Interview starts..

First of all, which of Atari's early games did you work on? [pre -85'ish I guess, but also if you feel something was ground breaking/you are especially proud of, after that]

I designed and programmed
- 4 Player Soccer (4 giant trackballs, B/W, look down cabinet).
- Missile Command
- Tempest
- I, Robot
- APB (All Points Bulletin)

What was your speciality? [any part you feel you are especially good at?] [concepts, math etc..(shared DLL libraries...=)]

My specialty was in having fun, addictive fun. I had no technical specialty, considering my degree was in psychology. I liked doing explosions and special effects.

Atari..

How did you guys at Atari go about constructing a game back then?
[was it a hard hieracically controlled development vs.you where given a subject to [freely] construct a game around]

Either come up with your own game idea and get it okayed by mgmt., or pick an idea from a catalog of approved ideas. The catalog of approved ideas was created by all of us at regular brainstorming retreats to Pajaro Dunes and other cool places.

The routine? [Started with a concept, gave it so many manhours to get going - or as at Nintendo, a number of groups work independently with paralell development of games and at the end of each time period [every two months] the groups prototype game get reviewed by a board (in competition with ten other games) that thumb's up or down it]? [in short, the working procedure]

The programmer/game designer worked with a hardware guy who usually designed a unique new piece of hardware for each game. Then there was a project leader to take care of red tape, coordinating cabinet design/manuals/field tests/etc. There was a technician to keep the PC board working. That was about it. That composed a team which worked on just one game. Reviews were on just 1 game/team..

Pick an idea. Write up a game proposal. Get it ok'd by mgmt. Take a couple weeks to bring up a playable simple version. Mgmt. reviews that and oks it or axes it. If ok'd, continue with the whole game. Regular reviews by mgmt. to make sure still fun. Kill if not. Then field test and focus groups. Read collection reports. If makes big bucks, you're in there with a winner, and finish it up. Otherwise kill it, or big changes and keep going. That was a big problem later on at Atari: they forgot how to kill projects that weren't fun, and would let them go on forever.

So what was the working atmosphere like? [all suits with strict working hours vs. lax jeans etc]

No suits. Jeans/shorts. Very intense competition to do hot games. Massive voluntary overtime in order to make your game as hot as possible in the limited amount of time allowed by management. Lots of Friday parties/beer/volleyball. Lots of divorces. All the major competition's arcade games available to be played at all times for "research purposes". Ha.

---

[More specific] about Tempest then: Did you work alone on Tempest, or where you a team?

I designed the gameplay, graphics, sounds, and did the programming.
A hardware engineer designed the special vector display circuits and math box.
A project leader coordinated things.
A technician kept the hardware running.

Who did what?

Sorry, but I can't give names. See above.

How did you go about with the programming?

It was done in assembly language with structured macros (for loops, if/then conditionals were supported, avoiding goto's).
I did all the programming. I'd write the code on programming sheets and turn it in to the typists who'd type them in to the DEC computer, then give us a tape with the resulting compiled/linked program. We'd then take that to a black or blue box for debugging. We'd mark changes on a listing, give it to the typists who'd edit our files and give us a new tape, repeat ad infinitum.

Any part of the implementation that you remember struggling with?

The action at the top of the hole when the enemies are flipping over on top of you was a continual user interface problem.
The game play actually went through 2 phases: the Tempest gameplay was the second plan after the first plan was discovered to be unfun.
The initial gameplay was to be simply a First Person Space Invaders. I got it up and running, but it wasn't much fun. Too much like Space Invaders (surprise surprise). So after the review they said do you want to kill it? or have you got any other ideas? I said I've got this nightmare I have where monsters are coming out of a hole in the ground and I've got to kill them before they get to the surface and kill me. Let me just take the 1st person space invaders, wrap them onto the surface of a hole (tube) in the ground. Mgmt. said OK. I did it and it was a lot of fun. I first tried rotating the tube instead of the flipper, but that made everybody want to puke, plus it was a lot more work, so we went to moving the flipper around the upper edge.

Was there a lot of effort put into the actual visual design of the game? [flipper should look like this, blaster has to be yellow - with polls etc]

Yes, but not extraordinarily out of proportion to the rest of the tasks, considering the whole project was done in less than a year.
I designed all the monsters, playing surfaces, colors, etc. and I'm not an artist by any stretch of the imagination. We didn't have any staff artists yet at that time. I just liked flashy electrical glowing pulsing kinds of things.

There was a rumour that the programmers wanted monsters, bigger and detailed ones, and that the cabinets were already designed when - at the last moment - those ideas where scrapped and Tempest got that 'electric' design.
- Care to comment? [what _was_ the monster theme?]

Totally wrong. First, there was only one programmer. Second, the monsters on the cabinet were designed at the last minute after the game was pretty much cast in concrete and preparing for production. The monsters in the game pretty much looked that way from the very beginning, and there never was any plan to do anything differently. The big restriction was in the number of vectors that could be drawn and moved around per frame and still have a fast frame rate. Simple graphics helped the game play fast.

