Sid Meier's Memoir!

by Sid Meier's Memoir! (retail) (epub)

  For the first time, flying a plane had become more like playing a videogame, instead of the other way around. I couldn’t cram any more instrument data into the bottom half of the computer screen, so the military had helpfully moved it up into the top half for me. The mountain had come to Mohammed.

  Now it was my turn to deliver the news that Bill had been waiting for years to hear: I was ready to make an F-15 flight simulator. He grinned like a kid on Christmas morning.

  4

  D-DAY

  NATO Commander (1983) * F-15 Strike Eagle (1984) * Silent Service (1985) * Crusade in Europe (1985) * Decision in the Desert (1985) * Conflict in Vietnam (1986)

  ONE OF THE THINGS MICRO-PROSE was famous for was our game manuals, which over time became as long and informative as textbooks. In the beginning, though, they weren’t especially tangential. Sometimes we had to explain to novice computer owners—everyone, basically—that the computer was working as intended, such as Spitfire Ace’s reassurance that the game would finish loading “in about 4 minutes.” Other times we were establishing gameplay conventions that are taken for granted today, like “The screen will flash red when you are hit” from Hellcat Ace, or Chopper Rescue’s promise that “More points are awarded in the more difficult levels.”

  Mostly, though, we just wanted to be helpful, because nobody likes a game they can’t win at least some of the time. Formula 1 Racing advised players to “be careful taking the corners in fifth gear: fourth is recommended,” while Hellcat Ace encouraged you to “Line up your next shot immediately, don’t wait!” The manual for Spitfire Ace even let you know that “The sky and ground are light blue and green respectively.” Maybe we went a little overboard with that one.

  The flying games, in particular, had pages full of information on special aerobatic maneuvers, partly because Bill had the knowledge and loved sharing it, but also perhaps because he was afraid the neat tricks we’d put in might go unnoticed if we didn’t point them out. As each game added more details, each manual got longer. By the time we released Solo Flight, we were including narratives about the experience of flying that were well beyond what the game could simulate.

  “Losing your attitude indicator in instrument conditions can be one of the most frightening occurrences in real flying. Combine this emergency with engine failure and smoke in the cockpit, and the pilot would be happy to use his silk elevator (parachute) to get his body back on the ground in one piece!”

  What had taken sixteen pages with Solo Flight now stretched into thirty-six with F-15 Strike Eagle. We taught players that a speed of Mach 0.9 was actually a variable threshold, equivalent to “661 knots at sea level” but significantly less at higher altitudes. We provided complex graphs outlining the G-force difference between a seventy-, seventy-eight-, and eighty-two-degree turn. We listed the stall speed, service ceiling, and armaments of each completely accurate enemy aircraft, as well as the slant range of their surface-to-air missiles. The manual’s centerfold diagram identified all twenty-nine indicators on the cockpit screen, followed by the ten different actions that could be accomplished with the joystick, before launching into a lengthy section on the difference between ailerons and rudders—even though the simulator “automatically interconnects these control surface movements to apply the correct amount of up elevator.” We thought of everything.

  Well, almost everything. Nowhere in the thirty-six-page manual did we mention how to land the plane.

  This was critical information; it was impossible to land the plane on your own. We had never found the right balance between accuracy and playability. Landing the aircraft is the most difficult part in reality, but killing the player at the last moment of an otherwise successful mission was not a way to earn fans. So the compromise we came up with was to have the computer take over and automatically land the plane whenever you approached your home base from a reasonable distance and altitude. Unfortunately, we took it for granted that people who didn’t make airplane games for a living would know what counted as reasonable in that situation. Oh, well. It still sold pretty well.

