Bruce de Graaf interview

By Scott Stilphen

I peer behind the curtain and bust the myths of Mythicon with their only game designer, Bruce de Graaf.


Q: What’s your educational background?

Bruce de Graaf: BSEE and BSCS (double major) from UC Berkeley 1974.

Q: Were there any programmers or games that inspired you?

Bruce de Graaf: I got a big kick out of the text-based Star Trek game being played by folks at HP.  I put a version together for VM/370 and it gained some notoriety within IBM itself ("An IBM 'Pogrom' Product").  I was told that in one audit, the application accounted for some significant percentage of all, worldwide IBM cycles).  It was never commercially released - it's somewhere in IBM's archives (maybe). IBM made everybody sign a "Whatever you do is OURS!" document.

Q: Did you work for anyone prior to Mythicon (after school)?

Bruce de Graaf: IBM.  I liked PL/I and annoyed the company enough so that when I graduated, they decided to get revenge and interviewed me.  Maybe it was that infamy (from my Star Trek game) that prompted the call from Mythicon.

Q: What was the development process at Mythicon like?  Were you the only programmer there?

Bruce de Graaf: Initially, I was the only one.  In an attempt to be "clean," I was on my own to decipher the 2600's workings.  The development process consisted of assembling 6500 code on a TRS-80, UV erasing a 4K EPROM, blasting the image to that DIP, transferring it to a ZIF socket on a hacked-up 2600 cartridge, and seeing what happened.

Q: What were the hours in your department?  Was your schedule fixed (i.e. 9 to 5) or could you basically come and go as you pleased?

Bruce de Graaf: Are you kidding?  The initial discovery consisted of weeks of 24 hour days!  I could come and go as I pleased … if I didn't want to get anything done.  The worst part was that the air conditioning in the building went off at 8PM.

Q: Was the Atari 2600 system the only one you worked on?

Bruce de Graaf: For Mythicon and for games, yes.  I've worked on everything from the 8008 (yeah, not the 8080) through PowerPC, x86, and ARMs.  I've always liked the 6500 architecture, and that's certainly a factor in my favoring the ARM.

The big deal for Mythicon was the use of chip-on-board (COB) manufacturing technology; it certainly wasn't the games.  Mythicon had a manufacturing cost advantage (less than $1 each, whereas Atari’s cost $13 each) that meant that it could offer cartridges at retail prices that were below the simple manufacturing costs of anybody else.  Also, due to very sharp work by the sales and finance people, Mythicon was able to sell through outlets (even grocery stores) other than high-margin, department stores.  I cobbled together a game (Star Fox) and, later, we decided what it should be called.

Q: I know that Atari started using the chip-on-board process around 1982 (by then, most of their carts were 8K in size), but far be it for them to pass on the cost savings to consumers.

Bruce de Graaf: I have often mused how in-cartridge intelligence could have enhanced the 2600.  A technique known as "jamming" was used to simply override some of the 2600's signals and thereby change the display characteristics, but jamming has implications on the life of the console.  Some people were able to code for six repetitions of two players (16 pixels) and two missiles (2 pixels) for a total of over a hundred pixels per scan – no pattern replacements, just redisplays.  If in-cartridge circuitry could replace the patterns, that would pave the way for some kind of BASIC-like programming of games and other applications.  Yes, this is putting "lipstick on a pig," but it would have opened the gates to the home developer.  Just imagine – we could have been Apple!

Q: What was the easiest/hardest part of designing it?

Bruce de Graaf: Designing is easy!  Cramming something playable into 4K and delivering in a short period of time is where the real art comes in.  The inspiration for it was, I had to blow up something, and I kind of liked lasers… but Star Fox ended up being a very long way from my original thoughts.

Sorcerer and Fire Fly were essentially the same code base with different graphics.  The first test versions appeared to shoot bombs out of a flying horse's butt - had to make some quick rewrites there.

