Locking Down

My parents were in town all last week and so being a nice son, I offered them my bed instead of having them pay for a hotel room. This led to me sleeping on the couch. That’s a lie. This led me to attempting to sleep on the couch. Let’s hope when I get married I don’t get into trouble too often because I think anyone larger than 5’8″ is going to have problems.

If I’m nonsencial, blame the sleep deprevation.

I went to the Seminole Hard Rock Casino on Saturday and was amazed at the variety of slot machines. There must have been hundreds of thematic variations on the exact same play mechanics. Being someone who analyzes games to a fault, slots are a little too simple for me. Even Roulette is more interesting. I eventually gravitated to a Pick ‘Em Poker machine which I destroyed by first doubling up and then, after taking a break, adding another 50%. High on my sense of self, I stroke confidently over to the real poker tables and had my lunch eaten. Turns out I’m pretty good when playing against a stat table, but against real people I’m less effective. Lessons learned.

There’s this article on the Gamasutra network from an EA guy talking about how they are looking at Talent Acquisition. Since it’s from an EA guy, it gets poo-pooed on the other blogs as bullshit. He has a list of “Top Myths” that he wants to debunk:

Top Myths About Working In Gaming

– Work Hard/Play Hard
– Getting Lost In The Crowd
– Jobs, Not Careers
– Old Demographic Paradigms

It’s the first one that irks me so. It’s not that it really is a truth, but it isn’t a myth either. The author says:

More effective time spent in preproduction, especially focused on locking down game design, helps to make scheduling more predictable and manageable throughout the project lifecycle. While the battle to lessen crunch is ongoing, the industry has turned the corner and is improving work/life balance and quality of life.

This is definitely truth. We [as an industry] are working harder to make life more manageable and predictable. I haven’t been forced to work a weekend since 2006. Nice.

But here’s the rub: “locking down game design”. This means an increased investment in the waterfall method – that we must know exactly what we are building before we set out to build it. In non-entertainment software design, sometimes that can be an appropriate approach. But even the best designers in the entertainment industry cannot type out a design on paper and know that it will be fun when it is coded and put into action. It can’t be done.

So in order to deal with the high cost of turning over employees along with the public relations disasters that can go hand in hand with it, management demands predictability. Predictability requires safe, understood designs. Even with the most understood, safe designs you will have incompatibility and unpredictability, however.

The only way to get exciting results is to encourage innovation. But you can’t encourage innovation and encourage predictability simultaneously. Or can you?

Perhaps if you abandon the waterfall method of planning, you can embrace this unpredictability. Perhaps if we are focused, so focused that we may alienate a few customers, then maybe it can be done. I don’t know. My speciality isn’t management. But I do know that right now the industry is trading innovation for work/life balance. That isn’t a value judgement – it is just an observation.

The other three myths I can stand behind. We aren’t at the point where they can be totally debunked, but progress has been made in the short time I’ve been here. Baby steps.

2 thoughts on “Locking Down

  1. In our game, for our UI at least (Zack knows which game I’m talking about), I think that we actually were able to help a lot with the crunch by having the artists and designers get a look and design for the UI clearly established early on (prepro/MS1), though individual screens did change substantially. By the time we got to alpha, there had been several milestones where we did have excess time on our hands. Yet, from my perspective, our crunch time wasn’t all that easy. Why not? Well, our extra time got wasted on two fronts: fixing a bunch of issues with features unrelated to this year’s version (I guess everybody, myself included, assumed they would “Just Work”, but we were mired in a bunch of bugs in code that we didn’t write), and also several producers took our extra time to add features during alpha that hadn’t been planned for.

    So while I totally agree with you that you can’t have things 100% designed up front, having a clear target established can help a lot, as well as having certain software components created early on that could be used in many places, can certainly reduce pressure towards the end of the cycle. And we have made progress on that front over, say last year. At least, on our game.

    But you can’t then go and waste all of that buffer you’ve created thanks to good planning by adding features at then end of a cycle. Having extra time because we’ve planned well doesn’t mean that we have time to add more stuff.

  2. Some elements benefit from being designed upfront, but gameplay is most of what I deal with and it certainly does not. Having a target is definitely essential. I’m not advocating going loosey-goosey into the whole project. But in design meetings we sit around and argue minutiae on what will be fun or effective and it is a waste of time that the original author seems to advocate. You need to get shit in the game before you can decide how to make it better or if it is good enough.

    The problem is that like your legacy code issues things also just “come up”. And when you schedule for 100% or more in the hopes that you will pare down to 100% when things come up you never finish the last 20% of a feature that could be great and you always work until you are just about to go nuts. Producers adding features late in the cycle wouldn’t be as big a problem if iteration time was sufficiently scheduled.

Leave a Reply

Your email address will not be published. Required fields are marked *

Human? * Time limit is exhausted. Please reload the CAPTCHA.