  Omission of vital game mechanics aside, the manual for F-15 Strike Eagle was special in another way: it was our first game to attempt manual-based copy protection. Digital rights management, as it’s known today, remains the eternal battle between creator and user. We come up with a way to protect a game; someone figures out how to break in. Rinse and repeat. Of course, Bill was adamantly against any practice that took money out of our pockets, but given the amount of software pirating I may have done in my younger days, I didn’t have a peg leg to stand on when it came to casual sharing. There is something, however slight, to the argument that pirated games are a form of advertising to people who wouldn’t otherwise have bought them. I purchased games on several occasions in the early days after being exposed to free versions, and I wouldn’t have learned nearly as much or as fast about programming if I hadn’t had real, unencrypted data to play around with. (Back in those days, the player’s computer did the compiling on the fly as the game loaded, so the data on the disk was not only visible, but completely editable given the right tools.)

  That being said, there are plenty of habitual pirates whose motives go beyond curiosity. I can’t condone profiteers, and no one’s a fan of the actively hostile users who hide malware inside the tempting download they’ve provided. Fortunately, we didn’t have to worry about the latter back then, since the first computer virus wouldn’t appear in the wild for another year or so after F-15 Strike Eagle was released. Ironically, that virus was originally intended to be an aggressive form of copy protection itself: when the program detected what it thought was a pirated version of the authors’ software, it would begin erasing critical parts of the user’s hard drive in retaliation. Occasionally, it targeted the innocent. Nicknamed “Brain,” this well-intentioned but poorly executed virus included the full names and contact information of its creators, because they saw no need to hide from the pirates they thought they were targeting.

  F-15 Strike Eagle screenshot.

  © 1985 MICROPROSE, WWW.MICROPROSE.COM..

  But even if we were too virtuous to conceive of malware in 1984, unauthorized sharing was rampant. The science fiction novelist Orson Scott Card once wrote that a particular MicroProse game was so good that “even people who drive 55 mph might consider stealing it.” Sometimes, a game of middling popularity would end up with more legitimate sales than a big hit, because it was harder to connect with someone else who owned it. Widely celebrated titles were easy to find on local bulletin boards, and according to some estimates, up to 80 percent of the copies being played might be pirated.

  Videogames were considered artistic works under copyright law, thanks to the 1982 court case Stern Electronics Inc. v. Kaufman, but in reality there was no enforcement. Entire businesses were dedicated to breaking software protections, and they conducted themselves openly without fear of reprisal. Softkey Publishing, for example, was so successful that they could afford to circulate two separate monthly magazines full of code-breaking instructions just for software on the Apple II computer.

  Working against us was the fact that data storage was limited, making the programs so small that a determined pirate could comb through each line of code by hand. The term “open-source” only had to be created after companies figured out how to make things closed-sourced, and genuine encryption belonged only to the military. There were a few data layout tricks we could use, like storing the information in a skewed spiral on the disk instead of in straight lines, but these were never too hard to figure out, and they sometimes made legitimate copies of the games unreadable.

  Working in our favor, however, was the equally limited pace of data transfer. The fastest modem on the market in those days cost about $600 (over twice as much in today’s dollars), and could transfer data at the blindingly slow speed of 1,200 bits per second. This meant that a typical 48K game (that is, roughly one-third the size of the Wikipedia page explaining what a kilobyte is) would tak
e five or six minutes to download—not too bad for the potential game thief. But a single digital image, made from a real drawing and not ragged pixel art, could easily be as large as the game itself, making our manual full of pictures effectively impossible to send over a phone line.

  The manual was a minor loss if it contained only instructions on how to play—games were expected to be intuitive, and many players just figured it out as they went along. But if the manual contained crucial information that the player couldn’t progress without, then its absence would break the game without altering one bit of data. It’s hard to pinpoint exactly whose idea it was, but there are several examples of “manual lookup” copy protection in games from 1984, and none from 1983, so clearly it caught on fast.

