MW3 vs BF3, a lesson in Game Feel I guess

Eurogamer.net has an interesting article comparing (though they fail to use the term) the game feel of Modern Warfare 3 and Battlefield 3. The article brings a game quality that nobody talks about into context, discusses the rivalry between EA and Activision(-Blizzard), the future, and even some pc vs console discussion.

The Game Feel part of the discussion has to do with response times. Apparently Digital Foundry sat down and measured out the response time of MW3 and BF3 and found that MW3 responds about twice as fast as BF3, partially because MW3 runs at 60 FPS and BF3 runs at 30 FPS. I think their interpretation is that this response time is secretly addictive to players (even if they may not realize it) and draws them to MW3 like children to Chuck E Cheese's. I think they're on the tokens. In exchange though, Battlefield 3 implements a whole slew of rendering techniques that Modern Warfare can't handle in the time allotted, so we have a clear tradeoff.

The really interesting part to me though, is near the end where Digital Foundry points out that BF3 runs at 60FPS on a PC, that's not using the 5 year old technology that Modern Warfare is optimized for. They then claim that come next console generation, Activision is going to be at a technological disadvantage because of the engine Modern Warfare uses. EA will topple Activision, and the hobbits will begin taking on their real nemesis: Sauron.

The article is very interesting, but I wonder if the predictions are true. I think it's a little presumptuous to assume that Activision won't try to improve the engine at all for the next generation. Activision probably saved a pretty penny on development costs for MW3 so I don't see why they wouldn't invest in an upgraded engine for Xbox 720 or whatever. Even if they didn't, I'm sure Modern Warfare 4 would sell amazingly well on name alone.

Thoughts on Marvel Versus Capcom 3

So Ultimate Marvel Versus Capcom 3 releases next week. I think people are pretty excited about it despite a fairly long list of complaints. The biggest and probably most legitimate complaint is that it's too damn soon to be releasing a sequel. Marvel Versus Capcom 3 was released in February. This is easily one of the shortest life spans for a fighting game ever, and probably one of the smallest gaps between a game and it's sequel ever, which kind of sucks. I think my biggest problem with it relates to Malcom Gladwell's 10,000 hours.

In his book, Outliers, Malcom Gladwell says it takes about 10,000 hours of deliberate practice to become really good at anything. This includes Chess, playing violin, or making banana sculptures. In general it takes about 10 years for a normal human to accomplish this quantity of practice. The generation of fighting games before Street Fighter 4's reawakening of fighting games came out about 10 years before Street Fighter 4. Over those 10 years, players developed their skill and plumb the game's depths like ghetto scuba divers searching for treasure chests filled with crack. I'm a little bummed that MvC3's crack chests will not get the same chance that MvC2's had.

What I really wanted to talk about though was the design of "vanilla" Marvel Versus Capcom 3. This game has gotten a lot of hate in the fighting game community because old pros can get "random'd out" by scrubs, kind of like the tea party members in the republican presidential candidate race. But I think people have just failed to appreciate what a spectacle MvC3's design is. The game's goal was to capture the essence of what made MvC2 "hype", and they pulled it off with x-factor.

To give a little back ground on x-factor, you should probably watch this and this and this (that last one is a little long and stupid, sorry). MvC2 and the evolution of it splay is pretty interesting, and could probably be a post all it's own, but there are two important things to know to appreciate MvC3: 1) MvC2 was really really imbalanced. There were the 4 "god" characters, a bunch of pretty okay characters and a bunch of really bad characters. That's just a function of MvC2's production though, the goal was to have as many characters as possible, not really to have them all balanced. But this did lead to a lot of "cheap" stuff. 2) Comebacks like the ones in those videos happened more often than in any other fighting game I know of.

So MvC3 was designed to recapture what made MvC2 great while trying to make it more of a balanced game/something people who don't want to waste 10,000 hours of their life can appreciate. There are a bunch of characters again (with the additions from ultimate, it will be around 50), and this time almost all of them have techniques and abilities that are really really good. Rather than having 4 god characters and a legion of trash, the game has a bunch of really good characters, and like 5 bad characters.

