Retrotopia Store | FAQ & Tech Support



Classic Gaming Expo
Plaza Hotel, Las Vegas
10:15 am, Sunday, August 10, 2003

DON DAGLOW designed and programmed the award-winning Intellivision cartridge Utopia before becoming a manager and then director of Intellivision programming for Mattel Electronics. He left Mattel in 1983 to become a producer for Electronics Arts, then Broderbund, before forming his own software company, Stormfront Studios.

JOE KING was co-designer of Magic Carousel and Hover Force with programmer Steve Ettinger. He also provided graphics for a number of other Intellivision cartridges.

KAREN NUGENT did the graphics for BurgerTime and a number of other Intellivision games. She recently ended 14 years of work at Walt Disney Productions in the animation department.

KEITH ROBINSON designed and programmed TRON Solar Sailer before becoming a manager at Mattel Electronics. He went on to design Intellivision packaging and instructions for INTV Corporation through the end of 1989. He is now president of Intellivision Productions, Inc.

STEPHEN RONEY designed and programmed the Intellivoice game Space Spartans with Bill Fisher. He and Bill then joined John Sohl on B-17 Bomber. After Mattel Electronics closed in 1984, he and Bill founded Quicksilver Software. Today, Steve is a game developer at Quicksilver as well as Vice President of Software Development for Intellivision Productions, Inc.

JOHN SOHL designed and programmed Astrosmash before starting work on the Intellivoice game B-17 Bomber. He left Mattel Electronics in 1982 to join Activision.

DAVID STIFEL designed and programmed Game Factory for the Intellivision Entertainment Computer System. Since Mattel he has pursued an acting career. He has had featured roles in the recent motion pictures Minority Report and Gods and Generals.

DAVID WARHOL designed and programmed Mind Strike for the Intellivision Entertainment Computer System. He then joined Connie Goldman to complete the game Thunder Castle. After Mattel Electronics closed in 1984, Dave founded Realtime Associates which produced original Intellivision games for INTV Corp. until 1990.

ROBINSON: How many people have heard us speak here before? About half. You've heard us tell a lot of the rosy stories, the great things about working at Mattel. But I remember, [David Stifel] and I had a conversation one time...about how Mattel was willing to kind of exploit the programmers a bit. I think it's another viewpoint worth sharing.

STIFEL: I wouldn't say it was Mattel so much as something I noticed in the industry in general: the idea that, "Well, you love what you're doing, so of course you're going to work all these extra hours without getting paid."

ROBINSON: Toward the end of Game Factory you were pretty much there around the clock.

STIFEL: I wouldn't say 24/7 but at least 16/7.

ROBINSON: There was definitely...I think programmers were put into competition with each other. "This guy's willing to be here all night, this guy's putting in the time..."

WARHOL: I'd like to step up and defend the incompetence of what we were doing at the time. There were no schedules, there were no due dates, there were no development plans, there was no, "Oh, if you get this much done by this day, you can get that much done by that day." You pretty much built it and you played it and you discovered it as you were going. Sure, people would have spent a lot of time - like they still do now - toward the end of a project bringing everything together... But it wasn't really because they said "You know what? I think we're just going to exploit the masses." I think it was pretty much "Hey, let's release some games." "How do we...?" "I don't know, let them deal with it," kind of thing.

ROBINSON: [to Don Daglow] Senior management...would you like to weigh in on that?

DAGLOW: [official voice] Speaking for senior management... [normal voice] I think for the games business in general, it's an ongoing issue. One reason I got in here last night instead of Friday night as I had hoped, is that we're on crunch right now finishing a product that's going to be announced in about 10 days. And so everybody's working late nights. I'm bringing food into the team. At about 9, 9:30 each night I bring in snacks for people - people who are working six and in some cases seven days a week. So the more things change the more they stay the same. And the irony is we still have these dialogs. We have different companies look at it different ways. Because we're a small private company, called Stormfront Studios, what we do is, many of the people have stock in the company. So they have some stake in it. Back at Mattel, none of us had that stock. It was purely work-for-hire. Your salary...

RONEY: At the very end we had two or three shares...

DAGLOW: At the end, you're right. I had three shares.

WARHOL: I still have my three shares!

DAGLOW: I still have my three shares.

WARHOL: That's my retirement plan.

DAGLOW: But the other battle that I fought for two years – Gabriel Baum [VP of Intellivision] and Mike Minkoff [like Don, Director of Intellivision Programming] also – we fought for two years before we actually got people's names on the back of the boxes. It was only, what, the beginning of '83?

ROBINSON: The middle of '83.

DAGLOW: The middle of '83. And we fought that battle with withering memos and meetings and everything else.