  Many of the early examples were just tedious, along the lines of “what is the 12th word on the 17th page of the manual?” The better ones tied the information into the game somehow, presenting it as the words to a magic spell, or the answer to a riddle put forth by a tricksy enemy. For the decidedly non-fantasy setting of F-15 Strike Eagle, we went with top secret weapons authentication codes. Though we scattered them throughout the manual to avoid easy photocopying, our attempt was still too simplistic. There were only fifteen codes you might be asked to choose from, each consisting of a single letter. It was too many options for a player to guess randomly, but not too much information to copy by hand and include as a small text file along with the pirated data.

  By our next game, we got smarter. Instead of providing codes, the manual for Silent Service required the player to visually match silhouettes of imaginary destroyers. The squared-off, black and white shapes were simple enough to store in the game’s memory, but nonetheless too complex to describe verbally or convert into text.

  Even after home scanners entered the market in the late 1980s, and data speeds increased to the point that images could be easily shared, the hassle of it all was enough to deter most casual copiers. I think people in general are honest, as long as the dishonest choice isn’t ridiculously easy. When it comes to elite hackers, the biggest roadblock in the world isn’t going to stop them anyway, so we don’t sweat them too much. It’s not great that it’s happening, but every copy protection scheme has been broken eventually, and somehow, game creators have survived.

  Silent Service instruction manual.

  © 1985 MICROPROSE, WWW.MICROPROSE.COM.

  Though the destroyer silhouettes took up only a few more inches than the weapons codes, the manual for Silent Service expanded again, proving that there was more to our compulsive writing than just the obstruction of thieves. The discussion mostly covered the tactics of real submarine missions—this was my first game based on stealth, rather than speed and firepower, and the fact that aiming torpedoes involved lots of trigonometry made it even better. Bill, however, subscribed to the old military joke that the acronym for anti-submarine warfare stood for “awfully slow warfare,” and he was not satisfied even after we added options for speeding up the game clock and automatic aiming. Bill preferred to do battle at Mach 0.9, and the careful strategizing needed to win in Silent Service bored him immensely.

  “Can’t you just get on the surface and shoot it out with guns?” he complained.

  “That’s not really the point of this game,” I’d remind him, again and again. But he wouldn’t let up, and finally, I added a deck gun to the main submarine model just to appease him.

  Not long after that, Bill was demonstrating the game to a major buyer, and it looked as if the AI would get the better of him, which is a bad thing when you’re trying to sell your game. Retail executives are not interested in how clever or intricate your programming is. Most of them never play games at all, relying instead on their ability to read others’ emotional states. If you’re defeated during a demo, the whole room feels the letdown with you, and it’s almost impossible to erase that first impression of disappointment. Of course the game shouldn’t always be easy, but the time for losing comes later, after the rewards have been firmly established. In those first crucial minutes, the player absolutely must win, whether it’s a kid alone on his computer or a gaggle of salespeople around a conference table.

  Bill, ever the showman and still not a fan of slow tension, had chosen to take on multiple enemy ships at once, and all of them were lumbering toward him in what passed for a terrifying charge at submarine speeds. He fought and dodged as best he could, but eventually the last one had him blown to the surface and out of torpedoes. With seconds remaining before his doom, he switched to the deck gun and obliterated his enemy in a spray of gray and white seawater squares. The whole room cheered, according to Bill, and possibly put him up on their shoulders for an impromptu parade. More importantly, they bought the game.

  From that moment on, Bill was a designer, and any time he felt one of our games was lacking in excitement or cheap thrills, he would yell “deck gun!” to argue his point. It became a running joke that lasted for years.

  Artillery wasn’t the only area where Bill stepped in to improve Silent Service against my instincts. He also decided that it was time for MicroProse to hire an artist.

  I was, to be honest, a little offended. Sure, I was no Van Gogh, but I had been doing our game art for years and felt like I was pretty good—more than sufficient for sixteen-color graphics on a sprite grid, at any rate. Heck, I was so good, I didn’t even have to plot my pictures out on graph paper like some designers did. I just visualized what I wanted, and entered it straight into the computer! I was especially proud of the graphical menu I had designed for Silent Service, where instead of choosing from a list of things the captain could interact with—radar, periscope, damage reports, and so on—I had drawn a full-screen interior of a submarine conning tower, and a little human-esque captain that you could move back and forth between the different areas of the room. I was good, darn it.

