[In this article, Gamasutra presents excerpts from Jordan Mechner's The Making of Prince of Persia -- a new e-book which compiles his journals from the late 1980s development of the seminal game -- with a newly-conducted interview about his experiences.]
In 1986 Jordan Mechner began work on the game that would define his career, Prince of Persia, on the Apple II. Over the course of three year period of development, Mechner would partner with Broderbund Software, move to California to work on the game, and radically rethink and retool the game.
These extracts of his new e-book, The Making of Prince of Persia, clearly show how the game took shape over the course of its lengthy development process. To compliment these selections, hand-picked for Gamasutra by Mechner himself, Gamasutra is happy to present a brand new interview about his experiences.
Of the book, Mechner tells Gamasutra, "In a way, I feel that that gives a lot truer picture" of the development of Prince of Persia than a newly-written book would.
"If I were to write my memoir of Prince of Persia now, with hindsight, I would put everything in perspective, give everything the appropriate weight. But since these journals are what I actually wrote on that day, at that moment, there’s a lot of ping-ponging back and forth. One day this is going to be the greatest game of all time, and then two days later I don’t know if I should even finish it."
Notably, Mechner speaks and writes about those who helped shape the game, which he developed himself -- but not without assistance. In particular, he credits his friend and collaborator Tomi Pierce with many important insights. You can read more about Pierce on Mechner's blog.
Excerpt One: The Move to California
September 18, 1986
Looked at a house in Mill Valley, on a shady road winding through the redwoods. When I rang the doorbell the lady peered around me and said, "Is your mother down there?"
She spent 15 minutes showing me the house, but I don't think I ever quite convinced her I was serious.
September 23, 1986
Spent much of today working on the logistical problem of how to get the footage from a VHS tape into the computer. I finally (tentatively) settled on photographing the frames one by one with a regular 35mm camera, getting prints made, then (after retouching as needed) digitizing the prints with a regular Sony video camera. It sounds like a pain but I think it's the best way.
September 25, 1986
Another solid workday. Today I stayed till around 7 and got [homemade animation and drawing tool] DRAY pretty much finished. I tested it out by digitizing a page out of Muybridge. It'll do what I need it to do. It could use another day of work. Actually, I could keep working on it for a month, if I didn't have so much else to do.
September 26, 1986
[Director of product development] Ed Bernstein called his last P.D. meeting this afternoon. He's leaving to head up Broderbund's fledgling board games division. DOUG HIMSELF [Co-founder Dug Carlston] will be taking over as acting head of P.D. He'll be taking my desk, the better to stay in touch with the people. So I'll be moving into Ed's office. Life is strange.
P.D. is throwing Ed a goodbye party. "Better the devil we know than the deep blue sea," Steve said.
At lunch, Doug said: "You seem to have a very strong entrepreneurial bent." I was surprised, and said something about how I'd probably inherited it from my father.
Coming out here was definitely the right thing to do. In Chappaqua, I was in a rut. Now, I'm in the thick of it. It's great.
September 27, 1986
I have a car.
September 28, 1986
I have an apartment.
September 29, 1986
Today I moved into Ed's office. Obviously, this is a temporary arrangement; eventually some new guy will be hired to run P.D. and I'll get booted to some other part of the building. But while it lasts, it's great.
Besides vast amounts of space, a couple of armchairs for visitors, my own phone, and a door that I can close, the office has the most important thing of all -- equipment. A printer. An amber screen. An Apple IIc. It didn't occur to me until I was actually confronted with two Apple IIs on my desk and I had to figure out what to do with the extra one -- but it's perfect. Now I can run programs without destroying the source code in memory. It's... (gulp)... a development system.
October 14, 1986
David Stenn read my screenplay. He said it has promise but would need at least one more rewrite to be saleable. Perhaps sensing my disappointment, he said: "Look, it's great for a first script -- it really is. I wouldn't show you my first screenplay. You obviously have talent, you should stick with it."
He was more impressed with the reviews of Karateka I'd sent him. "You're in the right business," he said. "What do you want to get into this one for?"
October 15, 1986
Bought a camera at Whole Earth. It was more expensive than I'd anticipated -- $250 with the lens -- but it's a good camera, and I imagine I'll find some use for it even after the game's done.
I shot my first roll of film (David turning around) and had it developed at the local one-hour photo stop. I think this will work. The real problem, obviously, will be going from a sheaf of snapshots to the 280 x 192 Apple screen, and the loss of accuracy entailed therein. It almost makes me want to do it in double hi-res.
