Friday, December 12, 2003

Churning Addendum


New Manager In A Strange Land. I don't remember it being that short.

Something we've gone back and forth a couple times on the team is how strict we should be with checkins AKA commits. Much like a society swings back and forth between law and anarchy, the team swings back and forth between a strict checkin policy and a lax one. With the lax policy, everybody complains about the build being broken. With a strict policy, everybody complains about the time they have to spend validating their data, and how they can't check in because the build has been locked.

Our current policy is on the strict side. FWIW:
- when the build is broken we lock the depot - only people who are attempting to fix the build can commit
- the depot is locked an hour before the end of core hours, so people cannot "commit and leave"
- people who break the build and don't manage to fix it immediately are put on the blacklist; for the next week, they can only commit if they have permission
- a few special people are on the "awesome" list - they can commit at any time for any reason.

Despite the strict policy the build is frequently broken on at least one of the consoles for long periods of time, but it hasn't been in effect that long. We've also upgraded our build machines so they can churn faster - it could take an hour for the machine to detect a break, and another hour before the machine would validate the fix.

Sunday, December 07, 2003

Notes on Prince of Persia


Okay, I finally got enough time off of work to finish it, and now that it's over, I am inconsolably depressed. I tried playing THUG and Call of Duty, but nothing fills the hole that Prince of Persia has left. To me, these other highly acclaimed and polished games seem like stick figure illustrations compared to the glorious masterpiece that is Prince of Persia. I wish I had been on that team, just to see how they work, because they've got it down.

Now, there are a number of things that are considered "good game design" that are missing here, such as: meaningful choices; emergence; nonlinear narrative. In fact, the whole game is an illusion: it tricks you into thinking that you're clever, by providing puzzles that are just hard enough to not be obvious but easy enough that almost anyone can solve them. It tricks you into thinking that you're a badass, by providing exhilirating combat moves and death defying acrobatics. While playing it, I felt like, "I am the greatest." Never mind that when my non-gamer wife played it, she was almost as great as me. I was suckered in by the illusion.

This is what Mark Barrett calls magicianship. (At least I think it was Mark Barrett.) And it shows that the Sid Meier definition of game doesn't suffice, whereas the Rollings / Adams definition -- a sequence of challenges -- still does. Like Zelda, it's another example of how simple puzzles really can be satisfying.

I'm almost tempted, after playing Prince of Persia, to say that the great, erudite game designers are just wrong, and that puzzles are better than problems, and a single highly polished, highly crafted linear game is better than a not-so-polished nonlinear one. I would almost argue that we should stop thinking of ourselves as game makers but instead think of ourselves as makers of virtual funhouse rides.

But that's silly. Different games appeal to different people. I seem to be more easily taken in by the funhouse ride than the actual meaningful game experience, but plenty of other people are not fooled and prefer the game that they can truly get involved in on a deep level. All I can argue for is that you pick an audience and try to make that one audience happy, which Prince of Persia does.

Another game design rule, one of Noah Falstein's, is "provide parallel challenges with mutual assistance." This one Prince of Persia also breaks. Playing it, I realized that the rule is just a means, not an end. The rule should be: "Don't let your player get stuck." Don't have shelf level events. One way to do that, is with the parallel challenges technique. Another way is to playtest the crap out of your game and make sure that most of your playtesters can make it all the way through without consulting gamefaqs. Prince of Persia does this well. The time-rewinding feature, obviously, is a great mitigator of frustration. Also, at each save checkpoint, you have a vision of the future, where they show you the solution to the next room. (Although I loved the idea that these two 'cheats' were both tied in to the time-travel motif, I started skipping the visions, because they were spoiling it for me. I sometimes went back to them if I got stuck. I never had to consult gamefaqs.)

Let's talk about the wall run. To me, it's the wall run and not the rewind that is the heart of Prince of Persia, because it's the wall run that turns the 3d game into a 2d one. One big problem with 3d platformers is simply aiming your character at the next platform. The arcing wall run allows Prince of Persia to take the old two-dimensional platformer gameplay and wrap it around the walls of a three-dimensional space, and thus, like Viewtiful Joe is able to bring tried-and-true two dimensional gameplay from the old world into the new.

People may say, when Spider-Man 2 comes out, that we stole our wall run from Prince of Persia. We didn't. We stole it from Shinobi.

Speaking of stealing, Prince of Persia steals back everything that Tomb Raider and Ico
and Soul Reaver stole from it, and then some. All the elements from Prince of Persia are here, but we also have Lara's three-dimensional block crawling, Ico's female companion, and Soul Reaver's vampire stake. And Zelda's reflecting mirrors. Was it Picasso who said he didn't borrow, he stole?

One thing that bothers me: the ledges and poles that you jump from and to are at different relative locations when you encounter different challenges, and the Prince always makes the jump. On the one hand, this is wonderful: if I think the Prince can probably make a jump, he probably can, and when I tell him to do it, he does it succesfully. It's like the game is figurative, where other games are literal. The A button means "jump over the chasm" in Prince of Persia, whereas in Mario it means "jump three feet." It's actually hard to make him screw up. On the other hand...this is yet another factor that makes Prince of Persia
feel less like a game and more like a funhouse ride, as the skills of precision and pattern recognition (is that a jump I can make?) are not really required here.

A lot of reviewers have been dissing on the combat. Maybe this is because the combat is an actual game, and they don't feel like it has any place in their funhouse ride. Contrast with Ico, where the combat was simple button-mashing. Maybe I'm experiencing halo effect, here, but I even liked the combat. I liked it a lot. I'll even say it is the best beat-em-up style combat I have seen in a 3d game with a free-floating camera. (Devil May Cry and Viewtiful Joe are better, but they're fixed-cam. And they're Capcom. Who can compete?) The combat has a number of orthogonal elements that create meaningful choices: attack, tortoiseshell block, dodge, vault-over-opponent, a vampire-stake move that makes sure opponents stay dead, a wall jump attack (that I almost never used), a one-hit-kill counter-attack, the ability to rewind time (of course), the ability to slow down time (which I almost never used in combat but did use on some of the rotating-blade timing puzzles), and some special attacks. There are three resources: life, number of time rewinds you get, and--let's call it 'magic'--which lets you use special attacks, etc. Like you manage your lives, inviso, and smart bombs in Stargate
you manage these three resources in combat. When you kill an opponent, it gives you one more rewind and a little more 'magic', so you're not overly encouraged to hoard.

Many complain that, when fighting, you are assailed by too many enemies. I'd argue that the resource management aspect of combat only comes into play because you're assailed by so many. It takes a while for them to grind you down, and for you to start wishing you'd used a rewind to save some life, or wishing you still had a rewind left.

(BTW, as much as I liked the combat, I think we can take 'em. Time will tell.)

Prince of Persia is, in many ways, GTA's opposite. GTA has freedom; Prince of Persia has none. GTA is outdoors; Prince of Persia is indoors. GTA has a branching storyline; Prince of Persia is linear. GTA you shoot; Prince of Persia you slash. GTA has a lot of real-estate at low detail, with low production values, and bad animation. You get the idea. I hope Prince of Persia didn't differentiate itself right out of the market. They've been on the Yahoo buzz index for seven days now...that's something. Especially since it's the kind of game you don't need to look up on gamefaqs.