Bob Smith, who passed away in 2023, was a longtime video game developer who got his start writing games independently before taking a job at Atari. From there he became something of a stalwart member of the development community for over 20 years, taking positions with Imagic, Bally Sente, Electronic Arts, Accolade, 3DO, and Skyworks before retiring in the early 2000s. This interview focuses primarily on his work with the Atari Video Computer System (2600) but also touches upon his time outside of that platform as well.
Kevin Bunch: I guess my first question is how did you get started at Atari itself. What led you to that company?
Bob Smith: That’s a good one. We have to go back a little further than that. I built things. When I graduated from high school, I get a couple of years of college and decided I wanted to go build sailor boats and sail around the world so I bought plans and started building. I had gotten married at this point and I’m 20 years old. It’s a long time ago. Finished the boat, discovered my wife had incurable sea-sickness so I said, “Okay.” She wanted to start a family so I said, “Let’s move to New Mexico and I’ll build us a geodesic dome,” which I did. We had a couple of kids in the meantime and about then– Now we’re in the mid-70’s and finally microprocessors had come to the fore.
I had looked to computer science years ago and they handed me a deck of cards. Punch cards was not my idea of computing and certainly, certain business software was not my idea of computing. I finished the dome and I had been doing work for a lumber yard and talked him into buying a kit computer, which I built, and then I coded them an inventory package and an accounts receivable package for their lumber yard.
While I was doing that, I learned assembly language, which of course was the only game in town. BASIC was a joke. You couldn’t do anything reasonable in BASIC. It had to be in assembly language. I taught myself 8080. The machine I had built was a Processor Technology Sol computer, S-O- L. It ran an 8080. At about that time, RadioShack came out with the Trash-80, the TRS-80, which I’m sure– If you’re into the history, you know all about Trash and it was running a Z80.
While I was coding for the lumber yard, I wrote a game. The Processor Technology actually came with two assembly language games which were thoroughly addicting. Just dumb. Death from Above. Which was probably– which is a– with big video-game thing. That was the theme of the first game I saw. I thought, “That’s pretty cool.” and then so I learned how to program the Processor Technology in assembly language and as it turns out, the RadioShack had a very similar system of characters mapping the graphics. All the graphics at this point are character mapped. You have a map of character usually 80 characters wide by 20-24 lines tall and you could use letters and notes, letters and special characters. That’s how these early games were done on business computers, essentially, because there were no graphics cards.
Then I discovered that the TRS-80 had a way to put special characters in place of the actual characters. In other words, you could sort of do graphics and it was six pixels per character or something, I’ve forgotten how many pixels there were per character, but it enabled you to talk to pixels below the character level. I wrote what we called the first-person driving game at that point and by that I mean an over-the-shoulder view of a racing track.
It looked a lot like– I’ve never seen this before actually. Night Stalker or Night Driver. Night Driver was probably the closest, that was the Atari one. I wrote that for the Processor Technology mainly for my own amusement. Then when the RadioShack came out, I’ve learned that it was very, very similar, and of course, this was a Z80, so it was actually a superset of 8080, so 8080 [code] was working just fine. I modified it to use subpixel graphics that the TRS-80 was capable of, finished the game, and sold it to a magazine called Creative Computing.
This was one of the early, early magazines. They bought the game and sent me $100, I think it was, as an advance on royalties and I sent them a tape. About that time, now we’re talking 1979 and 1980 and we were living in Northwestern New Mexico at 6,000 feet elevation. I had finished the dome by this point. We had two kids. My wife said, “I don’t really like New Mexico.” She was born in the Bay Area in San Jose so she was basically a Californian. She said, “I’m just tired of this. Let’s get the hell out of here,” so we did.
I sold the dome for peanuts. I think I made 2 cents an hour for the time I spent building the dome. We had 20 acres at that point so I had built a 20×40 shop. I had a shop in the dome and 20 acres. We sold that and moved back to California.
At that point, I was pretty conversant in 6502, the assembly language that as it turns out the 2600 is new. I had learned this after 8080. People had told me, “This is what you need to learn. Everything is in 6502.” Chuck Peddle had made a brilliant little processor. Simple, clean, and cheap. $2.50 for a processor was unheard of at the time.
We came west, or back west. When I was building boats, I was building in Santa Clara actually, right near the new 49ers stadium in Santa Clara. We were back basically the same old neighborhood and I interviewed in two places one was Atari, and the other was a Swiss power grid company who did software for power grids. It was a very staid, very stable company. I didn’t have a college degree at this point. I had a couple of years of college but basically, I was self-taught in computers. I decided that, yes this will be very stable and as it turns out, both Atari and the Swiss power company offered me exactly the same deal, $16,500 for a year.
This is 1980 now. This is the very beginning of 1980. Of course, the Swiss power company said, “The first downturn of the business, Atari will lay you off like that. There will be no questions asked. You’ll be gone.” I went to Atari [chuckles] and I never looked back. It was smart, creative people. I couldn’t ask for more. I was working for wonderful people. Dennis Koble. I actually got an offer from Atari Consumer and Atari Coin-Op. Ed Rothberg and Steve Calfee in Coin-op made the offer and Dennis Koble and Brad Stewart in Consumer made me an offer. Frankly, Coin-Op scared me, because I didn’t have an engineering degree, remember. I was self-taught on the 6502, and I was pretty confident in my abilities, but boy, those guys scared the heck out of me. I went to Consumer because I thought there that I could probably get by there a little bit easier.
