How Myst developers managed to fit the whole universe on one CD-ROM

Hi, I'm Rand Miller, co-author of Myst. We set ourselves the task of creating the real world of “Myst”, but painfully hit the real limitations of the CD-ROM. I started playing games since the beginning of high school, and I'm not young anymore, that is, it was a long time ago. The games that I then had on the computer are unlike modern ones. I was very captivated by the Lunar Lander game. I still remember her. It had no graphics. It was a line of text on a CRT screen telling me how far I was from the lunar surface, how fast I was moving, how much fuel was left, and also a question mark. It was necessary to enter the amount of fuel that you want to burn, press the input, and repeat so on until you fall to the surface.

At the beginning of high school, a college friend took me to the computer center at the University of New Mexico. I saw this game on the screen, and it was real magic. For me it was magic. And he said: “There are other games,” and that you can make games yourself. “Now, you write programs to do it.” For me, this changed everything, and since then I wanted to do it. Cyan was formed around the mid-80s, and at that time it seemed to us that this is a very ambiguous name, so we can do anything. My brother Robin and I had the idea of ​​a blue sky, and we liked it, so we chose that name.I wrote to my brother: “We need to turn this into an interactive book. You need to make a book, it's awesome. ” He was an artist, musician, he knew enough about computers to be dangerous, and I knew enough about graphics and music to be dangerous, and it was a great team. We started with The Manhole. He drew the first page of the book, on which were a manhole cover and a fire hydrant on a painted background. What happened next became an interesting transition for us, turning us into what we have become. He did not turn the page. Having drawn a hatch and a hydrant, we were no longer interested in what would happen on the next page. We were interested in what was under the hatch and why this hydrant was needed. After that, he drew pictures with the hatch cover pushed back and the vine growing somewhere upwards, and also the picture closer to the hydrant with a small door below.These were just still images that he sent me when we corresponded. And at some point we suddenly realized that we were not creating an interactive book, but creating a world. I think it was one of the first entertaining CD-ROMs, if not the first.

Further, our evolution moved along the Cosmic Osmo. It was a very natural transition. We learned a lot about our HyperCard platform, simultaneously doing “Manhole” and figuring out what suits and doesn't fit this world. Cosmic Osmo was a chance for us to develop further, while remaining close to my heart for my brother and me. It was a test of the capabilities of what HyperCard and computers were capable of at that time. Everything was interactive. I would do projects in HyperCard today if it were available. It was such an elegant, amazing tool that I think Apple did not fully understand how to position it on the market. At its core, HyperCard can be thought of as a stack of virtual cards in which an invisible button can be applied to each card. It was possible to draw on the map an image of, say, a fire hydrant,and put an invisible button on it. And if you click on it, then you will get an enlarged image of this hydrant. Gradually, this turned into what we were doing: a chain of still images.

One Japanese company contacted us. She liked The Manhole, and we had a meeting with us where we were told: “Can you do something similar for an older audience?” We replied: “Yes, of course, no problem.” The proposal consisted of seven pages, it was a proposal to create "Myst". In fact, these seven pages were maps with a top view of each island with small notes. We did not know how proposals were made to create games. Before that, we only dealt with our games. It seems that then we asked for 250 thousand dollars, and they replied: "Well, we will give them." “The game should be as good as the 7th Guest. Can you make her better than 7th Guest? If you have not seen 7th Guest, then it is worth a look. ” At that time, this game was not yet ready, but it was already advertised. They showed us a preview of this media project on a CD-ROM,and we said: "Well, yes, yes, of course." After that, we immediately started to work. As I said, “Myst” became a very natural evolution of our previous worlds, but at the same time it turned out to be a leap. Because from previous worlds, we learned that pieces of the plot can be embedded in the game. While working on The Manhole, we did not understand this at all.

But later on we realized that we liked it, and these fragments of history allowed us to maintain sincerity. Somehow, they provided realism to the world, because in the real world everything has a story. Everything has a reason. When we were ready to start making the game for an older audience, we realized that it was important. What we need to associate with the objects of history. And on a floppy disk, this was not at all possible. There simply would not fit anything. One of the attractive features of the game was that the player does not die, does not increase his level, is not forced to start all over again, and the only way to ensure a long passage was brute force - expanding the spaces available for research. If we succeeded, then the answer to this question would be a CD-ROM.

