The "Red Line" in Game Demos
So here's my latest thought-dump.

The Devil is in the Details
Maybe I just have a fragile ego. I read a bad review, look back on my Game-In-A-Week effort, and look at the miserable state of my current project, and think, “Oh, man, I am the Ed Wood of videogames.” The fact that my current Torque project bears a passing resemblance to Plan 9 from Outer Space doesn’t help any. And when I got back to it, after being away from it for a while to do the “Game In A Week” project, I found that the control scheme I previously thought to be “okay” was utter crap. What was I thinking?

In Tim Burton’s movie “Ed Wood,” Wood (played by Johnny Depp) explains his philosophy about movie-making, which was that people only care about the story, and that they don’t care about the details. Whether or not Wood actually said something like that in real life, it’s important advice to remember – and avoid!

Attention to detail is the mark of most quality products. And the lack of such attention is the mark of some utter, dismal flops. My most painful gaming memory is the 24 hours in which I possessed – and for over two hours played and desperately tried to enjoy – “Tresspasser,” the Jurassic Park game. There were undoubtedly some very, very cool technologies in there, and I’d like to believe there were some seeds for some brilliant gameplay. Unfortunately they were contained in one of the most appallingly poor game products I’ve ever seen. From framerates that drove a nearly-top-of-the-line machine to unplayable levels (two frames per second? With a then-modern 3DFX card?), to the ridiculous “waldo” interface, to the super-stretchy arms, to the guns and boxes that were impossible to stop from rolling end-over-end down a 10 degree incline, to the two-item interface, to the completely brain-dead and boring dinosaur AI. Absolutely unforgivable lack of attention to details!

Unfortunately, when you are really close to a game, it can get easy to get used to glaring problems, and ignore those details. Another problem is that with limited resources (and everyone has this problem, from indies to dev teams on $50 million projects), there’s only so much attention to details you can provide – eventually you need to prioritize, know which deficiencies still need improving, which can be safely ignored, and know when to say, “It’s good enough.”

The simple answer is to get outsiders to test our game for us – but getting useful advice from testers can be quite a challenge. During the Test Fest for Void War and Outpost Kaloki, I found that I got more useful information watching players rather than hearing their suggestions later. It’s difficult to sponsor an event like this every time you need a fresh perspective on your game, so having more tricks up your sleeves is important.

Borrowing a Trick from Another Medium
In college, I once sat in on a panel of popular science-fiction / fantasy authors at a conference. The question was raised about how to improve the quality of one’s writing. The universal answer was ruthless editing and revising.

One of the authors (was it Barbara Hambly?) mentioned a trick she uses within a group of writers / editors that trust each other. They read the manuscript WITHOUT the expectation that they will finish the whole thing. They draw a red line at the point where they lose interest and quit reading, and send it back to the author. The author then finds out where everyone quit reading, tries to understand why, and revises the story and sends it back out again. The exercise is repeated again and again, with the goal of pushing that red line further and further towards the end until eventually they find themselves sucked in and finish the entire story. Now I imagine they include additional notes in red ink about the story – problem areas, compliments, etc. But the bottom line is literally the bottom line.

Can this be done with games? A short-story might take an hour or two to read, and it reads the same way every time. A game, on the other hand, may have 10, 15, 20, or even more hours of gameplay, and certainly shouldn’t play exactly the same each time.

But as independent, “shareware” developers, we do have a demo. Often the demo allows around an hour’s worth of play – when it ends, the player should be left hungry for more. While the full version of the game must be a solid extension and worth the money the customer is planting down for it, our success is directly tied to the demo. And I have lost count of the number of demos I have uninstalled from my machine long before the demo time expired. If it were a short story, you could draw a red line where I lost interest.

It certainly seems like there’s an obvious analog in the gaming world to this ‘polishing’ exercise in writing. For my current project, when it reaches “alpha,” I am NOT going to ask my associates to slog through as much of the game as they can, courteously pushing themselves past the point of frustration and boredom so they can try and provide me with a more complete list of issues – and in the process also blinding themselves to the same ugly details that I’m also blind to. Instead, I’m going to ask them to play only to the point where they would make the decision not to buy the game – and tell me where and why.

Now this will have to go hand-in-hand with a more traditional testing process during beta. And I can’t neglect the full version – but chances are the same things that suck or rock in the demo will also suck or rock in the full version, so the benefits of the process shouldn’t be limited to just the demo.

Now, I am still a little ways out (unfortunately) from Alpha, so if anyone else tries this technique out before I do, please let me know how it goes!

Ron Gilbert’s Engine-In-A-Week Experience
I have started following Ron Gilbert’s blog, and I saw today that while I was experimenting with a game-in-a-week with PyGame (Python’s SDL binding), he was doing something very similar, ripping out the guts of Sauce (his update to the classic SCUMM engine) and replacing the graphics side with SDL. I have a helluvalotta respect for Ron as a game developer and designer, so I found his report fascinating. And curiously enough, he’s about to delve into the mysteries of Torque next. I can’t wait to hear his thoughts. You can check it out at:
http://grumpygamer.com/9848959


And the Next Hit to My Productivity Is…
Last night my copy of Gran Turismo 4 arrived. Gran Turismo 3 and Dance Dance Revolution Extreme have been the two “killer apps” for my Playstation 2. Combined with the PS1 compatibility so I can play my old titles, those games pretty much justified my purchase of the box. I lost so many hours in GT3 I lost count – and I am NOT a racing-game fan! At first blush, there seems to be some weirdness with incomplete text that looks like it should be scrolling but isn’t – and some funky HUD display issues on replays that might simply be an attempt to be artistic and edgy. I will try to be disciplined – I will try to be strong – but I’m expecting the addiction to begin anew.

Labels: Game Design

- posted by The Rampant Coyote @ 12:05 PM (Permalink)