Things you shouldn't do

edited in General
So I was reading about the creation of Starcraft, and ended up on a blog post made back in 2000 by Joel Spolsky.

The short is that starting from scratch on something that has been publicly available is a bad idea. Even crossing this to game design makes sense. The pacing, in most good games, in standard and formulaic. Call of Duty only gets incrementally better because it only needs to.


  • edited
    The short is that starting from scratch on something that has been publicly available is a bad idea. Even crossing this to game design makes sense. The pacing, in most good games, in standard and formulaic. Call of Duty only gets incrementally better because it only needs to.
  • He might have meant to link the article. Might have.
  • (only ribbing him because he goes all grammar nazi on everyone else)
  • He deserves any number of ribs :)
  • The article itself has some good points, although it doesn't always carry across to game development 100%, the logic is pretty sound... I just worry that the same logic could be used to argue for building a monolithic engine that you use in all your projects from then on, because I doubt people would understand that the article pretty much states that it's impossible to get solid, working code without lots and lots of testing and real-world exposure.

    I just think @Karuji didn't put much time into the OP, so it doesn't make much sense. Care to elaborate on what you were trying to get across, sir?
  • Sorry for the poor OP: I was busy with two other things at the times — ironically one of which was studying Japanese giving @BlackShipsFillT Engrish comment a deliciousness to it —

    @Tuism I linked to the article in the word blog. 95% of the article link I have posted have been in bbcode, but here is the link again

    @BlackShipsFillyT I have grammar nazied once :P


    The reason I really liked has a number of reasons. I was initially reading about the development of StarCraft, and how they rewrote the engine from scratch as opposed to using the WarCraft one. They had a great deal of trouble with doubly linked linked lists which prompted the author to quote and link to Mr Spolsky's blog. The quote being.
    It's important to remember that when you start from scratch there is absolutely no reason to believe that you are going to do a better job than you did the first time. First of all, you probably don't even have the same programming team that worked on version one, so you don't actually have "more experience". You're just going to make most of the old mistakes again, and introduce some new problems that weren't in the original version.
    At the same time I was reading Mike Lopez's article Gameplay Fundementals: Harnesed Pacing and Intensity. One of the major parts of the article was the planning of level design, and the incremental steps of building a the fun in the game as a player progresses.

    This is something I have been interested in for a while. Please look at the image below before I continue rambling


    The "normal" progression is the one of the exponential curve, but the preferred progression, as stated by Lopez, is the green line. Some months ago I gave a talk at the JHB meet about how we need to concentrate on how we grow player skill and mastery in games, my interest also in finding the link between this macro progression, and the micro loops of gameplay feedback.

    This related to be due to my reworking of Colour Master, which I started from scratch due to not having the original GMK file. I have the same core mechanic, but I found the new game rather unsatifying to play as opposed to the original prototype from '09.

    So with this in mind I looked at the old and new versions of CM. The old version had been played, and had been iterated to solved certain problems, while the new version is sitting on my computer waiting for the some more features, which I hope will make the game more fun.

    Looking at Colour Master I realize that I kept the core colour changing mechanic — the micro of the game — while I had started the macro of the game from scratch.

    This led me to look at Call of Duty which only has incremental increases in design with the release of each game.

    So if the pacing of a game is a known formula, then there is likely to only be incremental increases in the development of games. The introduction of regenerating health by Halo, is really just and increment in how things were designed, but it changed the way modern FPS games work since there was no longer the worry that a player on max health and shields would breeze through an area, but a player on low health and shields would have a hellish nightmare

    Another example of where the iterate instead of start from scratch would be Diablo 3 and Tourchlight 2. Diablo 3 bears little resemblance to its predecessor. Where Tourchlight 2 really refined and polished the first one.

    When I was at GDC this year I went to a talk by one of the designers of Deus Ex: Human Revolution, as much as the game is rather different from the first two in the series the designers really drilled into the first two games, looked at what made them tick, and what the fans did and didn't like about those games.

    Another thing is that I was watching a talk by Chet Faliszek in which he showed how the cartoon style demo evolved, and through slight changes he reached his eventual appearance.

    On the other hand the demo did start out looking like a normal solider who was wearing military camos so the iterate philosophy isn't an absolute, and as it is often said sometimes you just need to kill your ideas.

  • I believe we need to change the css or something so that links would stand out a lot more. I've missed SO MANY links like this, even after careful scrutiny.

    On topic, TL2's Charge meter is a very good example of how to implement the wave-form pacing. You build up momentum, run, run out of steam, leave. You're incentivised to continue your run because you know you're maximising your damage potential. But that also results often in running into situations beyond your ability, and your wave crashes. Fun stuff :)
  • @Tuism I posted in the make the site betterar thread about the linking.

    The charge in TL2, from what I remember of the beta, is a good example of the wave form, but that is a micro level. The waveform (on a macro level) deals with the overall game progression, so as the narrative increases in drama and tension so does the difficulty, so after defeating a boss the game might dial down the enemy attack and health values by 10% to really drive home the fact that you're a badass.

    Borderlands2 did this really well, when they send you back to the starting area on a side quest, and you breeze through enemies that previously were a tough choice, they also gave something that initially appeared as purely for sceptical to have a practical value. (I really love Borderlands2, can you tell yet?)
Sign In or Register to comment.