I spent the first week at Atari learning the 2600 which was basically incredibly esoteric hardware. The TIA, which was the Television Interface Adaptor, was *the* chip, and that had– There were three chips on the 2600. One was the 6502. They actually used a 6507 because it was cheaper and had fewer pins, and it also meant it couldn’t address as much memory, but that’s another story. The TIA, which basically had all of the television specialized interface and a 6532, which had all the input/outputs. That’s where all the joysticks were attached and all the I/O went through there. The sound was on the TIA so all the custom sound was on the TIA. All the switches, the expert switch and reset switch, and all that went through the 6532.
Can you ask me another question?
[laughter]Kevin: You figured out how to work the very bizarre machine that you had plopped in front of you. What was the first project you worked on there? Was that Video Pinball, then?
Bob Smith: It was. Dennis handed me a manual and handed me a copy of the 6 char, which was Carol Shaw’s code that could display six characters across the screen, which for the 2600 was a very big deal. It changed the 2600. I spent a week learning the 2600 and reading and rereading. I discovered that it usually takes me about three readings of technical documentation before I start to get the drift, and this took me at least six. This took me many, many times before I finally got it. I was reading Carol’s code examples, and actually, I was reading the code from Coin-Op in 6502 because they were so thoroughly educated and well-versed in the 6502 and wrote brilliant code. So I learned a lot of stuff from the Coin-Op people.
After a week of that, I went back to Dennis and said, “Well, okay, I think I cut the basics, let’s try it out here.” He handed me a list, and at the very top of the list was Video Pinball. I said, “Well, I am but an egg, so I’ll take the top one off the list since that’s what you think I should be doing,” so that’s what I did. I spent the next week working it out with graph paper – boy, how much to go into the 6502 architecture… Essentially, we had two different styles of background. One was very blocky and had only 40 pixels across and as many scan lines as you wanted. That was the 6502, so I made part-
Kevin: That was the low-res background graphics.
Bob Smith: Right, low-res background. That’s what I formed the bumpers and the low-res things in Video Pinball. The flippers and the numbers and all those things had to be what they called players, which these days we call sprites, which were essentially eight-wide graphic that could be as tall as you want, which is another 2600 issue. I designed it, and the whole key to the 2600 was vertical separation. There were only two players, there were only two high-resolution objects on the screen, but since you had such control over the electron beam as it scanned the screen, you could actually figure out when this first high-resolution area had finished displaying and reposition it and use it again, and again, and again, and again.
If you look at a brilliant thing like Rick Maurer’s Space Invaders, unbelievable. He reused these players over and over and over again to get all of the marching aliens. That was all reused players and pinball used the same kind of thing. That was a very common technique for later games. I’m sure you know that the 2600 was designed to do Tank. It was designed to display two tanks, a crude background, and two missiles. That was it. Through magic, that was basically what it was, people like Rick Maurer and Carol Shaw, the Fantastic Four, the Activision Four. Kaplan and– Boy, why do the names escape me? Dave Crane, and Al Miller, and Bob Whitehead, were all brilliant, brilliant 2600 programmers.
That’s how I started with Video Pinball.
[laughter]Kevin: Did they give you anything to use as a reference? I know Atari had a video pinball arcade machine and a dedicated system and whatnot.
Bob Smith: Actually, they had a corporate game room. We were in 1262 Borregas at the time, and Corporate was right across the street. We never went to Corporate. Never. Except they had a game room attached to the lobby because they figured, “We’re a game company, we do coin-op games.” They had a Superman pinball down there. They had Lunar Lander and a lot of– Remember this is 1980, so these were the early Atari games. They had a Superman pinball, so I went in and played 8, 10, 12, 15 hours of Superman pinball because I was never a major pinball player.
I understood all the progressions and so on. The interesting part of pinball is working these things out. Get points here, and that adds to this, and that opens this gate and does this and that. I tried to build some of that into Video Pinball, and when I was done actually, I took it to Mike Albaugh, who was a great programmer. He had been a long time with Atari Coin-Op, and he said, “You know, this isn’t pinball, but it plays really well,” and I settled for that.
Kevin: [laughs] I was going to say you did a pretty good job getting a physics engine in there sort of and–
Bob Smith: Uhhh, sure…
[laughster]
Kevin: It bounces around.
Bob Smith: Yes. The entire movement was interesting because there really is no physics engine because the 2600 was on a frame-by-frame basis. It was basically, get your speeds and apply them every frame and add gravity and reflections, and there was no decay as you can tell because it’s so lively. It doesn’t stop bouncing. Probably the hardest part of that was the collision detection. Because the 2600 had hardware collision detection, it could tell when a ball and a sprite were overlayed on one another, but that was about it. The problem with the flippers and the ball was that the ball was going so fast that it could travel and never be presented on top of the flipper, but it needed to collide with the flipper. That all had to be done in software instead of hardware, which was a shame because, like I said, the 2600 had built-in hardware collision detection.
Kevin: Yes, I could see that. Actually, it’s funny. I was messing around with it the other day. I actually had the ball get stuck in the wall briefly, and I thought for a second maybe it had soft locked, but then it popped right back out.
Bob Smith: It wouldn’t surprise me if I had to put what I called a “flecking special case” to get it out of the wall. That was frequently a problem with things like that when they traveled so fast that they would start interfering with the collision detection and start doing a collision on the other side of the surface instead of the surface you want.
Kevin: I think I’ve seen you mention elsewhere that it sold something like 2 million copies.
Bob Smith: Pinball was my best-selling cartridge. It did 2 million, and Mike Moone still owes me a trip to Hawaii, actually. He told me at one of the creative sessions. I’m sure you’ve heard of Atari’s creative sessions. He said, “You know. If Video Pinball hits 2 million, this next creative session is going to be in Hawaii.” It never happened, so he owes me a trip to Hawaii.