Even in the early stages of development, we were able to acquire software for 3D rendering. This was one of the turning points, we realized that now the future is becoming different. I, yes, probably Robin too, had a memory: we work with this software - we create a ball on the screen, put a table, arrange a light source, and all this is frame-shaped, very simple. Then we click on the Render button, and we get a picture with the generated shadows and reflections, refractions in the glass. “Wow, now everything is different. You can do it like this. We can render for "Myst" image. It’s not necessary to draw them by hand. ” We realized that the game will be in color. This will be our first product, which can only be played on machines with color. Black and white monitors are useless here. Robin worked on a black and white Macintosh, but he was hacked.The machine did not have any expansion slots. It was necessary to remove the back cover, under which was a crazy board with a chip, which was connected to the processor by contacts, and allowed to display colors on the monitor. So we managed to run “Myst” on the old Mac SE, it was hacked to get colors.

We encountered very interesting problems, and one of the most serious problems was this crazy idea of ​​a CD-ROM. It was a relatively new technology, and it meant that most people had a single speed CD-ROM drive. It transmitted data at a speed of 150 KB per second, and in fact it is nothing. On this single-speed drive, you can get a maximum of 150 KB per second, and we had to make the game work. The biggest problem was not even streaming data. The most important was the head positioning time. The richest had double-speed drives, I’m not even sure if they already invented four-speed ones. But this is not important, because we had to sell the game to a wide audience. And it was not even about sales. We wanted people with the simplest multimedia computers to play the game. We didn’t even knowwill it work out, will the project become playable. How long will it take to switch from one image to another when you click the mouse? It should also be noted that before developing games, I worked for several years in a bank, so I learned about one psychological principle. It seems that they were talking about him at some IBM conference: “Two seconds is how long people should wait until they get feedback, from pressing a button on the screen or clicking a mouse.” And after two seconds, they begin to realize that there will be no response. They begin to wonder whether it is in them or in the car, so you need to adhere to this rule. But we had no idea how long the transitions between player movements would last. It might take 10 seconds to go to the next image. We just didn't know that. CD-ROM recorders were hard to get. They cost thousands of dollars.We didn’t have it. Our publisher Broderbund could have a recording device, but until we started the project, we didn’t even use it. These were interesting times when there was no testing at all. Robin generated images. Of course, he always did this, but now he was like on steroids. He generated images at home, sent them to me, and I connected them together in HyperCard. And Robin sent me full color images. He had a very powerful Mac with lots of memory and a large hard drive. I also had a powerful computer. And both of them still slowed down. Everything worked incredibly slowly, especially for Robin, rendering images was very slow. It was a wonderful time when Apple released a new car, and we decided that we could afford it, because it will be a replenishment of our collection. Then the following happened:Robin created a queue of multiple images for rendering. In other words, he could not render them immediately in the process. He had to accumulate them in the queue, and at one stage of work we purchased a new computer. When he received the new computer, the old one was pushed aside and was used as part of the rendering queue. Perhaps another computer was added to the rendering queue. When Robin stopped working or was interrupted for lunch or coffee, he immediately turned on the rendering on his machine, adding it to those two that were already rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.he could not render them immediately in the process. He had to accumulate them in the queue, and at one stage of work we purchased a new computer. When he received the new computer, the old one was pushed aside and was used as part of the rendering queue. Perhaps another computer was added to the rendering queue. When Robin stopped working or was interrupted for lunch or coffee, he immediately turned on the rendering on his machine, adding it to those two that were already rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.he could not render them immediately in the process. He had to accumulate them in the queue, and at one stage of work we purchased a new computer. When he received the new computer, the old one was pushed aside and was used as part of the rendering queue. Perhaps another computer was added to the rendering queue. When Robin stopped working or was interrupted for lunch or coffee, he immediately turned on the rendering on his machine, adding it to those two that were already rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.When he received the new computer, the old one was pushed aside and was used as part of the rendering queue. Perhaps another computer was added to the rendering queue. When Robin stopped working or was interrupted for lunch or coffee, he immediately turned on the rendering on his machine, adding it to those two that were already rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.When he received the new computer, the old one was pushed aside and was used as part of the rendering queue. Perhaps another computer was added to the rendering queue. When Robin stopped working or was interrupted for lunch or coffee, he immediately turned on the rendering on his machine, adding it to those two that were already rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.that have already been rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.that have already been rendering. So it was possible to implement distributed rendering, and all free machines were engaged in the generation of images accumulated in the queue. To complete the work, everything had to be rendered constantly.