SOHL: I was thinking back, I started at Mattel in 1980 and during the next six or seven years I was working in the game industry. And every New Year's Eve I was working on a game, because it had to be done for the Consumer Electronics Show in January. And even after I left Mattel, I kept thinking, "Here it is New Year's Eve again and here it is I gotta get this thing done by tomorrow."

ROBINSON: Don, the people who are willing to do the 24 hours and are there the whole time - don't they tend to get rewarded more? Don't other people see that, gee, the way to get rewarded is give up your life?

DAGLOW: In the way it's evolving on the coasts of the United States right now - we're really here to talk about the old days and we kind of keep bringing up the new days – but the way it's evolving now is that if you're an American developer, you're facing competition from Asia and Eastern Europe and to some extent the UK. And in Canada the video game business is subsidized. So anybody's who's not working flat out as a team basically is not in business against that competition. And you also have to produce work of really superior quality, but you can't charge what it costs to do business in the United States. In terms of the business of 2003, that's the way it is. Whereas in '83, I think it was more a matter of a combination of people not ever having done this before – so kind of figuring it out - and just big corporation [bureaucracy]. What was it, in the first year we had five different offices? Because we were shuffled around in the old warehouse and they'd keep putting up new partitions. We were an afterthought. We were going to go away soon the first year. So there wasn't a lot of planning. They didn't have time for people to form evil thoughts about us. They didn't think we'd be around long enough to have an evil thought about.

ROBINSON: But there was a lot of pressure...Joe and I wound up...I got divorced during Mattel. [To Joe King] You got...

KING: Yep.

ROBINSON: I think the hours certainly affected my marriage. Now, Steve, you managed to get through family life. How did you keep a life going?

RONEY: Well, I don't think I...I never worked 24 hours a day.

ROBINSON: And yet you got more games out than we did! [As if stunned by the idea] I guess...quality is involved somehow; talent is somehow involved in this, too, isn't it?

RONEY: I mean, you know, you'd work 12 or 16 [hours] when you'd need to, but there's a point at which it ceases to be productive. And you just have to kind of know where that is. And sometimes when it's down to the last...[to Keith Robinson] I think you were the one who started the rule that if the deadline was end of business Friday, well, if you haven't gone home, it's still Friday.

ROBINSON: There were two rules: One is that Monday morning is the same thing as Friday evening, because if you promise it Friday night they can't do anything with it till Monday anyway. So if you promise it Friday night you have the weekend to work on it. And the other one was, yes, if you haven't gone to bed yet, it is still the previous day. The argument I always made was, if you watched David Letterman on Tuesday night – it's after midnight - do you say you saw it on Tuesday night or on Wednesday morning? If you'd say Tuesday night, then I'd say, "Well in that case, we've established what you are, we're just haggling over price. It's still Tuesday."

RONEY: Everybody worked the late hours when they needed to. But people who work 16 or 20 hours a day just can't keep it up that long.

ROBINSON: Now, you're still in games...

RONEY: Technically, I'm in games again.

ROBINSON: Right. And Dave, you're in games. But primarily do you see programming as being a young person's [business]?

WARHOL: I've been wondering whether or not the phenomenon of taking younger people who have more time - who don't have family pressures - whether or not the economics are built up now such that that's what you have to do. Or can you have a group of people who can make regular schedules without the kind of permanent commitment to that kind of crunch? Because some game companies still do tend to hire young, where people do have that flexibility. There is a place in LA where a six-day workweek is mandatory. But I believe that now we do know enough about games, after having worked on it this much time, that you can schedule things out. I always remember running into people like David Stifel and a couple other people who moved into other technology areas and they're like, "Oh I'm glad I'm out of games [and] into printer drivers," or whatever it was and it's kind of like, "Well, they still have deadlines and they still work long hours." So...

STIFEL: In the film industry, I can remember any number of times going on set at 6 am and coming home on the bus at midnight. In that industry they've even had to pass regulations based on a guy who fell asleep at the wheel driving home one night.

DAGLOW: Another perspective though, we do get to work on games and the reason we've done it for so long and the reason we did it back then is because we're passionate about games; because we love what we're doing. So a lot of the time, when you're sitting there doing it...when you see the impact on your life, then you sit back and go, "Oh, God, I got to keep perspective." But also it's a labor of love as you're doing it, and that's why it makes it easier to get your life disintegrated - because you love the game so much. You're doing so much you can lose perspective more readily.

SOHL: I believe I started when I was 28. And I don't think anyone in the group was much older that I was...

ROBINSON: Except Don.

DAGLOW: Damn...

ROBINSON: No matter what, he's always older.

SOHL: As I recollect, I was not that long out of college and staying up all night to finish a paper was kind of a thing I got used to when I was in college.

