Wednesday, June 8, 2011

Learning 3/5—Try it Out


This week, I've got a theme: Learning. There are five steps in the process of truly learning something and I'll be discussing one each day.
Choose to Learn | Don't Take Notes | Try it Out | Play with Alternatives | Teach It

Try it Out (Part 3 of 5)
Some ideas are easy to visualize. You can imagine how they would look and work, as well as what would happen if you actualized the idea. For example, what if I made a Lego roadster? It'd be red and blocky, and it would have wheels that spin but don't turn. I could run it around on the carpet or I could push it forward, but it couldn't turn and if it fell, it would break pretty easily but still be easy to fix. If I give a Magic creature trample, I know that it will make very little difference on a small creature and will mitigate the value of chump-blocking for a larger creature but, unless the card has a damage trigger, not much else.

Some ideas are deceptive. You can imagine them, but the accuracy of your prediction is questionable. What if I built a real roadster? Well, I don't know much about buying car parts and my budget's not unlimited, so I may not be able to get the 1967 cherry red corvette body I'd like. I know if I buy good parts I can make it fast, but how fast? And with what kind of gas mileage? Some of these are questions even an expert couldn't know without beginning the process. If I give a creature dreamstep, how will that affect the game? I can imagine it'll cause some must-attack-to-block situations as well as some confusing board states once you add more creatures to each size, but I really can't know how it'll actually play out until I see it in action.

The trouble is that there are a lot of ideas that fall between "I'm sure I know what's going to happen" and "I'm sure I don't know what's going to happen" and that's a dangerous space because it's expensive (time-wise) to properly experiment and we don't want to waste time but it's far worse to assume your idea is going to work one way only to find out too late that it works differently (or not at all). Assuming something's going to work exactly as planned is what causes plans to go awry and designers to fail. I know, I've been there.

I spend a lot of time developing code for a living and I can tell you that it doesn't take long before you learn to do the hard stuff first and then fill in the gaps around it because otherwise you might get to the hard stuff and discover it's actually impossible stuff or that it has to be built so radically differently in order to actually work that all your previous work is wasted. I also spend a lot more time just writing stuff that may or may not work, trying it and fixing it until it does work than sitting quietly trying to figure out the exact right solution before trying anything at all. The reason being that your theoretically perfect solution is never perfect and it's just faster to try something that might work, observe how it actually works and then modify it. Note that I'm talking about individual snippets of functionality and not a large project — that is exactly the kind of task that benefits from more upfront thought because on that scale it's more expensive to have to go back and re-architect everything later.

I've also made that mistake in game design. If you followed the #GDS2, you watched me make this mistake and you saw it earn me the first elimination. The funny thing is, it's not like I was oblivious to the value of playtesting at the time, I just had managed to convince myself that I didn't have enough time to warrant it. While it's true that our time was very limited, there was no excuse to skip playtesting. It's so vital to the process that you just make room for it.

Playtesting is actually doubly important in game design because every design has two goals. The first is to interact with the rest of the game in a balanced, relevant, interesting manner. This is the mechanical aspect of it and failure in this department leads to stale environments like the current Standard with Jace's Cawblade driving down tournament attendance in droves. The second is to be fun. While it can be difficult to predict how a card or mechanic will interact with other cards and what kind of Limited and Constructed play will result, it is possible with training and aptitude to at least get a good estimate. But no one can predict how fun a game/mechanic/card will be in play. To clarify, it can be easy to spot whether a card can be fun or not, but you can't know whether a card will truly be fun without playing with it and—more importantly—letting other people play with it.

It may seem I've gotten far afield from the subject of learning, so let me tie it back in for you. No good teacher ever explains a concept and then doesn't give the students exercises or homework to try it out for themselves. You don't get a driver's license without trying out the rules of the road via your permit and they don't let pilots fly planes without running through many simulations first. How do you get good at football? Toss and catch footballs ad nauseam with your buddies, run a lot, and play football as much as you can. How do you become a rock star? Well, that requires a lot of luck and perseverance, but before those can even be relevant you've got to master your instrument and you can't do that without playing the hell out of it.

Practice makes perfect. Period.

[2019 Edit: I've since learned practice makes permanent. It drills in a certain way of doing things, and if that way is flawed, you're integrating that flaw into your method. A good segue for…]

Check back tomorrow for part four: Play with Alternatives

No comments:

Post a Comment