October 19, 1986
Shot four more rolls of film: David running and jumping in the Reader's Digest parking lot. One year ago tomorrow. Red and orange leaves... God, I'm homesick.
October 21, 1986
Today I wrote the first lines of code of the game (not counting the hi-res routines). It Begins.
October 23, 1986
Everyone in the office has been playing a lot of Tetris -- a Russian submission for the IBM PC. It's a classic, like Breakout. But I don't think Broderbund is going to publish it. The knaves.
October 25, 1986
Yesterday I implemented the running animation. Next I'll do the jumping... then the stopping... then the "jumping from a stopped position"... oh boy, this is great!
I restrained myself from taking all my work papers home with me yesterday... and I'm restraining myself from going to work today. There must be Balance.
October 31, 1986
Ed was pretty thrilled with the rough running and jumping animation, now under joystick control. So was Tomi. Lauren, Doug, and Gary didn't act all excited, but I think they were secretly impressed.
I love the quality of the just-digitized roughs, but I'm having trouble preserving that fluidity and realism when I clean it up and stylize the figures. This is going to be a problem.
I beat out Ed and Steve for the #1 spot on the Tetris high-score list.
The Mets won the World Series.
November 9, 1986
God, I miss New York.
Fifth Avenue... Christmas shoppers... rich ladies in furs laden with shopping bags and kids... crisp cold autumn air... the smell of burnt pretzels... St. Peter's... the steel drum players wearing woolen gloves with cut-off fingers, breath condensing on the air...
I'm looking out the window at the San Francisco skyline across the bay dotted with white sails. It looks unreal. Like some kind of paradise.
November 10, 1986
Called Kyle Freeman in L.A. (he's at Electronic Arts now) and asked him what he'd charge to license his Apple music subroutine. He spent half the phone call dumping on Broderbund. I realized after I'd hung up that this was the first thing I'd done independent of Broderbund since I got here. Interestingly, it actually strengthened my confidence that Broderbund is the right place for me. It reminded me that I am independent.
November 18, 1986
Digitized the running skidding turn-around that was so amusing on videotape. It looks okay. I'll need to redo the straight running, but I think everything else will work as it stands.
About half the animations are in now. Next step will be getting the character to interact with the environment (climbing a rope ladder, pulling a lever, etc.)
At this juncture I think I'll redirect my attention to the game design.
How old were you when this all started?
Jordan Mechner: I had just graduated from college. I was 21, but I looked about 15. I couldn't help that.
Someone says that you looked like you grew up, later on in the book.
JM: Yeah. The ordeal of the software development had matured and aged me in a matter of months.
In this part, you talk about how you're going to get the footage that you recorded onto the computer, and it sounds like you really had to improvise.
JM: Yeah. Today, of course, you can point your cell phone at something, press a button, and you're done. But to actually get frames of video into the computer was a multi-step process. I'd done it on Karateka a couple of years earlier using Super 8 film, to get the rotoscoped animations into the computer.
But for Prince of Persia, in the intervening years, VHS had been invented, so the process that I found worked best was to go through VHS to still pictures, by taking pictures off of a freeze frame on the TV screen, getting them developed at the Fotomat, and then pasting them up and using a video camera to capture one still image into the computer, and then cutting out those pictures frame by frame. Kind of laborious.
It sounds incredibly laborious, actually.
JM: But so rewarding, after weeks of work to get all those steps in place, to finally see a little character running and jumping on the screen, and realize that, yes, this was going to work.
These days, there's a lot of technology in place that enables workflows in games. Did the fact you didn't have any technology to rely on free you, creatively?
JM: I think there's something about multitasking... I took three years to make Prince of Persia by doing, sequentially, all the things that a team of three people could have done in a fraction of the time. So in a way, it was inefficient, because I'd get an idea, then I would spend a month building the tool to let me do it, and then I would get back to putting the idea in the game.
But having so much time, I think, really gave room for the game to grow organically. In reading the journals you'll see there were key changes that I made to Prince of Persia when I was a year in, two years in to development, which totally changed the feeling of the game.
For example, adding sword fighting, which was not part of the original game design. Once the game was fairly far along, and much of it had been implemented, it really became apparent that this would inject an element of action and suspense into the game that was missing. So, yeah, I think time is a luxury that makes up, in a lot of ways, for not having readymade tools.
You talk about having a "development system", and how that's not something you had had before. In those days, with the Apple II, you could just program on it. It's not like now, where you have tools that cater to this.