KING: We just really enjoyed what we did. It's 20 years later and this is really the best proof of everything: the fact that we're here today trying to share bits and pieces of that enjoyment with you, who reflect all of that back to us. You know, that the work we did is still appreciated and enjoyed. Even in the midst of whatever the hardships were - of deadlines, or challenges, or the time commitments or constraints - we just had a blast doing what we did. When I first started, Keith had hired me, he was my manager, and I asked this really stupid question. I said, "Where's the time clock?" And he just laughed at me. I had never had a job where there wasn't some type of ogre watching that clock and making sure that you clocked in on time and clocked out late. And he just laughed, he said, "You know what, there is no time clock. Just kind of hang out. Why don't you play games for a couple weeks, just get the feel...?" It was like, "Play games for a couple weeks?!?"

WARHOL: Honestly, you'd wake up in the morning and it was like "Wow, I can't wait to get to work." This was just so great. You looked forward to it.

KING: Absolutely. And the fun. Just the basic fun of what we were doing was probably the most seductive part of it. Because I think it was, without anybody's planning, people were voluntarily hanging out there 60 hours a week, maybe more, because of what Don mentioned: just the passion for what they were doing and just the sheer enjoyment of it. We were doing things that no one had done before and we were given - within the guidelines of whatever our assignments were - a tremendous amount of personal freedom to inject our own creativity. No one had ever given us those kinds of opportunities before. I'm not sure how many of those opportunities are left. It was just a lot of fun.

STIFEL: When I talk about spending hours and hours there, no one was standing over you with a whip. It was all very self-driven and often you were competing against yourself. The ultimate game was making games. That was the ultimate meta-game.

ROBINSON: Right. When I would be tired of working on my game - "Oh man, I just can't work on this thing anymore" – I'd take a break: I'd go into somebody else's cubicle and bug them and try to help them on their game. When we have the credits on the games, that's never one person who worked on it entirely. People were always coming in and out - whether you wanted it or not - helping each other on games. My game, TRON Solar Sailer: Steve worked on it, and Gene Smith worked on it, and Russ Haft was always coming in and coming up with suggestions.

DAGLOW: I got in the way some…

ROBINSON: Don started the game…

DAGLOW: I was saying, I got in the way some and you had to overcome [my contributions].

WARHOL: There's a Steve Roney story that I always remember. There's a programming technique called table-driven programming where you just have lists of all the answers or whatever it is in your code as opposed to trying to compute everything. And I'm standing in a hallway and I'm talking about a problem and Steve says, "You should have a table." And I'm kind of like, "Yeah, right. What's a table?"

NUGENT: Well, I can say the creativity was so strong and so much more fun than working [for Walt Disney Productions]. I worked at Disney in animation for fourteen years and it was so much more fun to work at Mattel. The creativity and the just wanted to go to work. You thought about games driving to work, driving home from was just a lot of fun.

DAGLOW: I think the one negative feeling I remember having, sincerely, was fear. And the fear was that I couldn't keep up with some of the other people at this table. Because some of the people at this table are phenomenal programmers and I was kind of in the hacker tradition – self-taught.

NUGENT: Is that why you went into management?

ROBINSON: That's why I went into management...

DAGLOW: There was a certain amount of violence associated with my being ushered into management. But, you know, I needed the help [with programming]. John Sohl, sitting two seats to my left here, is the one who got me going and saved my butt. I never would have made it without him looking out for me, teaching me so many things I needed to learn. And the fear of not being able to keep up with these guys... I've told this story before, so, sorry to those who've heard it before, but one of my great memories after doing this as my career now for 23 years is walking by at times and seeing different team members – probably most often Steve Roney to my left and Mike Minkoff who's not here – but it was many different people and their battles. Making things fit into 4K - these tiny little cartridges – you spent a lot of your time compressing, not designing or getting great inspiration. If you remember that old game show Name That Tune? I don't know if you ever saw it; you had to hear a couple notes of a song and guess the song? What would happen is these guys would stare at a piece of code, and Steve might look at it and he'd go, "OK, I can make that work in nine decles." Decle was a measure, the equivalent of a byte. And Mike or somebody else would kind of stare at it and rub their chin and say "No, I, I, OK, I know how to do Seven decles. Seven." And then the next guy. And watch it go back and forth. And I'd look at it and I thought secretly to myself - I wouldn't vocalize this - "If I were to do this it'd take 32 decles." These guys are so great. So the one real negative emotion I remember is fear. And it was fear of not being able to keep up.

ROBINSON: One of the great things about John Sohl... He was the first programmer actually hired by Mattel to work on Intellivision. Rick Levine and Mike Minkoff had been doing handhelds and were moved over into programming, but John was our first actual Mattel Intellivision programmer. And he was obsessed with writing down everything as he learned it, because there wasn't great documentation. People at our consulting firm, actually David Rolfe who wrote the operating system, wrote about a forty page [manual], Your Friend the EXEC, how the [Intellivision] was programmed. But John was always writing down everything he discovered. My favorite one, which we posted - if you read the Intellivision newsletter - a couple months ago, was where [John] went through and tried to describe each one of the colors. And it's like, "Blue. The blue is a sky blue, kind of California sky, about 45 degree angle..." a very detailed thing. But all of those notes got printed up and distributed. So half of our documentation was just like [John's] guideposts, like an explorer having gone out first and here's the diary of how to do this stuff.