Kevin: You’ll hold him to it, I’m sure.
Bob Smith: Yes, you bet. I don’t know. I’m very happy, video games were very, very good to me, gave me more than I ever expected.
Kevin: Is there anything you would have liked to have done differently with that game?
Bob Smith: With Pinball?
Kevin: Yes.
Bob Smith: Actually, no. Pinball was a 4K cartridge, and I probably spent a month scrunching it, trying to get it all to fit. That was another issue. First, I had to make it fit into 4K, and the other issue was, I was doing software collision detect and the way the 2600 worked is you provided the vertical sync. Every 16.687 microseconds, you had to be there to turn on the vertical sync signal or the processor had to be there. You turned on the vertical sync, let it go for three scan lines, turned it off, and then did 192 scan lines until you came around to the top again.
If your processing went over time, the screen would roll, which completely freaked me out the first time I saw it happen on my Pinball, because I was taking too much time to do my software collision detect. I had to come up with a faster way of doing the software collision detect so the screen wouldn’t roll when it went through the worst-case test. Then I had to scrunch it down to 4K. There really wasn’t much more room for more in pinball.
I actually wrote a little sound tool when I was doing that game. Most of the 2600 sounds were algorithmic, just because that was the cheapest way of doing sounds. I actually did a cable-driven sound, which was expensive. It probably cost me a hundred bytes. That was really expensive for a single sound. There wasn’t much I would have added to pinball.
I think the one that I would have really liked a few more weeks on was Star Wars. I did the translation of the Star Wars Arcade for the 2600 but they only gave me five weeks to do it in.
I did the job, it looked good, but I would have loved to have had another couple of weeks. Maybe it was five months. I’d have to remember what it was now. Anyway, I barely made it. It was an 8K cartridge. I don’t know if you remember the arcade. You go through the was it towers first, and then the space battle, and then the gates?
Kevin: Something like that. I mostly remember the trench run.
Bob Smith: The towers, because it was an over-the-shoulder view, everything had to grow really big as you passed it. That was a huge challenge because we had to go into flicker mode. When you couldn’t get enough high-resolution objects on the screen, you had to present them every other frame, and flicker them which would get you more objects on the screen, but it also looked like hell. But I had to do that, and I thought that if I had few more weeks on Star Wars, I could have done a better job on both the gate screen and the tower screen. The space screen against the TIE fighters was pretty easy. That was pretty straight forward. That was pretty much, I had done one or two 2600 games that involved that kind of combat, so that was no big deal. The other two screens took all the time, and they had such large objects they were flickered like mad, and I would have liked just a few more weeks to work on it, to get those a little smoother. That’s the one that I really felt bad about.
Kevin: I actually was going to ask, we can skip ahead and do it real quick, I guess, since we’re talking about Star Wars. I was curious, how did that arrangement come about with Parker Brothers in the first place?
Bob Smith: That was Bruce Davis. Bruce Davis was a lawyer who took over [Imagic] after Bill Grubb left. Bill Grubb was the founding CEO, and then Bruce Davis, who was legal counsel for a while, became the CEO. He worked the deal, and as it turned out, Parker Brothers had, I guess it was five months I had to work on it, had five months before their license expired with Lucas. They had to have it shipped in five months. That was the gating item, that’s why I was in such a hurry to do it. At the time, the 2600 was going out of style, and we weren’t selling the number of cartridges that we were in the past so Bruce was trying to come up with contracts and he came up with an IBM [PC] Jr. contract. I don’t know if you remember the IBM Jr.
Kevin: In passing, yes.
Bob Smith: We actually had a special room at Imagic that had a combination lock on it as specified by IBM. The Peanut they called it, the IBM Jr., was so secret nobody knew about it. It had some cool features. It had a funky keyboard. It finally had some nicer graphics. The original IBM was pretty funky, even the graphic card was only like 320 pixels wide, so it wasn’t much. Go on.
Kevin: Actually, I was going to also note that you have this recurring theme in this period. You did your first-person racing game, then you’ve got Star Wars, and you’ve also got Star Voyager, which was the next one I wanted to ask you about.
Bob Smith: Well, Star Voyager was actually my first cartridge for Imagic. That was karmic, I think was what that was. I decided I could do a double-line res space game, and grow the enemy, and have a starfield and all that stuff. I could bring up a stargate.
The problem was, I tried to do, what did I use? 30 count Close Encounters [theme], I’ve forgotten now, but there was one note I could not get on the 2600. The 2600 was all divide logic. Divide by three, divide by four, et cetera, et cetera. It didn’t have a complete musical scale on it. When you tried to play music, there were some notes that just weren’t there. Star Voyager was a good example of that.
We got to our first CES. In June of ’81, we had split off to form our own company and spent the summer and the fall getting our reverse engineering together because we were leaving Atari, which meant we were going to be sued unless we had a good reverse engineering.
We left Atari, got our dev systems together. We got really nice in-circuit emulators instead of the funky systems that Atari was using, although Atari was in the process of upgrading. By January CES, we had our very first games ready. Rob Fulop had Demon Attack, which won the game of the year. It was a great, great debut for Imagic. Dennis, what did Dennis write? Did he write Atlantis or did he write Trick Shot? I think he wrote Trick Shot first, and then Atlantis next and I wrote Star Voyager. We got to CES, and as it turned out, Alan Miller of Activision had written a first-person, third-person, however you want to see it, space game. He had done it in single-line resolution, so I was thoroughly abashed.