The other big change is x-factor. I'm going to make a big statement here: I think x-factor makes Marvel Versus Capcom 3 the most innovative fighting game since, I dunno, Guilty Gear. X-factor is a one time use power up that increases damage and speed and can be used at almost anytime in a match. The buffs you receive and the duration of x-factor increase with each character you lose. The problem most people have with x-factor is that when you're on your last character, x-factor turns them into some kind of omnipresent samurai ninja master who will easily blow away the entire other team. Some people to say that the game doesn't really start til x-factor has been activated.

Those people are 100% correct.

But this isn't a bad thing. It's just really weird. People have spent a lot of time playing Marvel Versus Capcom 3 like a normal fighting game, when really it's much more strategic and mathematical. An apt simile is that most people treat MvC3 like a knife fight, so they get upset when their opponent pulls out a shotgun. But I think that's the wrong approach. You need to be waiting for the shotgun, and waiting for it should be part of your strategy at all points in the match.

If the game had gotten the 10 years it deserved, I think we would've figured all that out. Alas UMvC3 is nerfing x-factor, and pushing the game to be more 'traditional'. Bummer.

Retro Graphics Leave more to the Imagination

I've been playing through the DS remake of Chrono Trigger lately. I hadn't played CT in at least 5 years (probably more like 10), and it's really interesting to go back and check it out now that I'm not a kid.

I think the biggest thing that stands out to me is how many visual tricks the future Squenix use to manipulate the player's imagination to fill in the blanks of what's happening on screen. For example: At a late point in the game, a character is about to take off in Chrono Trigger's version of the airship from a docking bay (inside another airship, which is kind of like an aircraft carrier in the sky). The character is sitting in the cockpit, the ship is sitting on metal planks with metal walls (you know, a docking bay) and prepares to launch. Then the docking bay opens, how do I know it opens? The game applies a simple bright grey semi-transparent layer in a square over the bottom half of the scene. In the players mind, this immediately demonstrates the doors opening and light shining in.

Even when you understand the tricks Squenix is using, it's still a great effect. I think because it encourages your imagination to recreate the scene in your mind (probably by mixing images from movies or 3d games). This effect is one of the secret appeals of retro games which nobody talks about. Modern 3d games have so much painstakingly thorough detail that you admire it and then just kind of ignore it, like your friend who always "gets sick" whenever nobody's paying attention to her.

I'm not sure what it is exactly. Perhaps we just naturally interpret images in our heads, and the more abstract/indistinct it is the more striking and imaginative it is once we've got it in our heads. Think of it like looking at clouds. Whenever I see a cloud that looks like a baby riding a crocodile (for example), it always sticks with me more than a banner for party radio attached to a plane.

To be honest, when MegaMan 9 came out I began desperately hoping that more AAA game companies (at least Japanese ones) would follow and begin pumping out retro games again. It seemed like it would make fiscal sense. Modern game programmers can probably churn out all the code for a 2d platformer in a month, the art takes maybe two months (done in parallel of course) and then you have a whole 10 months left for a standard production cycle to tune/polish the hell out of the game. Realistically this means you would probably spit out a new game every 3-4 months, and they would be goldmines.

Alas, MegaMan 10 didn't sell so well, and since then I don't think there have been any similar projects. Bummer.

Someone who gets it

I know nobody reads this blog, but I just want to make a note of this amazing piece reprinted on kotaku, written by Joshua Wise.

One word: Bravo.

Thoughts on War in Video Games

Kotaku posted a short, sweet, but slightly useless editorial today. Useless mainly because the point of the article is a quick, delicious burst in the middle surrounded by more dancing than Step Up 2: The Streets. The article, written by a PhD on the eve of Modern Warfare 3 (sequel to the fastest selling game of all time), is lamenting the fact that most war games (or war in any media really!) reduces the horrors of war, the complexity of its consequences, and the politics which surround it to a bunch of guys playing paintball in a corridor.