KING: I just cleared out a storage locker that we've had for about 20 years and I found a copy of that: Your Friend the EXEC, along with…

RONEY: You were supposed to turn that in, Joe.

ROBINSON: [shaking head] Joooooooe...

DAGLOW: Would you come to my office later, please?

KING: Along with...

ROBINSON: He didn't mention we had to put a time clock in just for him.

KING: ...uh...folders on at least a dozen other games I had forgotten I had even had my finger in. But I wanted to emphasize again the collaboration between everybody that Don mentioned. Nobody said they didn't have time to help you with whatever you were working on. Whether it was the programmers streamlining each other's code or the artists contributing on different games. Everybody shared. Everybody contributed to everybody's effort. I think that's another thing that raised the bar on what we accomplished.

RONEY: When Keith was mentioning about the original documentation, the other documentation was about the assembly language. One of the instructions at the very end was listed as, "Well, you know, you're never going to use this but since it's there I really should document it." And I essentially took that as a personal challenge. And I did find a use for that instruction, actually, which was part of the way to save space doing all sorts of funky things with bits. The early games I think were probably able to fit - maybe were able to fit - [but] as we got more and more sophisticated – we got more sophisticated faster than we got larger ROM sizes. Typically if they said you could have an 8K game, when you got to where you thought you were finished, you were probably going to be at 9. If you were lucky. If you were at 12, you were going to have to cut something out. But if you were at 9, there might actually be a chance that we could fit that down into 8 without actually losing that much information. And that's where we would spend a lot of time. John Sohl and Mike Minkoff who Don mentioned and Gene Smith and I were probably the four main people who would try to go in and try to crunch stuff down. You just got to where you had little tricks you could do with the processor and the instructions.

SOHL: Was "DECREMENT R7" the instruction?

RONEY: No, "DECREMENT R7" was not the instruction he said you would never use. "DECREMENT R7," which is actually used at the end of B-17 Bomber, I think...

SOHL: And Astrosmash...

RONEY: A couple of games...R7 is the program counter, so "DECREMENT R7" means just stay...its essentially an infinite-loop instruction. It was not separately documented; it was just decrement a register. The one that was, "you'll never use," was something like, "EXPLODE STATUS WORD" – one of those instructions that takes bits from all over the place and smashes them into a word in a particular format.

STIFEL: I remember Keith had finished TRON Solar Sailer and then he showed up for work the next day wearing Mickey Mouse ears saying, "I'm into decle cutting now. I have to do the decle cutting." For a couple of days, Keith was in a very bad mood, just wearing his Mickey Mouse ears.

ROBINSON: Yeah, the creative part is done, so now it's time for the technical: how to get this game which is now 13K [down to 12K].

WARHOL: I had a technique: when I was designing code it would be all sorts of notes and I would just copy it onto another piece of paper and consolidate it. And I would just keep copying it and shrinking and shrinking it. And when I finally got a clean copy then I would just sit there and type it in. So that's how I managed to keep it refined and [to size].

ROBINSON: Well, my stuff was a whole lot sloppier, which is why Mike Minkoff put Gene Smith on my project: to help me clean up the stuff... When I was cutting decles, I was just, "There's nothing else to cut!" I'm looking through the graphics...

RONEY: You can always make it smaller, Keith.

ROBINSON: Yeah, that was one of the rules: You can always cut another decle. Which made Mike Minkoff think there must be some single command which is essentially "PLAY ASTROSMASH" or "PLAY BOWLING." If you could just find it... My game was TRON Solar Sailer and the Solar Sailer is going along these [beams]. In the movie, the MCP sends out this pulse along the [beam] and tries to blow up the ship. And I had that in the game. And it had to come out. But I had never taken out the graphics for it! So here I am trying to cut, trying to cut, and there's no space left, and one day I'm going through the graphics and here's all this definition of this animated pulse and I'm going "I haven't used that in three months!" And I pull that out and suddenly everything fits! So it was just a case of not keeping track of what was in my own game.

SOHL: I did B-17 Bomber which was intended to be one of the first voice games...

RONEY: It was one of the first voice games.