By the way, the Happy Face mode was not some special test mode; it was a non-scoring mode that reduced the need to lean forward and reset the machine at the end of each game - a feature that seemed useful after 15 or 20 beers.

Q: Was Bill Bryner (who did the sounds for all 3) the only other person you worked with?

Bruce de Graaf: Poor Bill … he had a great melody all ready to go, but then he found out how many (few) bytes he could actually use.  Another fellow (Thorston "Bradshear"? – The Mighty Thor), joined later.  He too was appalled at how little memory was available.  Thor was a "bit-pusher" (hired to design and implement games), like myself.  He was a very nice person and quick to pick up the details.

Q: I know of at least 2 games that were planned but never came out – Loot and Star Fox II.  If I understand correctly, Star Fox II was to be 2 games, with Loot as the second part?

Bruce de Graaf: Yep, that was the maze game that could be attached to Star Fox or just hard-coded into the cartridge (never got to the point where we made the decision).  Star Fox II was much better code (than Star Fox) and it simulated a 3-D experience - your viewpoint was some distance behind and above the craft.  I wanted a game that would allow you to play starship-fighter and then land on a planet and go 'rogue-ing and looting'.  The idea was to have two games in two cartridges.  The SF-II cartridge would have a plug on its back that could accept the Loot game (sort of a cross between Adventure and Rogue) or the Loot game could be used by itself.  The hardware was little more than a latch.  It turns out that the hardware for bank-switching 8K games would have resulted in both SF-II and Loot being on the same ROM die and being even cheaper.  I did have a neat state counter algorithm that could store the layout of the Loot maze and the location of treasures and monsters and stuff in only 3 bytes.  The 8K ROM was a pretty simple idea: address a particular location in order to switch part or even all of the 4K ROM image to another ROM.  I had a prototype of a very tiny part of the 3-D Star Fox: atmospheric flight just before landing (and switching to the maze game) or just before space.

Q: Your idea of connecting two cartridges sounds similar to what Sega did years later with their "lock-on technology" on the Genesis game Sonic & Knuckles.  With your games, were there any features you would have liked to added, or any known bugs or glitches that gave you trouble (or never got resolved)?

Bruce de Graaf: Infinitely long list follows:

• Star Fox was supposed to be 3-D!

• Landing on a planet would switch to a maze-game: Loot.

• I still didn't know about the 6500 push stack of the flag trick

This was a way of comparing the scan line index to a reference and pushing the flags onto the stack (which just happened to be pointing to one of the 2600's control registers) – nearly halved the number of cycles necessary to decide whether or not to show a "player" or "missile".  Because of that, Star Fox's players are coarse and blocky.

Q: Do you remember what early or tentative titles the games had (if any)?

Bruce de Graaf: Other than Loot, there were no other names.  As I said, we named them when they looked like they worked well enough to flog for retail.

Q: A CES company press release stated there would be 8K versions of all 3 games.  Was this actually planned, or was Star Fox II the only game to be ‘upgraded’?  Were there any games or projects that you worked on that ultimately never got released or even finished?  And what of the plans to release games for other platforms?

Bruce de Graaf: Nope, we ran out of money.  Well, not exactly… we HAD funds in the pipeline (sold games for which we hadn't yet been paid).  We needed a line of credit to tide us over until some revenue could appear.  Various banks declared the doom of the game industry – even with money slated to arrive, we were not approved.  All of us had to find honest work in order to eat.  I tried to work both jobs but just couldn't do-the-time (very apt phrase).

We examined the TI platform and had some code (in comparison to the 2600, the TI-9900 was unlimited).  One buyer sniffed us but didn't bite.

Artwork for Star Fox, Sorcerer, and Fire Fly. 

Q: Do you recall who did all the artwork for the game’s packaging?

Bruce de Graaf: I don't know.  I always thought it was "Tammy and Tina" of TNT Graphics.