To be fair, I would say that Modern Warfare 2 kind of approached this problem but it looks like Modern Warfare 3 is just going to recycle the same trick. I agree with the PhD though, war games barely touch the real consequences of war.

But this is really a criticism of games in general. As it stands, the consequences of your actions in any game are completely limited by the code the team was able to piece together by launch, which usually means consequences are limited to "it's easier to kill X". It's very hard/expensive to make a game that can even remotely model geopolitical interactions, environmental shifts, or technological innovation unless it's the full focus of the game. I think the best we've seen is a few RPGs (Persona 3, Persona 4) which manage to track the player character's social life and consequences in a relatively realistic way. It'd be nice if we had an action game where your actions had huge effects on the game's world.

I remember back in the day I really wanted to make an action game where you played a mercenary and you would take contracts for various battles in a fictional continent's history. The game would have unbelievable replayability because you could dramatically change the course of history. It was pretty baller, then Atlus published Spectral Force and I got the feeling the idea wasn't so great. But maybe that game just sucked (it WAS another generic anime tactical RPG like all of the trash NIS released after Disgaea, rather than a decent Dynasty Warriors or amazing Devil May Cry experience).

Anyway! I think what Dr. Huntemann really wants is what hilarious Faux-Game Designer Peter Molydeux once quipped: A system where you kill a soldier in Modern Warefare 3, and then see his son crying in the crowd when you boot up FIFA. But given that AAA game companies are still in the process of hiring one man to design hands, I don't think those kinds of reactive systems are going to get the Modern Warfare treatment for awhile.

On Board Games vs Video Games

I had a pretty interesting thought recently on the big differences between board games and video games.

Besides the mainstream stuff like Monopoly or Chutes and Ladders, my real introduction to board games came when I met some folks in college who showed me the meat and potatoes of board game geeks. Settlers of Catan, Carcassone, Dominion and the like. Initially I felt board games were substantially more strategic than video games, which was refreshing like a mint julep in a sauna. They also made the necessity of Schell's 4 Pillars of game design (my term) more obvious (particularly the pillar of "technology"), so analyzing them was a good exercise. But playing board games felt different from playing video games, and trying to figure out why was like trying to bob for apples in a shark tank, impossible using traditional techniques.

I believe now, that the difference is a matter of player approach. I'll put it this way: Board games encourage modeling, video games encourage discovery.

What do I mean by modeling? Making models, ala Mouse Trap? No. I mean mental models. Board games exist almost entirely in the players heads. I will argue that a box of Monopoly pieces is really just a set of physical mnemonics and a random number generator. You open the box and you begin to play using an agreed upon rule set, which you can modify on the spot (with bullshit rules like taxes go to Free Parking, for example). If someone makes an illegal move, someone has to notice that it doesn't fit the mental model that the players are supposed to share, and prove it with the rulebook if necessary (which actually adds a fun meta game to an already meta genre). Board games are to rules what Kinect is to video games. There's virtually no feedback from the game itself about any of the actions you take. Board games are the drunken father who let's his kids play with shotguns because it shuts them up for five minutes. I think this is also why board games seem more strategically focused. You're forced to absorb a set of logical premises for manipulation, which are more pointed and discrete to allow stronger mental modeling.

Compare this with video games: Video games I would say are discovery driven because only in the most simplistic of video games can the rules be conveyed with an adequate level of precision in less than 30 pages. Because you have a computer acting as ultimate arbiter of the model of the game (replete with large fluffy beard), you don't even have to bother memorizing the rules, and better yet, because of inherent flaws in computer software design, the game's creators might not even know all of the rules. They know the rules they wanted, but only the code knows (makes?) the real rules of the game. Furthermore then Bearded Judge allows players to "feel out" a solution to their problems by supplying feedback for whatever gameplay experiment they engage in. DISCOVERY.