Living within limits is what we have to deal with. And I do not think that we constantly perceived it like that. There were a lot of limitations, and they just seemed to us tasks that need to be addressed. I like to solve problems, and I think this is part of your work if you are going to make games. If you do not solve problems well, then it’s better to do something else. So, we had two tasks: we need to find the data on the CD-ROM, and load it from there as quickly as possible. Let's start with the images, because this is the most important. These were 8-bit images, because for most people computers could not display more than 8-bit color. This meant that we can display on the screen at a time no more than 256 colors. And the images should have been located on the tracks close to each other. We did not want to have to look for them in some distant places.You could not just put them on a disk in alphabetical order. Therefore, we ordered them by epoch in the immediate vicinity. We gave them names so that the image close to the current one is relatively close to the track. For images, we used two levels of compression. First, we used only 256 colors for storing images, so their size was reduced, but we also used the compression technique to reduce them and increase the download speed. We managed to reduce the size of each to about 50 KB. So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.Therefore, we ordered them by epoch in the immediate vicinity. We gave them names so that the image close to the current one is relatively close to the track. For images, we used two levels of compression. First, we used only 256 colors for storing images, so their size was reduced, but we also used the compression technique to reduce them and increase the download speed. We managed to reduce the size of each to about 50 KB. So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.Therefore, we ordered them by epoch in the immediate vicinity. We gave them names so that the image close to the current one is relatively close to the track. For images, we used two levels of compression. Firstly, we used only 256 colors for storing images, so their size was reduced, but we also used the compression technique to reduce them and increase the download speed. We managed to reduce the size of each to about 50 KB. So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.For images, we used two levels of compression. First, we used only 256 colors for storing images, so their size was reduced, but we also used the compression technique to reduce them and increase the download speed. We managed to reduce the size of each to about 50 KB. So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.For images, we used two levels of compression. First, we used only 256 colors for storing images, so their size was reduced, but we also used the compression technique to reduce them and increase the download speed. We managed to reduce the size of each to about 50 KB. So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.So we dealt with the image problem. We also had a problem with music, let's just call it a sound problem. We had music and sound effects that were constantly playing in the background, and we needed to constantly stream them.

That is, if, for example, the player goes through the tunnel to the generator and I need to play a strange, creepy sound, then you can implement music in different ways, but the easiest way is to simply place the music tracks somewhere on the CD-ROM so that the drive goes to these tracks and reproduced them myself. Let the OS or QuickTime stream them all themselves. But can they play the tracks and look for the next picture, come back, then look for the next picture, and so on, without interrupting the sound? Because it is very likely that I will load this sound, start looking for the next image, and if the size of the fragment of the sound is too small, then I will not be able to quickly return to the sound to load it into the buffer. Music will suddenly stop, there will be a pause, then the drive will find music,will return to it and start streaming again. Therefore, we tried to place musical fragments close enough so that we did not have to search for them too far. In addition, when I press the button, I want to hear a clicking sound, and if I open the door, there is a creak, when the generator is turned on, there is a growing hum, and since these short, instantaneous sound effects should play quickly, they need to be close to the CD-ROM. And we meticulously had to solve all these aspects in order to maintain a low volume of transmitted data, even to compress them.and since these short, instantaneous sound effects should play fast, they need to be close on the CD-ROM. And we meticulously had to solve all these aspects in order to maintain a low volume of transmitted data, even to compress them.and since these short, instantaneous sound effects should play fast, they need to be close on the CD-ROM. And we meticulously had to solve all these aspects in order to maintain a low volume of transmitted data, even to compress them.