We made way too many Star Voyagers. I remember a couple of CES’s later, some Australians came up to me and said that we had sold them so many Star Voyagers, and they had to eat them for lunch. They were kind of upset. There was nothing I could do about it, I was just the guy, the coder, and had nothing to do with the sales end of it.
By the time they had come to me, the market was collapsing, and people wanted price protection on all the cartridges we had sold them. That was when Imagic was going down the tubes, of course, Atari had already announced they were going down the tubes, so we just followed them down.
Kevin: I guess I was interested– Did you do anything at Atari after Video Pinball or did you start any projects before you left to go to Imagic?
Bob: Actually, what happened was they made me a supervisor even though I didn’t have a degree. I had just finished my very first game, but I was 30. All the other guys were seven years younger than I was, so Rob Fulop, Rick Maurer, the Fantastic Four– All these guys were like 23 and I was 30, and Dennis was the same age. Dennis was running the department at that point and he said, “Be a supervisor. I’m going to go to another department because confidentially, I’m planning a start-up.” and this was the beginning of Imagic. This was at the beginning of 1980. This was back when I– I guess this is probably May or June. This was when I was getting ready to ship Pinball.
I lost my train of thought.
Kevin: I think that covers the question that I have. When you were doing the Star Voyager, did you have any experience before that with the Star Ship or a Star Raiders games that had the same perspective?
Bob: Obviously everybody had seen Star Raiders. That was the one and only application for the 800 that was wonderful. It was a cartridge, right. That was the game for the 800. Later on, they finally let people have the 800 so “Good” Asteroids came out and “Good” Missile Command came out for the 800, but at the time, it was Star Raiders. My first game which I’d sold to Creative Computing was a first-person game with things coming at you, so my first game was in that style, and that was always the goal.
It was nice to do an overhead game. Unfortunately, on the 2600, you had no bitmap. So doing a map-like game, unless you’re going to use this blocky playfield, that was the only game in town. So doing first-person was really what I strove to do, so Moonsweeper was first-person. Honestly, Star Voyager was. I didn’t try to do it for Riddle or Dragon Fire. Those were both second or third-person games.
Kevin: Nevertheless, it was a trend for you. [chuckles]
Bob: It was a trend because at the time we were thinking, “How can we make this the most realistic and the most immersive we can possibly do?” These days, you just hire a shitload of artists and you’ve got more immersibility, if you’re willing to bring up that many objects and that many art objects. At the time, there were really no pixels to spend. There were a couple of high-res areas on the screen that you could multiply so you get vertically separated high-res areas, and there were special modes on the 2600 to get two copies, and three copies.
Kevin: I also did have a question on Star Voyager too. I remember there’s switches between the lasers and the torpedoes, and I was curious if that was just the solution you had to solve that particular problem or if you wanted to make that one harder than the other. What was the thought process there?
Bob: I have to admit, that completely escapes my mind. I’ve got to pass because I haven’t played the game in almost thirty years.
Kevin: That’s fair. [chuckles]
Bob: Sorry. I thought I only had one weapon on there but I could be wrong.
Kevin: For a long time I thought so too, and then I remember one time as a kid I turned the game on and this laser was shooting out, and I didn’t realize the switch had been flipped. I’m like, “What’s going on here?” Anyway, I did want to bring up Riddle of the Sphinx. You worked on that one after Star Voyager?
Bob: That was after Star Voyager. I think I did that before Dragonfire. I think Riddle was second. As I said, I have my resume here.
Kevin: I’ve been trying to track down when all these things came out specifically and it looks like that one came out first so–
Bob: Yes. Star Voyager, Riddle, Dragonfire, and Moonsweeper were the four that I did in that order, so Riddle was second. I wanted to do something different. I had a good idea for the kernel, the screen display code, that would allow me to have these things marching down the screen and have a missile go off in the opposite direction. That was Riddle. I lived in the Rose Garden area in San Jose which was actually– I lived three blocks from the International Rosicrucian Museum. I had an Egyptian museum three blocks from my house-
Kevin: That’s pretty sweet.
Bob: -And I thought, “I’ve got this idea for the kernels. I’ve got the idea for the game and so this is what I’m going to do.” It’s a funny, funny story and funny not, and that is the kernels, the screen codes that I wrote, are almost exactly the screen kernels that were used for River Raid. Which sold millions and millions of copies, and I probably sold 750,000 of Riddles. It was basically a brilliant piece of work but it was a niche work and so it didn’t get a– It wasn’t much of a shoot-em-up as River Raid was and it was lot more slower-paced obviously, but that was Riddle.
Kevin: It’s an interestingly ambitious game, keeping track of items in an inventory and [crosstalk].
Bob: That was the idea. At Atari, before I left Atari– In fact we should talk about this. The last game I did before I left Atari– Dennis Koble– Let’s see. No, wait. That was in Paris. Atari was exploding as far as personnel because the 2600 had taken off and when I first got to Atari, Space Invaders had first been issued. Space Invaders was a smash coin-op hit. That was in the news and we were doing a weekly software meeting at Atari and they would say, “Wow, Space Invaders has sold 500,000 copies. I can’t believe it.”
We come back next week, Space Invaders has sold 700,000 copies. Space Invaders was just going through the roof, so Atari had a lot of cash. They were expanding as fast as possible and this was before I was a supervisor. When I was over there, we are hiring like mad. Long story short, after I finished Video Pinball, I volunteered to move out into the parking lot with some other programmers in trailers because they didn’t have enough office space in the building for us. The trailer was actually cool. I shared my trailer with Carla Meninsky who was doing Battlezone at the time, so we had a cocktail version of Battlezone in our trailer. [chuckles] We got to play Battlezone.