Q: Are there Easter eggs in any of your titles?  Do you recall any fellow co-workers that did?

Bruce de Graaf: I tried, but there was just no room.  To give you an idea of the crunch, I scanned the Star Fox ROM image to find patterns suitable for explosions; there were no explosion graphics, just areas of the code that looked close enough.

Q: Granted, there aren’t many fans of the Mythicon games (and to be fair, that was the first programming you had ever done on the system), but if you had a chance to redo any of your games (what would you change (if anything)?

Bruce de Graaf: I would never again try to reverse-engineer without a logic analyzer!  That Star Fox/Loot game might have been nice.  To be honest, even then, I had no idea of how to even get the game into 8K.

Q: Did you ever attend any industry shows, such as CES or Toy Fair?

Bruce de Graaf: No, they strike me as hyperactive depressing affairs – depressing when you consider the desperation of the vendors, and hyperactive when you consider the actual signal-to-noise ratio.  I've been other trade shows and helped hammer booths, kiosks and even hardware together (yes, "hammer").

Q: What were some of your experiences working for Mythicon?  Any stories or anecdotes from those days that you recall?

Bruce de Graaf: Nearly crashing a pick up truck when its brakes failed.  The highs when things looked like they just might work and the depths when it became obvious that they wouldn't.

Q: Since Mythicon, have you stayed within the field of game design?

Bruce de Graaf: Good God, no!  I don't even play games.  Game development feels like drinking triple shots of espresso and then running 220 through your teeth.  No way!

Q: Can you describe your career, between then and now, and where you’re currently working?

Bruce de Graaf: A whole range of software development, from intelligent oil well drill heads to space engine controls, from horizontally micro-coded controllers through arrays of supercomputers.

Q: Do you still own any of your games for these systems, either as a keepsake, or to show friends or family?

Bruce de Graaf: I had some items but, years ago, a burglar made off with them.  He/she/it got a pair of skis with dangerous bindings, too – hope they broke his knees.  Any other related material is gone, mercifully so.

Q: Which of your titles are your favorite, and what types of games in general?

Bruce de Graaf: The game I liked the best was my version of Star Trek (written in a very small subset of ANS FORTRAN-IV).  In contrast, the 2600 games bring back memories of gritted teeth.

Q: Have you stayed in touch with any of your former co-workers, such as Bill, VP Dave Dimmick, or company president Larry Jones?

Bruce de Graaf: No.  I do remember Larry as the funniest one-liner generator I have ever met.  I feel that all of the Mythicon folks were superb.

Q: Approximately how many people were employed at Mythicon?  Do you recall the names of any other people who worked there (sales, marketing, finance, etc.)?

Bruce de Graaf: Whew!  This is like archeology!

Dave – Marketing
Larry – Sales and President
– CEO or CFO
Pam and her sister Brenda – "Den Mothers" for us lunatics
Bill – Sounds
Thor – S/W
Tammy (Tami?) and Tina – Down the hall in their own company: TNT Graphics

Q: What are your thoughts on how the industry has evolved?

Bruce de Graaf: Two thoughts, validated by the years:

1. "The game industry is dead".  Every single year it is pronounced dead by somebody who should know better.  I'd love to be "dead" as a multi-billion dollar a year enterprise!  "Give me a call; I'll be sitting around my pool … ON MY BOAT!" (Larry Jones).

2. If the platforms ever go open (anybody with a couple of hundred bucks could then buy a development system), the number of games will explode.  However, the avalanche of product would then cheapen the really good stuff.  To a slight extent, that may be happening to the film industry; to a large extent that HAS happened to the music industry.  Most musicians are itinerant and impoverished; under similar circumstances, game programmers could go the same way.

Star Trek IBM independent released (shareware)
Star Fox Atari VCS/2600 Mythicon released
Sorcerer Atari VCS/2600 Mythicon released
Fire Fly Atari VCS/2600 Mythicon released

Return to main menu