JM: No. We developed software on the machine that was the target platform. When I started, I was working at home, in my parents' house -- so all I had was the Apple II that I brought to college with me. When I got to Broderbund, I saw development systems for the first time. And some of the programmers had hard drives, which I'd never seen, or two computers on their desks.
That opened up new worlds. When I realized I could have two Apple IIs on the desk next to each other, I could use one of them to test -- I could make a change in the program and then I could swap the disk, take a five and a quarter inch disk from one computer to another, boot it up, and test the change I made without destroying the source code file. Yeah, that was a huge time saver.
It's kind of amazing.
JM: Peviously, of course, the computer didn't have enough memory to hold the game and the assembler -- the assembler that would enable me to edit the source code text file -- in memory at the same time. It had to be one or the other. I had to wipe out the machine in order to run the software.
It sounds like such arduous work, but you also have an obvious enthusiasm for it.
JM: Yeah, I think I really enjoyed the work. I mean, I enjoyed that sense of immersion to a task that I thought was exciting and motivating. So I really cherish those moments of payoff, when they come together.
But of course, in the journals, there are a lot of emotional ups and downs recorded there. Some days I think the game is great; others I think it's not good enough, and should I even bother to finish it? But reading the journal, the times when I was happiest, I think, were when I was really deep in the work, and working long hours, completely absorbed in what I was doing. I didn't mind the long hours at all. I loved it.
There's a bit where you talk about working on a screenplay. I know you've always been interested in film, and it's interesting to see the tension between your two ambitions this early on.
JM: Yeah. When I was in college, and out of college, I was really torn between these two careers. I had the idea that I had to choose one or the other, and that the time that I was spending making Prince of Persia was taking me away from film, and vice versa.
It's ironic, of course, because 20 years later it was Prince of Persia -- this game that started on the Apple II -- that gave me my first breakthrough as a film screenwriter. That was, my first screenplay that actually got produced, Prince of Persia: The Sands of Time.
Excerpt Two: Enter Shadow Man
June 8, 1988
SHADOW MAN. Credit Tomi with this one.
I was explaining to her why there are no enemies in Prince of Persia. The animations for the player's character are so elaborate, there's not enough memory left to add another character.
"Why not use the same animations for your enemies, the way you did in Karateka?"
"Wouldn't work so well this time. This character is designed to look cute. He has a very specific personality in the way he runs and moves. The enemies would have to be cute too."
"Can't you just change the face, or the costume?"
"Not possible. If I change anything, it's a whole new set of shapes. There's just no memory."
She wouldn't give up. "Couldn't you make him a different color -- say, black?"
I started to explain: "This is the Apple II..." and then it hit me: What if I exclusive-OR each frame with itself, bit-shifted one pixel over? I visualized a ghostly, shimmering outline-figure, black, with white face and arms, running and leaping, pursuing you. I described it to Tomi.
"Shadow Man!" she exclaimed.
Tomi, Robert, and Eric all huddled around my screen while I paged through my source code.
Me: "Uh, you don't actually have to watch me do this. It might take a while."
Eric: "No, we want to. It's a test."
In about two minutes I had Shadow Man up and running. He looked great. It was as if he'd always existed. Everybody was wowed. How could I have ever contemplated the game without him?
Robert suggested that Shadow Man could come into being when you run through a mirror.
You leap through the mirror; simultaneously your evil shadow self leaps out the way you came, and slinks off into the darkness. For the rest of the game he's lurking in the shadows, dogging your steps... until the end, when you don the magic amulet and become powerful enough to reabsorb him into yourself, thus gaining the strength you need to defeat the Grand Vizier.
"You'll sell a billion copies," Tomi predicted. "All I want is a Honda Legend. Coupe. Silver."
Tell me about Tomi and Shadow Man.
JM: She was really uniquely talented, and a brilliant person, and she had a great ability to look at something and just come up with the one pithy remark that summed it all up. And in the case of Prince of Persia, for the first year and a half, I was clinging to the idea that this would be a nonviolent game, that this would be a hero who was escaping traps and slicers and spikes, avoiding all sorts of horrible death in these dungeons -- but he himself wouldn't hurt anybody.
And every time I showed Tomi the latest and greatest thing I'd put into the game, she would glance at it, and say, "Combat! Combat! Combat!" and walk away. Which drove me crazy, because in my mind, there wasn't supposed to be sword fighting, or combat, in the game. And anyway, there wasn't enough memory in the Apple II to include that feature, even if I'd wanted to put it in. But as the journals describe in detail, eventually Tomi had her way.