Anyway, at that time, I wanted to do a D&D game. And the 2600, you understand. We’re talking what a strange idea that is. We were playing D&D in my trailer. We had Sandy Maywall. Nick Turner, he had a few aliases, he was a D&D strange person. A brilliant guy of course. Everybody in Atari knew he was smart. He dungeon mastered several times for us and then we started brainstorming a D&D for the 2600. Obviously, we could get 8k memory, which is still not a lot. Just the pictures of the monsters was a story. I started designing the 2600 D&D. About that time, Imagic was ready to go and I was invited to go along with Dennis Koble and Rob Fulop to do a Imagic, so I did. That was the end of the D&D. That was actually the last thing I worked on at Atari.
Then I did some more code– Here’s is another awful story. I did test code for the proportional joystick that they wanted to put on to the 7800 [Actually the Super Stella; Bob got the system names mixed up – Ed.]. The joystick that fit on the 2600 was basically five buttons. One button for each way you push the joystick and one button that you had on your thumb, right? That you pushed with your thumb, so five buttons. The proportional joystick would tell you how far you move the stick in any direction. To do that, they basically had potentiometers, rheostats, something that would give you that proportionality.
The downside was, every time you read them, the reading was slightly different. Even if you held the stick as steady as you could, when you read every frame, you would get a different reading. If you are reading it directly, basically you get a jitter. You wound up having to average the position over several frames, which made the stick very sluggish. The nice thing about the five-button joystick is you got your response immediately and that means in one-sixtieth of a second. Any time you were averaging or using multiple frames, it got sluggish, more sluggish and more sluggish. I wrote the code for the proportional joystick and I wrote the memo that said, “This thing is useless. Kill it.”
I don’t know what they did because I was gone by the time all this came out. Then the next thing I did, this was while I was supervising, was they were doing 7800 and it was all in wire wrap. You know what wire wrap is? Essentially, they put components on a circuit board and have long leads on them and then wrap it with wire. The chip was made into probably a solid tube about two feet wide, about three feet tall, and these were all circuit boards that were all wire-wrapped. Unfortunately, the funny thing about wire wrap is sometimes the connections don’t stay that good, so if it starts misbehaving, you hit it hard with your elbow and that sometimes fixed it.
Anyway, I was doing test code with a wire wrap 7800, and the chip designers had designed this new chip and it was going to replace the 2600. It was called 7800. It was an obvious outgrowth of the 2600. We still had sprites, players, but now a lot of it was in hardware. Every time a sprite finished displaying and caused an interrupt, which would give you control and you could set it up for the next time, reuse and so on. The 2600 had no interrupts.
Anyway, very fancy, and I was writing all the test code and so I said, “We’re going do a baseball game.” I get test code for a baseball game. It had really nice bitmap, a really nice bitmap. It had nice high-res background and it had all these reusable sprites, so you can get nine players on the screen.
Then, I was doing code and I discovered a case that an interrupt would happen, somebody would pick it up or the code would pick it up, and they would not be able to figure out who did it, which would completely just blow things up, completely. [chuckles].
Kevin: Oh, jeez.
Bob: Completely. I wrote the memo that killed that chip and I felt bad about it afterward, but I just said, “Okay, here is what I found and here’s the situation where you can get an interrupt caused by this player display system and being completely in the dark as to how to handle it.” So they killed the chip. We are talking about a year or two of work involving chip designers, and they were not happy campers.
Kevin: I can see why they wouldn’t be, but you know.
Bob: I’ve had projects shot out from under me before and that’s the way it is. Nobody like it but it happens.
Kevin: Yes.
Bob: That’s the end of my Atari story. That’s basically what I did just before I left Atari. Then about that time we did Imagic, so Dennis and Rob and I left and started Imagic.
Kevin: Riddle of the Sphinx, you said was sort of built out of your interest in these things and you wanted to do a D&D style game?
Bob: Well, I finally did a D&D-style game for 2600, and I did it about five years ago. The National Videogame Museum just outside of Dallas. What’s the name of the town?
Kevin: Frisco, I think?
Bob: Frisco, yes. Frisco, Texas. I gave them that cartridge. I said, “If you can sell this for proceeds for the museum, go for it, it’s all yours.” Not my best work. The sounds, let’s face it, I phoned in the sounds. It’s not much better with the sounds. Otherwise, it’s pretty cool. Do you remember Wizardry on Apple?
Kevin: Yes, yes.
Bob: It’s a Wizardry-style display and it’s a ten-level dungeon, so I got my D&D game on the 2600.
Kevin: Mission accomplished. [chuckles].
Bob: It only took me 30 or 40 years.
Kevin: I don’t know if there is anything else on Riddle of the Sphinx that springs to mind that you want to bring up?
Bob: Oh, let’s see. We had a contest with Riddle of the Sphinx.
Kevin: Oh, boy.
Bob: And that was how do you solve it. Finally, somebody from Los Angeles gave us this really nice scroll that he had made that had all the answers on it. We gave him his award at the Rosicrucian Museum, in front of the mummy case. [chuckles] Yes, Riddle was always my cult favorite.
Kevin: Yes, I can see it appealing to like a very specific hardcore type of fan.
Bob: Yes, we had some very interesting entries.
Kevin: Dragonfire, that’s a very different game. I saw you mentioned it on the Portland Game Expo, that it was inspired by Jacks? Like the game?