SOHL: Well, yeah, it was. OK, and it was supposed to be 4K of programming and 4K of voice code. And 4K of program was the standard size for everything. So it was going to be a double-size cartridge to start with. And as the game went along, I kept adding stuff. I figured we needed a map of Europe. And you had to have interceptors... So the program code kept getting bigger and bigger. And the people who were recording the voice had some difficulty squeezing it down. So we ended up reducing the quality of the voice considerably. It actually did sound like a human voice when we started. And we also cut out a lot of the phrases that we had. We had things like, "Takeoff," "Landing," "Wheels up!" - stuff like that. We didn't have space for it. [B-17 Bomber] did go out as a 12K game; it was essentially 8K of code and 4K of voice. Well, it took me a year, and it took Steve Roney and Bill Fisher about six months toward the end of it because it was clear just about...New Year's Eve! I remember going in New Year's Eve and Gabriel Baum came in and we were the only two people in the factory and he said, "Is this going to go out? I mean, can we show this?" I think it was obvious...that I wasn't going to be finished for about another year. So he took his two best people and put them on the project, and thank God for that because it did get finished.

WARHOL: An anecdote about how big these games are. 4K, 8K were the size of the games. Pick any sound effect in any game that's been released in the last four years and that sound effect alone is much larger than an Intellivision game. As a matter of fact, we're working on an Intellivision emulator wherein I sample a square wave and that sample of the square wave of the Intellivision making a sound is larger than any individual Intellivision game itself.

RONEY: When we were first doing Intellivision Lives!, Keith and I decided that every single Intellivision game ever written would fit on a 3 1/4 inch floppy.

NUGENT: And colors. You would have basically one square inch and you could only have two colors in that area. So there's quite a bit of difference today.

WARHOL: The hardware only supported two colors, but the graphics artist would know about the bleed between the different colors - if you put a bright red next to black you might get a little band of purple in between them - and they would actually use tricks like that to get even more colors out of the hardware.

DAGLOW: I'm not sure who did it, but I think somewhere in '82, somebody figured out how to control the interrupts in such a fashion that you could stack sprites and you could create four colors. Do you guys remember who actually figured out how [to do that]?

WARHOL: Sounds like Newstadt.

ROBINSON: I think it was Bob Newstadt and Karl Morris, one of the artists, who did that together. And released this memo that said, "Think you're out of colors? Here's some more."

DAGLOW: Before that it wouldn't work because what would happen is, the underlying sprites would get updated at different times and so the overlay would get broken up. Once they did that, then of course Marketing sees this and says "OK, this is new technology!" So they started labeling the games – what was it? – "Super Graphics!" They implied that something was different about the cartridges, that they were being manufactured differently somehow. In reality, we just figured out how to stack the things. They wanted America to think it was new hardware.

ROBINSON: We have some questions, I think, before we get too technical...Yes?

[Nance Crawford, David Stifel's wife, stands up in the audience.]

ROBINSON: Wait! Ringer!

NANCE: No, actually a friend who could not be here asked me to ask this question and then I got to thinking about it and not knowing anything at all about programming...

ROBINSON: "Mr. Stifel, when will your movie be out on DVD?"

NANCE: ...I came up with a dumb question or two of my own. The basic question is for Don, and it's kind of a long one, and then there's a panel question. How did you recruit game programmers when nobody had ever done it before? Was there a specific recruitment policy at Mattel to find people of varied backgrounds rather than the stereotypical scientific type of personality, and if there was a policy, how do you think it worked – how well did it work? And then farther on for the entire panel, do you think an eclectic recruitment policy would work today?

DAGLOW: Yes, it was an eclectic policy. We realized we needed a diverse range of talents. If I look just to the people to the left, some of these folks I had the pleasure and the honor to have hired. You find tremendous musical ability in some cases. In all of them you find a depth of knowledge of the world that exceeds just their own specialty. You get any of these people one-on-one and start talking to them, you're not just talking about video games, you're not just talking about art or engineering - there's a wide range of topics upon which they can discuss knowledgeably and they're interested. So, yeah, if you want to call it an eclectic range, I do think that showed up in the quality of the products. I think that policy probably predated me, because the five of us who were there the day they made it a department, certainly we were already an eclectic group. I still use that policy in the company I run today. I still deliberately use that policy; I still deliberately seek diversity.

ROBINSON: Can I point out one other thing about that, though? An inadvertent thing that led to that was when Mattel started hiring people for Electronics, they weren't sure this was going anyplace; they hired on the cheap. So the top scientific, really hot experienced programmers... Mattel couldn't afford. Steve, you were an experienced person, you had to come in at a cut or...