We had already developed a strange feeling, we could accurately represent all the calculations, and we knew right away whether this or that would work. For five or six years of work, we have developed this feeling, and fortunately, we were right. The workflow was very strange - we sent one or more hard drives to Broderbund, they burned the first Myst CD, and so we got a golden master disk. We inserted it into our CD drives, clicked on Play, crossed our fingers and started testing. The game started and most often worked. Ordinary people would not be able to do without medical assistance, but we had a very strange naivety that allowed us not to stress because everything might not work at all. It’s strange to understand that in a way this saved us. I don’t know if we would have undertaken the project at all if we hadn’t had such an attitude.Robin and I discussed sales: “I wish I could sell 100 thousand copies. Can you imagine that we will sell 100 thousand copies of the game? ” It was an incredible figure. We did not expect such a success. I remember when I first came to the nearest media store. After the release of Myst, I took a long vacation. I went to New Mexico on vacation for five weeks. And in New Mexico, we just released the game. I went to a media store in Albuquerque. I was wondering if they had our game. I went deep into the store, and there the whole top shelf of the gaming department was occupied by it. It shocked me. “Oh my god, it's just crazy. It can't be, it's so cool. ” Robin and I said: "It will be awesome if we sell 100 thousand copies, just awesome." And we quickly exceeded this figure in a few months. And the game continued to be on sale.She was in the top ten most popular. For a long time she was the first on the list, and remained in the top for several years. And so we quickly reached millions of copies. Several million copies. The irony is that, probably, only 50% of the people who played it got out of the island of Myst, because it was necessary to solve puzzles. But for some reason, this did not decrease enthusiasm.

Perhaps due to the fact that the first island was so intriguing, and the player got enough story to intrigue. Perhaps part of the audience was young at the time, and for her it all seemed magical. The game left such a deep mark in the souls of people. I meet people who were too young to play “Myst”, they said: “Yes, I remember, in my childhood my mom and dad played Myst, and I watched the screen, and I didn’t understand anything, but it was just it's interesting to see all this. ” And I am very pleased that the game has an impact today. I can’t convey how wonderful it is to be the creator of this game. We still think that “Myst” was an experiment. Like many of our other games, it evolved and led to Riven. We understood - to what we liked in “Myst”, we came a little later, when the plot developed a little further,when the story told more about the world and its people, it became a little more holistic. And the Myst budget did not allow us to realize this, but at Riven we had enough money to hire more people, to reveal the plot more. In addition, we made the puzzles more holistic, because in the first game they seemed disconnected from each other. Although many say that in “Myst,” even on the island of Myst, puzzles seemed logical, this was not so. But this can be compared to The 7th Guest, in which the puzzles were simply fragmented. I’m not trying to downplay her contribution, she was awesome, but at that time puzzles were just thrusting into the game, for example, playing chess and the like. We tried to integrate them so that they seemed natural, but that was not enough. The player should have felt that the puzzles are part of the story of this place.Therefore, in “Riven” it was very difficult for us to do this, during the development of the design we thought: “How to embed the plot and story into this world? How to embed puzzles into it? How to integrate all this? ” This was one of the biggest problems, and the lessons we learned led to Riven.

I want to say an interesting thing about the game industry. Everything has changed in her, and I like these changes. I started making games with my brother. We did what we would call indie today, what we could do on our own. And then we noticed a trend: everything is moving towards the fact that only people who can make games will get a lot of money from corporations-publishers, and in a sense it was sad. Yes, we were lucky, we were on the right side. We had money after Myst and we continued to work. But it seemed that turning away from the guys working in the garages, the industry as a whole lost innovation, that indie feeling. The advent of the Internet, and in particular of downloadable content, after which it was not necessary to print CD runs and try to distribute them in stores, fighting for shelving places,suddenly spawned a whole new influx of young companies of one or two people: “We have an artist and a programmer. We have to make a game. ” And I really like it. It seems to me that all this flowering indie has breathed new life into the industry, I am glad that this happened. In a sense, history has completed a full cycle. Now we are not dependent on the publisher. With our last two projects: Obduction, which was on Kickstarter, and the current Firmament, the same thing happened. We turned to Kickstarter twice, and our fans said: “Yes, we will take a chance to see your new game.” In a whole new series. Strong pressure is connected with this, but also a bold position: “Here I am, creating a completely new idea for the game.” Difficulties continue, 60-year-old I am working on Firmament, trying to understand what will happen in two years. I have the same difficultiesas before, I am conducting the same battle, and I understand that I can not do everything right, but this is the way, and this is completely normal.