Bob: Jacks was basically it, yes. My wife is a big Jacks player and I’ve always liked the fact that you toss the ball so you have a limited time, and then you have to do this gestalt. You have to, it’s not like, “I want to get this one and this one and this one,” but, “I want a pattern that will take the whole group away.” Missile Command had much of the same if you ever watch a really good Missile Command player. At Atari, I saw some really good Missile Command players, and it was the same kind of thing. It was kind of like a sweep action, and that’s what I wanted to attain with Dragonfire.
Just you add pressure, the dragon firing fireballs from below, and you needed to quickly find a pattern and sweep through the pattern as efficiently as possible. That was the concept. Later, we added the bridge scene because we felt that just the dragon scene was not quite enough. At Imagic, we always did our games in the open layout, so everybody was free to pick on our games if they so choose. It’s hard to take but you welcome it in the long run.
Kevin: Did you get good feedback on Dragonfire then? Is that why you added the second screen?
Bob: I got good feedback from everything. From the sounds, to the colors, to the action. One thing about 2600 games, they kind of lacked in graphics a lot of the time but their controls were always incredibly responsive, and that was why I always liked it better than the Mattel. The Mattel has this nice bitmap behind it, but the operating system was handling all the button pushes, so you had to wait two to three frames before you got any reactions from the operating system. The 2600, it was like right now. They read every vertical blink. Except for the paddle controls, that’s another story. Most of the time, you read your controllers at the start of every frame so things were always really fast.
Kevin: I guess a little sidebar real quick, did you ever mess around with like the Intellivision or Coleco or any other platforms that Imagic was producing games on?
Bob: Only to help people find bugs. I was really not interested in the Intelli. Do you know anything about the Intellivision?
Kevin: I know some stuff about it, yes.
Bob: Okay, it was a 10-bit GI processor. Ewwww, no thanks.
Kevin: It’s a weird machine.
Bob: Yes, I just stayed the hell away, and I helped with I think the Odyssey. I helped somebody with but I never did an Odyssey cartridge, but I helped some people with some Odyssey issues. I believe it was a 6502 [Odyssey2 used an 8080 – Ed.] so I can do some damage there. Somebody had a strange timing bug that he eventually found, but we all worked on it.
Kevin: Got you. Okay. Do you have any idea how well Dragonfire did?
Bob: Dragonfire was I think about three-quarters of a million.
Kevin: Pretty good.
Bob: Yes, for the fact that it was Imagic and we didn’t have the distribution that Atari did. Atari salesmen at the time were saying, “Gee, if you put an Activision or Imagic cartridge in your 2600, you void the warranty.” They were ruthless.
Kevin: Making up nonsense there.
Bob: Yes. We didn’t have even the distribution that Activision had, but we did okay.
Kevin: I guess, bouncing back, you mentioned Star Voyager you made a lot more than actually sold. Do you have any idea how well that one did on the market?
Bob: That was less than a million, I’m sure. I think Demon Attack I’m sure sold over a million, but none of my cartridges with the Imagic sold over a million.
Kevin: Gotcha.
Bob: They were all less. My best-selling cartridge was Pinball but that was because of the distribution that Atari had.
Kevin: I guess we can bump along to Moonsweeper which, I’ll let you know, that’s one of my personal favorites.
Bob: Oh, that my first 8k cartridge, and I tried to do perspective. Hit or miss because, obviously, the 2600 had no multiply or divide, so you really had to do your projection via a table or some other way.
Kevin: I was going to say, the effect of landing on the planet is very distinctive. Like it just sort of unfolds there.
Bob: Yes, that took me weeks to get that effect. That took a long time but I think it turned out okay.
Kevin: You got a song in there too even?
Bob: Yes, actually, that was the first game I think where somebody else did my music, because I had 8K and there was a young musician who was like a student intern at the time, and I said, “Okay, go ahead, do my music and I’ll give you a cut on my royalties,” and that’s how we did it. He wrote the music, or did I write music code? I forgot. I believe he did. I think he wrote the music code and then he put the tunes in.
Kevin: Do you remember his name off the top of your head?
Bob: It was Neil McKenzie.
Kevin: I might look it up. See if he actually went on to work in anything.
Bob: Actually, some other people that I’ve mentored have done extremely well, which is heartening to see.
Kevin: That’s cool. That was another game that had that perspective that you were talking about being interested in. Did you take any lessons that you’ve learned from like Star Voyager forward on this one, when you were working?
Bob: Always, always, and just little snippets of code that you pick up. Mark Klein, another really smart programmer at Imagic introduced us to phi, P-H-I, which is a really, really interesting mathematical way of approximating motion as it turns out. I’m not going to go into it. I’ve forgotten all the details of it, but it was a magic little piece of code that was incredibly small, to the tune of 12 or 16 bytes to make it work.
Kevin: Is that’s something you used in Moonsweeper, then it sounds like?
Bob: Hell, yes. As soon as we discovered it, we used it in every cartridge afterwards. Anything that’s that efficient gets used.
Kevin: Makes sense.
Bob: If you look at the first few bytes of every 2600 cartridge, you’ll find exactly the same piece of code because it’s the right way of doing it, and that’s the most efficient, fast way of doing it. The fastest way of clearing the 128 bytes of RAM.
Kevin: Okay. After that game, it sounds like you got roped into doing the Star Wars game, which we did talk about somewhere else.
Bob: Imagic was kind of on its way out at that point. We had tried a public offering and got shut down like 24 hours before we were supposed to go public, so there was never any real payoff for any of us, except royalties. I did okay with royalties, but at the time, taxes were such that Uncle Sam helped himself to half, so we kind of got burned out of the big fortune, but that’s all right. We had a good time. It beat a real job.
Kevin: I hear that. What happened with your career after Imagic went down? What did you go on to work in?