And it's hard to imagine what Prince of Persia would have been without it. Certainly, at the very least, I don't think it would have become a Jerry Bruckheimer movie.
It sounds like you were very much working alone, and you had your own work process, but it seems like you really did need to bounce these ideas off other people, and get input to take the game in the direction it needed to go.
JM: Yeah, it's tremendously important, and I hope that's something that comes through in the journals -- the extent to which even something that's a one person project really is a collaborative effort. And I think that's a lot truer than people realize, in many fields -- not just video game design, but even in fields like art, novel writing, music. I think any time something good is accomplished, it's not just the work of one person. It's the result of a time, and a place, and a community of people.
Excerpt Three: Hitting Bottom
November 11, 1988
"I like games where you can shoot things. Your game has no rewards except getting to the next level. It's all survival and no triumph." - Tomi
She's right about POP. It's empty and lifeless. I don't know if even the shadow man and sword fighting will change that.
On the other hand, I put in a new door which looks pretty good.
Oh, God. I want this game to be a hit. Like Karateka.
Maybe this whole modular-design approach is wrong. Maybe the thing to do is put in a whole bunch of hard-wired enemies, one after another, and forget the whole free-floating, random-access, 24-screens-per-level idea.
24 screens, if they're linked sequentially, could give a playing experience as satisfying as a whole level of Karateka. But they should be in the form of obstacles to be overcome one after another. For example:
- A chasm that has to be jumped
- A gate that has to be raised
- A guard that has to be killed
The way it is now, you're plunged into a huge arena with no overall idea of what you're trying to accomplish except "get out." It's too perplexing, especially at first.
Maybe after the first 10 or 15 levels, I could start introducing some real Lode Runner/Dr. Creep "puzzle" type gameplay. But in the beginning, it should be pretty much left-to-right (like Karateka) with a little bit of up-and-down. So the player can get his bearings.
November 12, 1988
Still not enough.
What's the point in running, running to get to the exit, if all it gets you is more of the same?
The princess waiting at the end is a reward only in the story. We need rewards in the game -- like beating a guard in Karateka. What makes a game fun? Tension/release, tension/release. Prince of Persia has neither. It's like going on a 25-mile hike. Every now and then, you get to step over a log or cross a stream. Big deal.
Running, jumping, and climbing, no matter how beautifully animated, hold your attention for maybe the first three screens. Then you start to wonder: when is something going to happen? Like: a guard to fight. An airplane to shoot down. Something.
There need to be sub-goals. Places where you can say: "Whew! Did it! That was a tough one!... What's next?"
- clearing a screen in Asteroids or Pac-Man
- beating a guard in Karateka
- solving a level in Lode Runner
Right now, solving a level in Prince of Persia has none of the feeling of accomplishment of any of these. It's more like "Oh... so that's the end. Oh."
What elements do All of the Above share?
1. You can tell at any moment, by glancing at the screen, how close you are to finishing, how much is left.
2. There are setbacks and successes on the road to ultimate success. You get a smaller version of the "Whew! Did it!" when, say, you clear a difficult area (Pac-Man), or drive a guard back with a series of blows (Karateka), or retrieve a hard-to-get sack (Lode Runner).
Conversely, you get the "Oh, shit..." reaction when you accidentally split up a bunch of bigger asteroids into more smaller, faster ones; or when you finish a pattern and see that you've missed one dot; etc. Some setbacks are fatal, some are just irritating. But when they happen, you feel they're your own fault.
3. You can hold off on the next task, waiting for the right moment, before saying "Okay... Now" and going for it... plunging into a period of higher tension, higher chance of either a setback or success.
Persia has none of these features at present.
If the sub-goal is "solving the level," you need a consistent visual indicator of how close you are. You don't just stumble onto the exit and say "Oh -- guess I'm done." Or stumble onto a sack of gold and say "Oh -- here's another one." That's why collect-the-dots games like Lode Runner and Pac-Man always show the entire screen at once. That's key.
But POP doesn't show the entire screen at once. That's a problem.
November 13, 1988
How can I be so up on screenplay story structure, and so blind when it comes to my own game?
A story doesn't move forward until a character wants something. So -- a game doesn't move forward until the player wants something. Five seconds after you press start, you'd better know the answer to the question: "What do I want to happen?"
There always has to be a range of possible outcomes, some better than others, so you're constantly thinking: "Good... Bad... Terrible." Every event has to move you closer or further away from your goal, or it's not an event, it's just window dressing.