RONEY: They were after...I saw an ad, I think it was one to three years microprocessor experience. Well, OK, I had six years experience and a Master's degree and none of it was really on microprocessors so I thought I might have been in trouble, and I was. And I had TOP SECRET clearance and all sorts of stuff that they didn't care about. So I crossed out the TOP SECRET clearance on my resume and wrote that I had a hundred board games. I sent in the resume and I called up [Mattel] every couple of days... After about two or three weeks I finally got an interview. Mike Minkoff interviewed me and he gave me this little test they called the "Minkoff Measure," which was essentially reading some assembly language and telling them what it did. Now, it happened that the Intellivision assembly language really looked a lot like a PDP-11, which is what I actually had been working on immediately before that, luckily for me. So I think I had a much easier time than if I had been a 6502 program like they were looking for. So I did very well on that and [Mike] was all bummed out because he said he'd like to hire me but he wasn't sure he could get me...because I made too much money. I ended up getting an offer without even a second interview, which surprised me... But I did go for essentially no raise. But that's kind of the way games is. When we shut down from Mattel Electronics, Bill Fisher and I started up Quicksilver Software. We weren't able to get work right away; I had to go get a real job, because I had a wife and two kids so I needed to make some money. So I went and got a job and it was basically only five or six years ago that I could finally afford to work for what [Bill] could afford to pay me. But working in games, because it is games, you can get people to work for less. It's like, my daughter tried to go into sports management and she tried to work for a baseball team, but they won't pay you anything at all because a lot of people want to work there. Games are the same way.

ROBINSON: I think that the base salary [at Mattel] was about $18,000. That was what they were trying to get people in for. You couldn't get experienced people [for that], so they weren't hiring the experienced people. The people that did come in, like Don was saying, were people who had other interests and had learned programming on the side someplace. Andy Sells had been working as the piano bar singer at Tony Roma's in Santa Monica when his wife got pregnant. He said, "I gotta get a real job someplace," and somebody said, "Why don't you go over to the Computer Learning Center and take a course for a few weeks?" That was his background. A lot of people came in with these six-month certificate programs at the time. Now what happened a year or two later when Mattel Electronics was suddenly making $200 million in 1982, Management came back and said "Oh, we can't have this entire company relying on people who don't have experience and are just out of college and who've never worked before!" So they said, "You people have got to start hiring people with more experience and we'll start giving you more money." So in [early 1983] a lot of programmers came in who suddenly were these very experienced programmers and were coming in at salaries like $35,000. And it caused some resentment from the people who had been there longer and who had worked on games and were more creative. These other guys came in with more experience and a larger salary but didn't know what they were doing. And I know that probably caused problems for you, Don, having to explain to a few people about that. I know it caused problems for me having to explain to a few people about that. But then it all came home to roost when they started laying off people and the people who were laid off first were the $35,000 people who wound up not programming anything. And the people who stayed around and managed to get the raises were the people who had the more creative backgrounds – who had come from law or music or Eddie Dombrower who had studied with the Joffrey Ballet, and these other people who had just picked up some programming on the side.

WARHOL: The last good response to Nance's question for me would be, as far as diversity goes, my final interview at Mattel was with Gabriel Baum, vice president of the division. For about five minutes, we talked about my qualifications. For about 10 minutes, we talked about what it was like to work at Mattel. And for about a half-hour we talked about the symphony in Europe from 1920 to 1950 or something like that. It's getting a sense of communication skills, getting a sense of being able to hold linear thoughts together. So really, pretty much today when I interview people, it's almost like the last thing I talk about is what's on their resume and what they're doing. It has to do with, can you make a connection with the person, can you hold a conversation with them, and that kind of thing. I do see that that technique was used and I do think it worked very well.

KING: I agree. And it worked both ways. From people like Steve with TOP SECRET clearances in aerospace and engineering coming to the company to uh...I was at the other end of the spectrum. My qualifications amounted to picking up trash at the beach with a pointy stick. But at the end, when I left Mattel, I ended up with a TOP SECRET clearance at Lockheed. So, again, the opportunities came full circle.

ROBINSON: I interviewed Joe, and before he even sat down, his first question to me was, "Is it as much fun to work here as I think it is?" It's the enthusiasm. I know, Dave, that's the first thing you always say: it's the enthusiasm for games, it's the enthusiasm for doing the job. That is almost more important than anything else.

FROM THE AUDIENCE: Were any of you involved with the Intellivision III or the Intellivision IV or can you comment on any of the hardware that was in development at Mattel?

DAGLOW: The Intellivision III was basically an Amiga three years before the Amiga came out. I'm over-generalizing a bit but basically that's what it was…

ROBINSON: Just to clarify things for people who have been to the web site, that was the Intellivision III but it later was renamed the Intellivision IV

DAGLOW: OK, Intellivision IV, I'm still...the Intellivision III...

ROBINSON: The Intellivision III was something that was shoehorned in to try to compete with ColecoVision.

DAGLOW: So what we now know as the Intellivision IV but back then called the Intellivision III was an Amiga-like machine. The industry was beginning to fade, we knew we needed something, so basically they brought it to us. It was already well-formed before we were ever told about it. In terms of hardware design, one of the regular patterns at Mattel was that they would design the hardware before they would ever ask any of us for input about what would make it be a great game machine. There was this east is east, west is west and never the twain shall meet [philosophy].