Bob: Oh, I stayed in games. Oh, let’s see. Now I’m going to have to go back my resume. Oh, dear, I closed my resume. It can’t be far. Here we are. Let’s see. Oh, yes.
Kevin: I saw, at some point, you were working at 3DO.
Bob: Yes, I was like seven years at 3DO.
Kevin: Almost seems like a lot of the old Atari devs ended up at 3DO.
Bob: That’s another story that you don’t know. Okay, let’s see. Through 1985 I was with Imagic, and then I did my coin-op stint. In ’86, I went to Bally-Sente and they had adopted the Amiga 508 chipset. I don’t know if you remember the Amiga but it had some really, really amazing hardware, the software not so much. They tried to do a multitasking system. They half succeeded, called Intuition as I remember, but the hardware was really, really amazing. Really amazing. It had display lists, meaning– Wow, you could give instructions to the graphic processor as every frame kind of thing. It was very advanced.
Anyway, we used the 508, so I did one game for them. It didn’t test all that well. It’s called Moonquake, and about the time I finished that, I needed surgery so I went off to have my gallbladder taken off. While I was gone, Bally-Sente went out of business so that was the end of Bally-Sente. Then I did independent contracting for a couple of years and I was working for Rob Fulop, who was doing something called Interactive Productions. That was basically doing online games. What I wrote for him was a huge package that took eight-by-eight cards. Are you familiar with card systems?
Basically, you would take a graphics screen, cut it up into eight-by-eight sections, and then make it into a deck of cards that you could put anywhere on the screen.
Kevin: Sort of like HyperCard, are you thinking?
Bob: Sure. Basically you can build screens out of these cards. I wrote code that would take a screen made out of these cards, would edit the card set, would use a card either upside-down, reverse left and right, or normal, and compress the card set so that you could send it over like a 1200-baud modem. It was so slow online that we had to– That’s why you use cards, because then when you had a game, you just ship in the cards and then ship in the card maps. That would tell where each card goes for each screen. It was a very efficient way of doing that.
Then in about 1988, I went to Electronic Arts. There, I wrote software renderers. Right about that time, they were starting to do polygon things and hardware rendering was still in the future. The NVIDIA cards, et cetera, were just a pipe dream at that point. I wrote really fast IBM PC renderers. They would take a 3D scene and hand it to me and I would draw all the triangles. I was probably 10% faster than any code I ever ran it against, but obviously, it was something whose end was nigh. As soon as hardware came out, that was totally passé.
Kevin: But you know in the interim, that’s pretty good.
Bob: Yes. No, no. It was actually amazing code. It was a written in assembly obviously because we wanted speed. I was doing more than triangles. I was doing n-sided polygons so you could have as many sides as you wanted. It wasn’t optimized for triangles the way hardware is these days.
Kevin: That’s pretty cool. Yes.
Bob: Anyway, I also started a game called UFO that I didn’t design and I hated the design. EA decided not to put it out so I just said, “I’m done.” I was living in San Jose and they were up in Foster City and I said, “No, I don’t want to drive this far for this.” I left EA and, of course, I did it like two months before they went public. Good job, Bob. Then I want to Accolade. Accolade was a little company that was formed by Alan Miller, who was one of the Activision guys. They wanted me to work on their graphic adventure engine. You remember graphic adventures?
Kevin: Sure. Yes.
Bob: Larry the Lounge Lizard I think was more of the famous ones. Accolade was working on an adventure engine where they actually had a piece of software that would do all kinds of things with one scene. With the masks so you could walk behind things, and areas that you could walk and trigger things to happen. It was a whole engine that was designed to do this scene-based graphic adventure. Mark Voorsanger, he made a skateboarding game in Atari that was very famous. Something and something, two names. Sorry.
Kevin: Skate or Die, maybe?
Bob: No. That was EA.
Kevin: I’d have to look it up.
Bob: If you look up V-O-O-R. I can’t believe that– He had an unusual name. Let’s see what I can find here.
Interviewer: Mark Voorsanger. Skateboard.
Bob: Voorsanger. Oh, cardiologist. I guess there’s more Voorsangers than I thought. Let me try that again. ToeJam and Earl.
Kevin: Oh. ToeJam and Earl.
Bob: ToeJam and Earl. Mark Voorsanger and Greg Johnson.
Kevin: Yes. That’s a great one.
Bob: That was 1989. Anyway, Mark Voorsanger had written the graphic adventure tool and decided, “Screw this. I’m going to do something else.” I came onboard Accolade to work on the graphic adventure tool. I worked on that for probably a couple years. We added a new point-and-click parser to it. When we first rolled it, it was, “Type this out,” and we have parser that figures out what you’re saying. We went to a context-sensitive point and click parser that was a lot slicker, and that’s about– Then Steve Cartwright did a second game and that was about it for the graphic adventures. They just didn’t sell all that well.
I was working as a tools guy for Accolade and I had brought up a programmer by the name of Russell Shiffer who was into music. At the time, MIDI was just coming into vogue, where a lot of games had a MIDI soundtrack where you could hook it into a Roland synthesizer, or it could generate MIDI events as the sounds go.
Russell was really good and he was really getting to learn coding, so he took my place. He’s actually one of the guys I mentored who did really well. He took over the tools and, about this time, the Super Nintendo came out and I decided that I wanted to go back to game design and away from tools. I said, “Let me play with this Super Nintendo.” Super Nintendo had something called Mode 7 which allowed for scaling graphics. You could scale things really big.