  Then I saw the conning tower screen created by our new artist, Michael Haire. His 3D perspective was truer, his color contrast was livelier, and his captain looked human, without any need for an -esque. It was better in basically every possible way a work of art could be better.

  Oh, I thought to myself. I guess we did need a real artist.

  Painful revelations about my own skill aside, I couldn’t help but be happy with the improvement to the game, and I consoled myself with the thought that this would leave me more time to spend on other aspects of programming. At some point, I had gotten it into my head that what Silent Service needed was a realistic map of the entire Pacific Ocean, including all the tiny islands no one knew the names of, and accurate water depths throughout. Now that art had been taken from me, I was even more determined to make the map special, and soon I worked out a programming trick based on fractals that allowed nearly infinite zoom, from a global view down to a rectangle of ocean just eight miles across. It wasn’t exactly “open world” by the modern standards of Minecraft or the Fallout series, because there was still only one thing you could go around doing, but it offered as much freedom as you could realistically get in a submarine game without attaching wheels and rolling it up onto dry land.

  Silent Service screenshot.

  © 1985 MICROPROSE, WWW.MICROPROSE.COM.

  Meanwhile, F-15 Strike Eagle was defying all expectations, selling hundreds of thousands of copies and winning “Action Game of the Year” in a readers’ poll from Computer Gaming World. This only made Bill’s hunger for aircraft games more intense, but I was ready to move on. It wasn’t burnout as much as I was just out of ideas. F-15 had included every type of missile and bomb in the American arsenal, and our mechanics were as close to the real thing as we could get without security clearance. The radar was top notch, and so was the chaff you could deploy to confuse the radar. Every item on the screen was rendered in 3D, from the landscape to the projectiles to the enemy aircraft. I saw nothing left to improve.

  Fortunately, other folks at the company were keeping a steady pace for now, with solid genre entries like Kennedy‡ Approach and Acro
jet. They afforded me a certain freedom to explore, without making Bill worry too much that we were breaking the formula for success.

  “Sid’ll figure it out,” he always said. Whether he meant I’d find new inspiration, or come to my senses, I wasn’t sure.

  His reluctance to diversify was not completely unwarranted. Before Solo Flight, I had made a brief foray into new territory with a game called NATO Commander, and it was, to put it politely, not my best work. Or, as I described it to one journalist many years later, “It was not even fun to play. It was just bad.”

  The idea had been to create a wargame on the computer that would eliminate all the drawbacks of the traditional tabletop versions. Wargames had developed out of the strategic planning done by actual generals, pushing around miniature platoons on a giant map of the battlefront. Training scenarios were developed for the officers who would eventually be playing with real lives, and later these fictional setups evolved into games made available to the public. Often, wargames are historically accurate reproductions of specific battles—giving you the option to play out Custer’s last stand differently, for example—and they never stray beyond the military technology contemporary to their setting. Other key features include dozens of easy-to-lose miniatures instead of a single player piece, a map that takes up the entire table and requires hours to set up, and a complicated rulebook that you and your friends can argue over for at least as much time as you spend playing the game.

  It was deficits like these that convinced me a computer could do it better. Along with instant setup of the board and a rulebook that never lost track of the exceptions, computers had the important ability to hide information from the player. Modern satellites may have nearly eliminated the fog of war, but for most of history, military commanders were blindly guessing about rival troop movements. Many battles took place only because of accidental encounters, and sometimes not even with the enemy. During World War I, for example, the British submarine HMS G9 stumbled upon the British destroyer HMS Pasley, and the two exchanged fire until the G9 split in half and sank, leaving only one survivor to inform the Pasley’s captain of their mistake.