FROM THE AUDIENCE: Just like today?

DAGLOW: Actually, Sony and Microsoft get better input from us now as third parties than Mattel did when I was Director of Intellivision Game Design. But they brought it to us and we immediately took some of our best folks and set up an initial range of games. We also had to preserve people on the regular Intellivision games, too, so we couldn't take all of our strength and put it there.

ROBINSON: The department that was developing Intellivision IV was physically not on the Mattel site, it was actually down Rosecrans [Avenue] from where we were. Dave Chandler's group, the man who was behind the engineering of the original Intellivision and hand controllers, he was doing the Intellivision IV, which as Don said was a 68000-based machine. But they needed artists to put up some graphics to show what it could do, so that's when they first came over to our department and that's when Joe and Daisy...the two of you did some screens.

KING: Yeah, Daisy Nguyen. One of the things we found recently opening up the storage facility was a lot of the development art and screen shots for the Intellivison III or IV...

ROBINSON: We've posted a couple of photographs of the screens, but one thing that's nice that it turned out Joe had was the documentation on how to do the graphics for the IV. Now the III, like I said, was shoehorned in. What happened was, Mattel shot itself in the foot by doing all of its advertising saying, "the reason Intellivision is better than Atari is because the graphics are better." Never advertising anything else except, "the graphics are better," "the graphics are better," "the graphics are better." Now ColecoVision comes out and has twice the resolution and Mattel just has to stand there and go "Uh...well...there is game play, too, you know..." So they tried to figure out how can we compete against ColecoVision? What Atari tried to do...they came out with the 5200 and immediately got slammed by everyone who had bought 2600s, because none of the cartridges would play in the 5200. So Marketing said, we need to have something that will compete with the ColecoVision, but will still play the original Intellivision cartridges to make everybody happy. So the idea, which was fairly simple...the Intellivision has a chip which takes care of the graphics, so they said, "Well, we'll just have General Instruments design a better graphics chip, but still use the same 1610 processor. That way the same games will work and this'll be great!" And General Instruments put that together and everything was going to be fine, but then they said, "You know, Intellivoice isn't selling so well, so why don't you put the Intellivoice unit in there also?" And the people who had been working with Intellivoice had a little clip on the top that if you removed a plate you could plug stuff in. That was for extra hand controllers. They had been developing wireless hand controllers, and they said, "Let's put that into the Intellivison III also." And then, "It's going to need new Executive software and you know the programmers use scrolling all the time so let's put that in," and then, "We can do things faster than updating every third cycle..." so the development time kept getting pushed out as they kept putting more stuff in. Had they simply updated that video chip and released something they probably could have gotten out and competed with Coleco, but it took them a year and by the time it [was shown] all the reviewers said "Too late. Too little, too late." Karen, did you do some of the Intellivison III stuff?

NUGENT: Yeah, all secret stuff probably nobody knows about.

ROBINSON: They did do one brochure that we printed some of on our web site and some, I think, on Intellivision Lives! With a couple of the screens that you did. Essentially what they had to do was take a [regular] Intellivision and design the screen as four Intellivision screens. In the brochure they would take these four Intellivision screens and moosh them together to make it look like it was actually one [double-resolution] Intellivison III screen.

FROM THE AUDIENCE: Even one second of sampled sound is gigantic, so how did you fit the voice samples for an Intellivoice game into 4K?

RONEY: It's not sampled, it's synthesized…

FROM THE AUDIENCE: But normally synthesis relies on a generic set of recorded phonemes, but you had male voices, female voices, personality…

SOHL: The original voice chip was using a technology at the time which they called "formant," which used certain frequencies and you had a filter involved in it, so it used compression technologies that were specialized for the human voice. The problem with that is, when you reduce the quality the quality gets really bad. So, it was pretty efficient for encoding voice but even so there just wasn't enough ROM to do it right.

WARHOL: They did use real samples. They cast actual actors and actresses and it blew me away that they would go through so much trouble to get such name talent. Somebody from Firesign Theatre was hired to record a few words like "Answer the phone," and stuff like that. But it still took the actual sample and broke it up into smaller pieces and then it used that to feed the synthesis module rather than to just play it back as a sample.

ROBINSON: Yeah, the chip itself models the human vocal cords and the whole larynx and everything. It actually has I think 17 or 19 different formulas that it uses to figure out how sound is created so the synthesizing equipment that we bought from General Instruments takes the recorded sample and then breaks it down so that each one of these models will recreate that sound. And it's not automatic. We had a team of voice specialists; linguists who had no computer background at all who had to come in and they would actually write out the dialog for each game in this universal phonetic language. Because they would then take each sound and say, well a vowel takes less information to reproduce than a consonant so we can use a higher sampling rate during a consonant and we can cut back [during the vowel]. So they could literally go through and have four or five sampling rates in one word. And then after they would do this, they would then bring people in and they would have tests. Can they understand the words? If they just hear it that's one thing. If it's put in context that's another...