Guess what, I decided I was going to do a first-person space game. [chuckles] I know you’ve never heard this before! You can go from a really tiny spaceship in a distance and blow everything up full screen because of this scaling. There were problems, meaning it was the only item on screen. Everything else had to be a sprite, so all my instruments and so on had to be sprites. The interesting thing we did with that is we raytraced the enemy ships, because polygons were still not mature enough to be able to perform like I wanted them to.
What I could do because I had plenty of ROM, so I raytraced 16 views of an enemy ship and stepped through them according the angle it was to my ship. It actually came off pretty well. The other thing I did is I actually got through Nintendo approval first time. I don’t know if you’ve ever done this, but when you sent to Sega, Nintendo, and all these guys, they always say, “You can’t do this,” and they’d send it back. You’d have to change it before you then resubmit it. This was the only game I’ve ever submitted to a major company like that, that came back first time with no complaints.
Kevin: Wow.
Bob: Anyway, that was WarpSpeed. I think they liked it because I used this Mode 7 and the hardware was still very new, so nobody else was using Mode 7. I was one of the first people to use Mode 7 on the Super Nintendo. As it turns out, this is later in my career, the Super Nintendo hardware is what went into the Gameboy Advance. The Gameboy Advance had exactly the same hardware basically that the Super Nintendo had, Mode 7 and everything.
Anyways, I went to Accolade for a while and then Dennis, who was my boss at Atari said, “We’re going to go work for Universal Studio. We’re forming a new company called Universal Studio’s Digital Arts.” It was Dennis and Roger Hector and a few more people that I worked with before. Roger was going to be the president. This was about the time that PlayStation came out. Now we’re into Sony PlayStation.
I went to Digital Arts for about a year, and finally, I just couldn’t take it anymore. It was too freaking strange. I like Roger. I have a lot in common with Roger, but I’ll never work for Roger again. That’s where it comes down to. Things were just too harum-scarum there.
Then I went to 3DO and I was there for many years. There, I was the supervisor. I worked on the PlayStation again. I worked on Army Men, which was one of their franchises that didn’t do all that well but it was okay.
I did work here and there at 3DO. In 2000, I decided, “To hell with this. I’m leaving,” and I moved to Oregon. Unfortunately, that was right when 9/11 happened so my retirement kind of went– My 401(k) just went to hell as you might imagine. 3DO called me and said, “Listen, we have this project we want you to manage.” I said, “Listen, I just moved to Ashland. My wife doesn’t know anybody here. I’ll do this for you for three months but you have to fly me home every weekend.” That’s what we did. I worked for 3DO for another year or so offline, which I didn’t like. We had a very crude VPN which just made things between that and the version control system– It was a nightmare. Just a nightmare working for 3DO.
I finally left them and went to work for Skyworks which was actually advertisements on flash machines [advergames – Ed.]. I liked the people I was working with. I was working with the Kitchen brothers and I liked them. I didn’t really care for the work I was doing, but I got a lot of Gameboy Advance experience. I’ve done quite a bit of Gameboy Advance. I’ve done bowling. I’ve done World War 2 bombers on Gameboy Advance. ATVs on Gameboy Advance. I’ve done a lot of Gameboy Advance work.
I’ve done more work since then, but the significant thing I did was not for a company. I had found an article somewhere online that said a 320-pixel wide Gameboy Advance actually has three times that number of pixels because to form each colored pixel, they used three dots. To make a long story short, I came up with a really, really good way of showing very high-density texts on the Gameboy Advance because I took advantage of 1200 dots across.
Kevin: That’s pretty impressive. You’re using these little portable screens too.
Bob: I had a very, very good way of showing text. What I did is I put the Bible on the Gameboy Advance and I made all kinds of slick things. You could look for a word. You can do all this stuff. You can set bookmarks and yada, yada, yada. Sent it to Nintendo for approval, sent it back. [laughs] They approved it, but I never got it done. I sold it to somebody in England and it never got produced. That was just another year of wasting time.
Bob: Now I’m working on Arduino. You know what Arduino is?
Kevin: Yes.
Bob: Okay. I’m working on an Arduino project that I’m not talking about. [crosstalk] I think for myself. I retired a long time ago and I always, always maximized my 401(k). I had a wife and family so I was very conservative unlike some of my brethren at Atari, so I came out of it okay. [laughs]
Kevin: That’s good. That’s good. If you ever end up talking about that project in the future, I’d look forward to hearing about it.
Bob: Okay. I think I’ve got you in my Skype contacts. I’d expect you’ll be arrested for it, but that’s cool. Remember, I’m in Portland.
Kevin: I’ll let you know. I’ll keep that in mind. [laughs]
Bob: Maybe you won’t get into politics at all, but let’s just say this is a computer project that I have never seen before. Never. Any more questions for me?
Kevin: No. I think that covers everything. I guess I did want to ask if some of the work cultures of all these different companies you worked with– These people that you work with way in the past– Did any of your past work styles carry forward?
Bob: I think so. It’s like when I was at 3DO, I got manager of the year one year, which I didn’t really think I deserved. I have a very laissez-faire management style and I’ve always agreed with a maxim that says, “Hire the right people and then let them do their jobs. Don’t micromanage. Don’t try to do their job for them. Don’t try to lead them. Hire the right people.” I’ve always tried to hire the right people. Unfortunately, I’ve had to fire a few people, which is never fun for anybody involved but that’s the best way to get the best work I think.
Kevin: Yes. I think I agree with that management style. [laughs]
Bob: Yes, and people enjoy working for me. We always had a good time. Yes, we had to spend a few all-nighters and so on, but generally speaking– Deadlines suck. I spent my whole career working under deadlines and I don’t do that anymore.