Talking hardware, what platform did you use when programming? [Apple?]

We used an in-house debugging system. It started out on the "Black Box", then we got the "Blue Box". The Blue Box used the Forth programming language for handling debugging, and worked pretty good (for that time in history).

So how did you emulate the released [hw] Tempest PCB with the vector generator?
[- a hacked (prototype PCB?) shared memory RAM in prom sockets?]

The Blue Box was hooked up to the arcade cabinet's actual PC board, which was usually out of the cabinet lying on the lab bench for easy maintenance. We debugged right there on the actual game. The only thing we did at our desks was write code with pencil and paper, design graphics (no tools) , think up sounds (no tools).... The only thing emulated was the ROM in the PC boards which was plugged into RAM in the Blue Box, so we could load in the latest version for debugging from the tape.

And you had a prototype Wells Garner color XY monitor standing vertically on your desk....? [what did your [hw]setup look like? chaos?]

Yes. I remember one time the prototype monitor stopped working. It turns out that some of the components on the monitor's PC board got so hot that they melted their solder, desoldering themselves, and they fell out onto the table (the board was mounted upside down). It was funny to see these components lying there under the board next to the black monitor.

By the way, how come Tempest became a vertical game?

The vertically mounted monitor was more appropriate because of the need to display a basically circular game play in the center, with horizontal bars of information up top and down below. A horizontal monitor just wouldn't have allowed as big of a playing area and there would have been lots of wasted space on the sides.

I understand the vector generator code is the same with all Atari vectors...?
- but not the [hw]vector machine?

Don't know.

As you know the monitor broke after a while on location, what did you guys at Atari do about it? [Did you all go haywire and tore your hair in frustration?]

That was a hardware problem. So I let the hardware guys tear their hair out.

What was the cause of it? [I mean, I have a Space Duel that (before I bought it) had been on location for some fourteen years (ok, working only in summertime, but anyway..), with the same monitor that Tempest has, and not a fault....]

Tempest was the first game to use the color vector generator. I'm sure there were things they discovered in production after doing Tempest and a few other games that enabled them to protect the monitors and make them last longer.

On the other hand, maybe the monitor was vulnerable to being destroyed by the software, and I did things to it that the other programs didn't, like when you blast down the tube, and vectors get drawn all the way out past the edge.
Perhaps that fried it.

You don't have to reveal this, but do you know who put in those infamous post_180,000_last_two_digits cheats?

That was me. But it was intended as protection against piracy. I had about 6 levels of protection to prevent things such as removal of "Atari" from the various screens by unscrupulous Japanese and Italian counterfeiters. I had checksumming going on in many places to detect alterations to the code. If anything amiss was detected, I would trash RAM at random (sometimes using the last few digits of the score as the random number), sometimes after a delay of a few minutes. Unfortunately a bad thing happened: just before shipping, I rearranged some of the screens, moving Atari more towards the center on one of them, and I forgot to change the 3rd or 4th level checksum protecting that location. We did release a fix for this after we discovered the problem.

What happened when Atari got the news about the cheats? [Did you write the updated ROM's - 217 and 222?]

see above.
I at first claimed that it was a hardware problem (of course). But then a hardware analyzer in one of the labs caught the actual occurrence of the software trashing the coin count. So I immediately ate crow and fixed the problem.

Are there any other cheats/easter eggs, that you know is still in there?

No comment.

Tricks?

No comment.

Bugs?

Of course not.

The initials on the lower five highscores [those that all have 10'some points at powerup], who are they?

ONM
LKJ
BVD
MPH	project leader
DFT[you?]	?
STL	hardware guy
MJP	technician
BEH

Did you play Tempest a lot?

Yes. A whole lot. It was my second favorite game after Robotron, which had that similar kind of total zen concentration thing going for it.

What was your highscore? Maximum level you came to?

Don't remember. Just remember playing it through somewhere around the 80th level.

Why do you think Tempest became such a 'cult' game?

Feeling of total immersion, almost like a trance or zen state, you get into when playing it. Good pacing. Ramped difficulty. Selectable starting level. Restart near where you died off. Visually stimulating graphics. Fast action. Great control - felt good to just spin it. Good sounds. Sinister theme. That feeling like you almost made it to the next level, and in the next game you could do it. Pulsing/electrical sounds/graphics/gameplay.

Did you have that feeling when the game was released?

Had that feeling before it was released, during development, when it was always being played in the lab. People were saying it before it went out for field test.

Because it really is a classic, Jeff "Llamasoft" wrote Tempest 2000 for the Jaguar you know.... A bunch of remakes for different platforms has been made... Btw, do they bug you when they make conversions/ports?

No.

Any arcade cab's at home?

Yes.

• Tempest (a prototype called "Vortex" which is now broken again).
  
• Missile Command.
  
• Solar Wars (video pinball).
  
• I, Robot

What do you do now?

DeBabelizer, so everybody else in the world can do super graphics for games. (left Atari in -91)

Thanks!