Were there specific criteria for making a game work at all costs vs. leaving it be if it had problems too weird to debug?


Boris asks, "were there specific criteria for making a game work at all costs vs. leaving it be if it had problems too weird to debug?"

I don't recall there being any formal criteria. There was nearly always a good collection of game compatibility bugs assigned to me, which I guess wasn't surprising since that was one of my jobs after all. Sometimes it was easy to figure out; sometimes it was hard. Sometimes it took a day; sometimes it took several days.

The bugs were assigned priorities and severities by the person who found the bug. For example, something that prevents the game from running is more severe than a visual glitch in a game that otherwise seems to run fine.

If the game was a major title, it was more important to fix, compared to a game that didn't sell as many copies. For example, DOOM was a pretty high priority to get working.

But overall, I was allowed to exercise my own judgment as to how much time to spend trying to fix each game. The fact that we had MS-DOS Mode available meant that there was a fallback if I couldn't figure out what was wrong, or if I figured out what was wrong and there was no practical way around it.

I"ll take the next few Mondays discussing some of the things that I had to deal with. It's cheaper than therapy.

Comments (10)
  1. Ray Koopa says:

    I wonder if you’re also experienced in DirectX game development, Raymond. You speak about “visual glitches”, which might just happen from changes the Direct3D team made (or a funny graphics chipset telling Direct3D it supports everything, to cite one of your famous articles). Wouldn’t you forward the issue to them then? Or were these “fix this game” tickets from a time before DirectX?

    1. DirectX didn’t release until a month after Windows 95 was released, and most games at the time were designed to run in DOS anyway (or, on the rare occasion, use the WinG library). It wasn’t built into Windows until the OSR2 release, which was the following year.

  2. Joshua says:

    This is gonna be tales worth reading!

  3. Mike says:

    I have a nagging suspicion we’ll hear about VGA registers. Trust me, you don’t want to go near them.

    1. AsmGuru62 says:

      Bit planes, barrels and latches!
      Lots of fun! Or not… it depends.

  4. Andrew says:

    Quote from original article: “The opening title sequence never worked on Windows 95, at least not without using MS-DOS mode. I console myself….”

  5. Yukkuri says:

    >I”ll take the next few Mondays discussing some of the things that I had to deal with.

    This promises to be interesting! I love the appcompat war stories

    1. Will says:

      Me too, my absolute favourite content here!

    2. xcomcmdr says:

      Same here, I can barely wait ! :)

    3. Scarlet Manuka says:

      Me too! Especially since I still play some of those games, thanks to DOSBox. In fact, I suspect one of the games I play may be Example 85 in the second bonus chapter. (This example is a space-themed game that needs 30 XMS handles during planet landing but doesn’t check whether they could all be allocated, with interesting consequences. The game I have in mind is a space-themed game that, under DOSBox, has a long pause at the planet landing stage, much longer than it had when playing in real DOS.) Is 1992 long enough ago to break the no-naming rule?

Comments are closed.

Skip to main content