered clear of the GTA games because the content sickens me, but even still, I think one of the real reasons that GTA‘s emergent gameplay is more appealing is because it’s based in the real world. It allows the player to live vicariously chainsawing cops, and renting time with women of ill repute, only to murder them to procure a refund. The Sims is a terrific example of the success of an emergent system. But, the reason The Sims is such a success is because people who aren’t ordinarily gamers buy it. Maybe emergent games allow you to get – not gamers – but the true mass market. That’s what The Sims did, and that’s where the real money is. Certainly though, neither Geometry Wars: Retro Evolved or Assassin’s Creed is targeting that audience though.

 Let’s look at another example, Shadow of the Colossus. Unfortunately, I haven’t had much time to play it, but from what I’ve played, it seems like there’s a pretty specific “correct” solution to defeating each colossus. They aren’t going for any sort of emergence. One who argues that Shadow of the Colossus is NOT one of the high points of the Playstation 2, which has seen more than its share of outstanding titles (including Sly 2), cannot be serious or didn’t play enough of it to get sucked into the wonderful world to which it exposes us.

Maybe I’m wrong. Maybe Geometry Wars: Retro Evolved is a fluke. Maybe it isn’t a sign that people want next gen graphics with simple gameplay. But, looking at what people are trying to accomplish, I’m worried that people will think more power means more complicated gameplay. When I see more complicated gameplay, I find it a bit of a turn off. I’m not saying they can’t be successful, or that I won’t play them. But, for example, I paid $200 for Steel Battalion, then read through a 40 page manual since there was no in game training missions, and when I started playing, I ran out of in game batteries even though I found no mention of batteries in the manual. (The mech runs on gas and is a manual, so you have to be careful about your fuel consumption.) Steel Battalion is trying to model realism with its complex system, but when my in game avatar died, and it deleted my saved game, that was a big turn off for me. I’m not trying to say that Assassin’s Creed will be a bad game. Maybe it’ll turn out terrific that I will regret my lack of foresight in publishing this article. I’m really looking forward to playing it at E3 next week. But regardless, I wonder if there’s some insight to be gained from the almost universal praise of Geometry Wars: Retro Evolved.