Loading ...
Sorry, an error occurred while loading the content.

RE: [XP] XP- Game design

Expand Messages
  • Phlip
    ... It was speculation that unit tests should provide test fixtures that acceptance tests re-use. That would also leverage tom! s recommendation to use the
    Message 1 of 13 , Mar 1, 2004
    • 0 Attachment
      BETIS Alexandre wrote:

      > That more or less sounds like "do acceptance testing
      > too, and not only unit-tests". Am I wrong? This is
      > the feeling I get from this post-mortem anyway.

      It was speculation that unit tests should provide test
      fixtures that acceptance tests re-use. That would also
      leverage tom!'s recommendation to use the game's
      scripting system as much of its acceptance tests. But
      that means every aspect of the game effort must be
      test-infected.


      =====
      Phlip
      http://www.xpsd.org/cgi-bin/wiki?TestFirstUserInterfaces

      __________________________________
      Do you Yahoo!?
      Get better spam protection with Yahoo! Mail.
      http://antispam.yahoo.com/tools
    • Robert Blum
      Hi Phlip! ... Not worth the effort. Really. Tests for game play elements are ridiculously fragile, and game play changes on a minute-to-minute basis in the
      Message 2 of 13 , Mar 5, 2004
      • 0 Attachment
        Hi Phlip!

        > Author the game play elements, leveraging the engine's
        > test fixtures.

        Not worth the effort. Really. Tests for game play elements are
        ridiculously fragile, and game play changes on a minute-to-minute basis
        in the final stages before a game goes into beta.

        > Author the game scripts, leveraging test fixtures on
        > the play elements.

        "Scripts" referring to cut scenes? I don't think so. That's utterly
        pointless. You watch it when you're done, and when you like it, it gets
        checked in. It's not going to change unless you either change the
        script (in which case you have to re-watch anyways), or the engine -
        which your underlying tests should catch.

        > Script playing the game, and collect statistics on
        > survivability.

        Cute idea. However, human players are a bit more ingenious than a
        simple script. Those statistics are meaningless, at least for the game
        I work on. Especially since many of the tweaks the scripted approach
        would find might very well be unreproducible for humans. (Unless you're
        able to control a game pad with millisecond-precision. Even the top
        players find that quite hard to do.)

        > The problem with that fix is nobody wants to hear
        > "rewrite an inhouse module that has worked for us
        > before" at the beginning of a project.

        Your fix would only address the engine - anything above that has proven
        time and again to be very test resistant. We've tried it, and it didn't
        work out. Any game team I know that tried XP says it didn't work out.
        Any game team that has written about XP says it didn't work for game
        play. (Well, at least the articles I read)

        I might very well be wrong. In fact, I'd love to be. So, here comes the
        call: Is there *any* commercial game that did XP on
        gameplay-components?

        - Robert
      • Phlip
        ... Thank you, Robert. My post was carefully designed to outrage all the eXtremo game developers, and flush them out into the open. ;-) ... Oh, that s another
        Message 3 of 13 , Mar 5, 2004
        • 0 Attachment
          Robert Blum wrote:

          > > Script playing the game, and collect statistics on
          > > survivability.
          >
          > Cute idea.

          Thank you, Robert. My post was carefully designed to
          outrage all the eXtremo game developers, and flush
          them out into the open. ;-)

          > ...Even the top
          > players find that quite hard to do...

          Oh, that's another thing. Game shops put their Quality
          Control budget into "top players" to manually test.
          This is why I can never figure out how to play any
          modern video games when I try.

          > I might very well be wrong. In fact, I'd love to be.
          > So, here comes the
          > call: Is there *any* commercial game that did XP on
          > gameplay-components?

          I think we are noticing that near-XP fails less
          reliably than XP here.

          =====
          Phlip
          http://www.xpsd.org/cgi-bin/wiki?TestFirstUserInterfaces

          __________________________________
          Do you Yahoo!?
          Yahoo! Search - Find what you�re looking for faster
          http://search.yahoo.com
        • Robert Blum
          Hi Phlip! ... Well, thanks for thinking of us :). ... Ah. Not quite. Play-testing a game turns you into a very good player, no matter what. It even works for
          Message 4 of 13 , Mar 6, 2004
          • 0 Attachment
            Hi Phlip!


            >> Cute idea.
            >
            > Thank you, Robert. My post was carefully designed to
            > outrage all the eXtremo game developers, and flush
            > them out into the open. ;-)

            Well, thanks for thinking of us :).

            >
            >> ...Even the top
            >> players find that quite hard to do...
            >
            > Oh, that's another thing. Game shops put their Quality
            > Control budget into "top players" to manually test.

            Ah. Not quite. Play-testing a game turns you into a very good player,
            no matter what. It even works for me - I've got no particular skill in
            the genre we're building, yet outplay average people in any game of the
            genre, just by virtue of being exposed to it all day. (Everybody on my
            team still beats me with both hands tied behind their backs, though...
            :)

            I agree that usability testing is sometimes not getting the attention
            it deserves - but it's hard to constantly get fresh people for your
            latest build, and it might interfere with your goals to have
            unexperienced players look at a game - see below.

            > This is why I can never figure out how to play any
            > modern video games when I try.

            It might be a matter of choice. There are plenty of out-figure-able
            games out there. I agree at the core, though - too many games require
            prior experience or a thorough reading of the manual. But again, that
            is a side effect of the market we're in. If you want a game to sell,
            you *must* get the 'multiplicators'. Early adopters who are very
            knowledgeable about games and influence their friends to buy it. Since
            they tend to be experienced gamers, that is our user. (By proxy - but
            still).

            Unfortunately, you're an innocent bystander in this and lose.

            Notice that established franchises and movie licenses often have a
            simpler interface - due to different customers. (Hardcore video gamers
            tend to look down especially upon movie licenses, the same way a
            'artistic' movie critic has a problem with Hollywood fare. Franchises
            have time to refine interfaces over multiple sequels)

            >> I might very well be wrong. In fact, I'd love to be.
            >> So, here comes the
            >> call: Is there *any* commercial game that did XP on
            >> gameplay-components?
            >
            > I think we are noticing that near-XP fails less
            > reliably than XP here.

            The above sentence did not parse for me. Why would we be interested in
            reliable failure? Aren't we trying to reduce the rate of failure?

            - Robert
          Your message has been successfully submitted and would be delivered to recipients shortly.