Did Atari have a QA team?

[godzillajoe]

This isn't meant to be another "because who the hell tested Pac-Man?!" type thread but as someone who does QA and software automation for a living, I'm just curious.

 

Or did the developer essentially test his/her own code?

 

What was the common practice in the industry back in the day

[jaybird3rd]

Atari did have a team of test engineers; David Comstock (who later worked on ET Phone Home, Cloak & Dagger, and Superman III for the Atari 800) was one of them. He talked about his job during his interview with Retro Gaming Radio a few years ago. I don't remember the details, but I believe he mentioned recording game footage with a VCR that was synced to the 2600 display, so the programmers could review the playback and study the exact circumstances which triggered a bug (such as one on-screen object passing through another when it shouldn't have). I think Video Pinball was one of the games he recalled as having an especially long debugging cycle.

[bikeguychicago]

My guess is that they didn't have a formal QA department since most of the coding was done by a single person per game. I would imagine that any QA done on a game was by the developer, primarily, with his or her co-workers acting as secondary testers simply because they were either interested in playing the game and/or they were helping the developer overcome a bug in the program.

[Andrew Davie]

Back in the early '80s I found it invaluable to record games when testing, with a VCR. Being able to replay a bug helped heaps. I did this personally when I was a bedroom programmer, and our QA department when I worked professionally ('84 onwards) also did this. I expect that it was done at Atari, too -- often a programmer won't believe there's a bug unless s/he is shown it in person

[BenjaminR]

Fascinating question! I work in the QA world too and I often wonder how Atari may have handled software testing back in the early 80's. It's interesting to read about recording the games with a VCR, that certainly would have worked. I'm always interested to learn about how testing was done long before the invention of all the free software tools I use and take for granted every day at work.

[X900BattleGrape]

Back in the early '80s I found it invaluable to record games when testing, with a VCR. Being able to replay a bug helped heaps. I did this personally when I was a bedroom programmer, and our QA department when I worked professionally ('84 onwards) also did this. I expect that it was done at Atari, too -- often a programmer won't believe there's a bug unless s/he is shown it in person

 

Ah the inevitable "works on my machine" blow-back from programmers etc. I don't think you could get a way with recording gameplay these days, given the complexity of the games and input devices, but back then it would have worked quite well I imagine. I'm in QA myself, so it's a fascinating question to me. With persistent online games this picture sums up a tweak on the old "works on my machine" saying.