Thinking about games this way does ignore the more tactile sense of video games, and the freedom inherent in board games (and incidentally, all non video games) which is another important distinction. Sometimes a video game is more than just a puzzle. Tricking the proprioceptive sense and putting the player in the game.If a board game is boring, you can easily change it to make it more exciting for your particular social blend. These are important differences if I were to host a real debate on board games versus video games. But I think this discussion of approach is interesting.

How Ghost Trick Solved the Adventure Game

A few months ago I finally managed to get my hands on Ghost Trick, which sells itself as an investigative game in the same vein as Pheonix Wright(with which it shares a creator). This isn't really a game review but I will say that ghost trick was certainly worth the money (granted I only paid $20 for it). The dialogue is pretty well translated, the story is really interesting and wonderfully convoluted (I guess that's something most people don't like though) and as far as I can tell, is logically consistent from start to finish. The art is great (except the horrendous box art), the music is great (even if it could use a couple more tracks) and possessing things feels great.

So the real meat of this post is why Ghost Trick is the secret genius of the DS-driven reawakening of the Adventure genre.

I didn't really grow up with adventure games. I was almost exclusively a console gamer, and stuck to action games. My first real adventure game was Myst when I was a wee lad but Pheonix Wright was when I really discovered the genre and began pursuing it. For the sake of narrative though, I'm going to use Pheonix Wright: Ace Attorney as a placeholder. I love this game to death. Because the actions available to the player are so limited, the game is able to put the player in the middle of several fierce courtroom dramas/verbal sword fights with a great backstory to lead up to them. The role of the player is to watch the story unfold, and sniff out contradictions (a whiff of which will send Pheonix into a crack-like frenzy). There aren't too many games that make you feel as smart as PW:AA does when you know what's going on.

But therein lies the problem. When the critical details aren't clear, you spend a maddening amount of time trying to figure out why the gun which obviously implicates the parrot isn't stopping the curry shop owner from talking (spoilers!(the spoiler is that that is not a spoiler)). Though I'm ashamed to admit it, I brute forced my way through more than one cross examination using the overly generous/easily exploitable quick-save feature. This is a common issue in almost all adventure games and completely destroys any immersion that's been established.

So now let's talk about Ghost Trick. Ghost Trick pretends to be about investigations (seeming very Pheonix Wright-ish), but it's a big lie. Ghost Trick is about haunting shit and saving lives. As a ghost, your power is to turn back the clock a few minutes and then try to possess things to prevent someone from dying. You push bowling balls off of shelves and turn on air conditioning units to move flags around to change the circumstances. Each of the game's levels is basically a small physics puzzle, with one amazing twist: There's no physics engine.

See the problem with Adventure games is that the conditions for progress are hyper specific despite a facade of choice and freedom. Hidden behind the veil of a realistic world with realistic characters and consequences is a set of finite and ultimately arbitrary switches. When you feel you have an answer that fits the situation, you go crazy because your set of level pulls doesn't match the one in the secret diagram. Games with physics engines have the opposite "problem". They can't cover every possibility because the situations are almost never exactly the same.

But that's where Ghost Trick triumphs! Whenever you make a guess at the answer to the physics puzzle, the reason it won't work is immediately clear. The ball doesn't bounce that high, the board falls too far, etc. When you restart a level you go back in time. Logically, every particle in the room is exactly the same as when you first saw it. Thus, the switches under the hood are reset just like the world whose representation you're interacting with.

I suppose the genius of it is presenting the fixed (digital, if you will) operations of the game space as physics (or analog) operations, then allowing the player to "rewind time", and close up any consistency issues that show up. This means that the player won't feel that something arbitrary and solvable outside the illusion of game space is preventing them from proceeding. It's an amazing solution to an old problem of gaming

I suppose it's hard to copy without ripping Ghost Trick off, but let's all at least give it the round of applause it deserves. Lord knows nobody played it.