WARHOL: You should stop that story right there.

ROBISNON: If you want to read the rest of that story it's on the web site.

SOHL: I think it's amazing how much effort they put into it. They probably had ten people downstairs working on it. The programmers – game designers – were upstairs. Downstairs were a bunch of systems folk. I never quite knew what they did, but that's where the sound people were. They spent thousands. They had specialized linguists. It's a pity that Intellivoice died so soon. I think it was doomed from the start because it fragmented the market. Because there were 3 million Intellivisions and 10% of the people bought Intellivoice, well, then your market is suddenly 300,000. So if your game's not that good, you end up selling 30,000 instead of selling 300,000 for [a mediocre] non-voice game…

DAGLOW: In fact, one of the great fights at the senior management level was should all games have Intellivoice added on as a bonus? Or should there be only voice in Intellivoice games and if you don't buy an Intellivoice it won't talk to you? And, unfortunately, I think the wrong decision got made. I'm still...not that I'm still mad about an issue that I got into a big fight about what? 21...22 years ago, but I'm still ticked about this. Because if we had been allowed to put voice in all of our games, instead of only putting them in the Intellivoice games, which were specialized...

RONEY: And not make them so they were not playable without the voice...

DAGLOW: Right, you were restricting your game design: "You must design a game so that without the voice it won't go." As a game designer, OK, you've just taken my palette of how I can design and cut it by 90%. So another great battle...

ROBINSON: The only game they released [with voice as a bonus] was the baseball game that Don actually first created where the idea was to make it a TV-type view...

DAGLOW: Developed by Eddie Dombrower...

ROBINSON: It's the only game that's voice enhanced. By having the play-by-play announcer – you don't need the Intellivoice, but it's sure great to have it on there...

RONEY: But you do need the ECS.

ROBINSON: could have added a lot to a lot of games. One last question...

FROM THE AUDIENCE: You mentioned senior management decisions. When you read the Blue Sky Rangers history, I know its all hindsight, but what jumps out at you is that some of these decisions by senior management just seem so bad...

WARHOL: We were years ahead of our time in terms of making bad management decisions!

ROBINSON: If we were making games today, Marketing would come in and have licensed the rights to the movie Gigli.

FROM THE AUDIENCE: ...well I was wondering about the reluctance to use arcade games. Because ColecoVision came in and started doing arcade games. Was there a decision, "Let's never do arcade games"? BurgerTime was great, of course, and it was very successful, but I was just wondering why more weren't done sooner?

ROBINSON: That was just money. That was just what Mattel was willing to pay...

DAGLOW: And they beat us to them. The licenses were mostly gone before we got to them.

ROBINSON: We tried. Mattel got in there and tried to fight for some of the licenses; we just lost. We didn't put up enough money; we didn't fight. See, the thing is, that was a way for any company to get into the market. Parker Brothers or SEGA who came in late into the home market, that was a way for them to get in there: just drop a whole lot of money on a license. No money left over to do actually a good game with it...

RONEY: We were too busy throwing our money at Hanna Barbera.

DAGLOW: One example, though, is we did have a shot at Gottlieb – they did a deal very late in the video game era. That deal turned out to include Q*bert. Which Coleco, I think, got.

RONEY: Parker Brothers.

DAGLOW: Parker Brothers got. I was the one who went and met with them and did the eval on that. They didn't show us Q*bert. They showed us two other games that were nowhere near as good. And I came back and said to Gabriel, "Mmmm, there's nothing great there that they showed me." It was a package deal that included Q*bert but I had never seen it. A year later I felt like a sap, but they hadn't disclosed the game to us.

ROBINSON: We just lucked out with BurgerTime. Because what happened was we had made a deal with Data East, who didn't really have great distribution in the United States...but when BurgerTime came out, Midway licensed it and put it into every arcade. That was just a case where we lucked out in terms of licensing a hit.

WARHOL: As far as management decisions, if you figure it was novel for us, it was also novel for them. It was a new, emerging industry and they wouldn't really know to what degree of importance certain factors would be or what consumers wanted. So they were probably doing a lot of the same kind of experimenting that we were.

ROBINSON: But as Gabriel Baum said - he didn't like the marketing deals that were made - his statement was [impersonating Gabriel's British accent], "I don't mind licensing...but our Marketing department goes into every meeting with their pants pulled down, walking in backwards."

RONEY: That'll teach Gabriel not to come to these things.


©Intellivision Productions, Inc.