The overall goal of POP is to get the girl. But that's not a strong enough magnet to pull the player through all that distance. It needs sub-goals.
Beating a guard in Karateka buys you time to gain distance. You want to get closer to the palace because the princess is there; every guard you beat brings you closer. It's simple, but it works. In psychological terms, it even follows the classic addictive pattern of diminishing rewards: each subsequent guard is harder to kill, and gives you a smaller reward for your pains, until you reach the intermediate goal (the end of the level), at which point there's a bigger reward, and things get easier again.... for a while.
Getting through a dungeon in Prince of Persia doesn't give that satisfying feeling of getting closer to the goal. Partly because it all looks pretty much the same. That, I can fix.
But there's another key element in story structure that also applies to games, and is missing from this one: The Opponent. Someone competing for the same goal as the hero, or trying to stop him from attaining it. The more human, the better. (The days of Asteroids and Pinball are over.)
In this case (we're short on time, so let's use the opponent we've already got), it's Shadow Man.
Some games boast a whole series of different opponents. (According to [screenwriting teacher John] Truby, this is characteristic of Myth, and it weakens the story.) We'll make the shadow man your opponent for the entire game. You're competing for hit points. Each blow you deal him weakens him. Each power dot you eat makes you stronger. But if he gets there first and he eats it, he gets stronger. So when you face each other with crossed swords, the balance of power is not predetermined (as in Karateka), but is the result of your own actions thus far in the game.
It links the combat with the running-around. It's brilliant. I love it!
(Forget the boring damn keys.)
Prince of Persia
You describe the game as empty and lifeless.
JM: Right, that's -- what was that, '88? This is another one of those unplanned bonuses of the roundabout way that the game developed.
Originally, I had developed a game with the idea that this was going to be packaged with a level editor, like Lode Runner, which was one of the games that had inspired Prince of Persia. And I thought that by packaging the level editor with the game, users would then have fun creating their own levels. And so I was thinking of it in a very abstract way, as kind of a puzzle game with action.
And at this point, at the point you mention, I realized that that the game -- although technically it had come together, and working in all the ways it was supposed to -- it wasn't giving the player a fun, exciting experience in the way I had hoped.
And so in a period of a few weeks, and months -- fairly close the end -- I radically changed the design of all the levels, the pacing, the rhythm of the game, the balance of action, exploration, and puzzle solving, and combat. I found the fun, found the balance that made Prince of Persia what it was.
But it was only possible to do that at that late stage because I had invested so many months early on in building a fully-featured level editor, that made it very fast for me to build new levels. You know, move enemies and traps around, and then instantly play the game, and iterate very quickly, so I could actually build a completely new level, and playtest it in a day. And that's what made it possible to rebuild the game at such a late stage.
That's something that's a lot harder to do today. Building an environment in a triple-A console game requires dozens of artists working for months to build the world, and make it look good. You have to make certain decisions early on, but because Prince of Persia was so modular, it was cheap to tear down a level and rebuild a new one from scratch.
Until I read this, I didn't realize what a huge influence Lode Runner was on you. Thinking of them as finished products, I wouldn't see the connection. But the connection, in the end, became more subtle, but also more profound, I guess.
JM: Lode Runner is deep in Prince of Persia's DNA. Lode Runner is like the light pencil sketch that you can't see anymore by the time you have a finished ink drawing.
You talk about trying to find the goal for the game -- the opponent. You write, "the more human, the better." That's something about Prince of Persia. It bucked the trends of the time. Prince of Persia is a very humanistic game.
JM: Yeah, I really wanted to make a game that would also be an exciting and visceral human drama. Even the animation -- sort of the original concept that launched me on making Prince of Persia was the idea that it would be a platform game where you could run, and jump, and fall to your death, and so forth, but that if you fell you would really feel that it hurt, that your character was flesh and blood, as opposed to the sort of the Lode Runner, Mario Bros. kind of characters that could float, that felt that they didn't really have bones to break.
Excerpt Four: The Key to Finishing the Adventure
December 2, 1988
Doug wandered into my office today and I gave him the joystick to play with. He was impressed. When he left he said: "I feel like I've had an adventure." I told him I'd have a version in a couple of months that would really be playable. He said: "Seems to me it's pretty close."
Also spent a couple of hours with Lauren E., and some good ideas came out of that.
I realized that the 50-level, Lode Runner approach is all wrong. What made Karateka so compelling is that it's easy. You boot it up and pick up the joystick and it's obvious what you have to do. You're