Wed, 18 May 2011

Shall We Play A Game?

Just watched the clip on YouTube, and while the famous line from the 1983 movie Wargames is usually quoted as "Would you like to play a game?", what the computer actually asks is "Shall we play a game?"

And yes, like David, the young protagonist of that movie, I'd love to. Specifically a computer game, since that's all the rage nowadays, and I've loved them ever since I saw my first one, probably "Pong" back in the 1970s. I've played endless games on computers of various kinds, written extensions for existing games, written entirely new games, and just about everything along the spectrum.

I'm in kind of a slump right now, with no game I have available really sparking my interest. And I began to analyze that state of affairs, and ask myself, what is it I really want in a computer game? To answer that, I had to cast my mind back over the years, and over my current roster of games, to get an idea which ones I liked and why. Here are my random thoughts and observations. Note that my gaming these days is entirely done on Linux, so any crossover to MS-Windows or Mac or DOS or mobile game apps will be purely by accident or for historical reference. I'm also not a player of online games or MMORPGs like World of Warcraft or EVE; just never got into any of those. I realize this puts me in a clear minority.

First off, while I enjoyed Pong in its day, and after playing enough of it became quite good at it, these days I tend to shy away from such games. Call them realtime, action, "twitch", or whatever you'd like, I cannot any longer enjoy a game that depends entirely on my reaction time. One genre I enjoyed a lot and still do is flight simulators, which definitely are realtime and depend on swift and accurate reactions, but they also have intervals where you have to think, not just act, and that's what I prefer.

One game I talked about online a few days ago was AV-8B Harrier Assault, a lovely game published in the early 1990s by a company called Domark. I played it for hours, and never got tired of it, and one of the reasons was that it was part strategy game and part flight simulator: You planned your assault, set your missions, allocated your resources, then went and flew the missions, and the outcome of your flight factored back into the strategy portion. For me, that's just about perfect, and any game I play, or design, will, I hope, have the same mix of thought and action.

I like a few simple games, like xmahjongg (a solitaire game based on the tiles of the ancient Chinese board game) and xjig (a wonderful and very configurable jigsaw puzzle, more of a "toy" than a game I guess), but those can hold my attention only briefly. Thus, another of the criteria I have for a "good" game is a certain degree of complexity.

But it can't be too complex, or I won't enjoy beating my head against it. Ideally, a game will have an adjustable level of difficulty, although in many cases the "easy" setting is some game expert's idea of "easy", and turns out to be quite hard in itself. If I really get into a game, to the point where I actually become good at it, I'll probably want to adjust it to a more challenging setting, but when I'm starting out, I want it dead easy, with opponents falling over if I just blow on them hard. Finding the balance between "easy for beginners" and "enjoyably challenging for experts" is hard, and I've never yet found a game that gets it right; in the vast majority of cases, the "easy" setting is just too hard.

I will readily admit that I suck at most computer games, so what's "hard" for me might be boringly easy for others, maybe even most others. But hey, let's give the handicapped a break, and have an "ultra-easy" setting or something, for those of us who find "easy" a bit too challenging. Many games don't even bother, figuring you just need to jump in and figure it out, swim with the big fish as it were, and many of those games I just pass by after trying and failing at them enough times.

Another thing I like in a game is the ability to pick it up and put it down easily, play for a few minutes or a few hours depending on my situation. Let's talk cases: One game I've played a lot in the past few years is The Battle for Wesnoth, one of the best free games out there. It's a delightful swords-and-sorcery game, with orcs and elves and mages and knights and all that, intriguingly complex, gorgeous graphics, just about everything you could want in a game. But one big problem I have with it is that once I get started in a scenario, I really can't "put it down" until I play it to its conclusion. I can't fault the game itself, it's all me. Each "campaign" is broken up into several "scenarios", and mechanically the app has an excellent saving and reloading capability, so I could put it down after only a few minutes. But I lose track, forget my battle plan, so whenever I pick up Wesnoth, I normally prepare for a couple hours with it at least.

Contrast that with another game I've become addicted to in the last several months, Dungeon Crawl Stone Soup Edition, which I'll abbreviate hereafter as DCSS. Even though playing a character through to a win is, for me at least, the work of days, maybe even weeks, each individual session can be as short or as long as I want. You wander around, encounter a monster or six, kill them, and boom, you move on to the next encounter. I can put it down without losing track, plus the game has reports you can request that will let you re-orient yourself to any given character even if you've put it on hold for a very long time.

A third game I've spent far too many hours playing is Oolite, a space-flight and -trade game based on an influential 1980s game called Elite, which I never played, though I did spend hours playing its sequel Frontier. It's about halfway between Wesnoth and DCSS in put-down-ability, with its action being broken up into separate jumps between stellar systems. While you can spend hours in a given system doing various stuff, mostly you just load up, make a jump, drive to the new system's space station, and dock. So your gameplay sessions are "chunked" by the jumps, but each chunk is typically fairly short, so you can make a few jumps or several, depending on the time you have available to play.

I realize that complexity and short playtime are opposing requirements, but obviously some games can do it, and I like that in a game, though the hours I've played Wesnoth show that even with its larger time requirements, I can still enjoy it. I just have to plan out my sessions more carefully, is all.

A related issue is the "overall goal" of the game. Wesnoth is the standout here: It has "campaigns" (which I'd more likely call "scenarios") which are divided up into separate "scenarios" (which I think of as "modules" or "missions" or something), each with a clear goal and victory conditions. Each campaign has an overarching storyline, and each scenario within it has one or two strictly defined victory conditions. And each scenario contributes to the next; how much money you have at the start of a scenario to buy units with depends on how much you used up in the previous scenario, plus usually how quickly you finised the previous one. Also, units that survived the previous scenario can usually be recalled for the next one. In some cases choices you make determine which scenario(s) you'll play next.

I like that structure. Contrast it to DCSS, which has a clear goal, but only one: retrieve the Orb of Zot, and don't die in the process, which is nearly impossible, at least for me. Online I see people talking about all sorts of wins, but I have yet to come close to winning even once. DCSS is one of the famous "roguelike" games, based on a 1980s text adventure game called rogue. That Wikipedia entry notes that rogue "seems to be tuned to about 1 [win] in 1000 attempts". DCSS seems to be about the same. Plus, as with all roguelikes, when your character dies, that's the end; you start over from zero. "Permadeath", it's called, and while it may be more like Real Life, it's not really something I enjoy in a computer game. So you can play for hours, days, weeks, developing and building your character into a mighty force, and then one slip and that's all lost.

Several other games I play from time to time have a similar "single goal" structure. Freeciv, a derivative of Sid Meier's extremely popular PC game Civilization, actually has two alternate goals: either destroy all the other players, or fly to Alpha Centauri. Or rather, build a spaceship capable of flying to Alpha, a huge public-works project that takes up a ton of time and resources; the actual flight is merely assumed. In Freecol, another clone of a Sid Meier game, this time Colonization, you can "win" either by eliminating all other colonial presences, or by throwing off the chains of your mother country. I don't have a serious objection to such long-term goals, but they do require a degree of dedication that I find increasingly difficult to muster. Give me shorter self-contained scenarios and I'm happier.

Some games, thankfully few, actually have no goals or scenarios other than ones you might wish to set for yourself. Some examples are Simutrans and OpenTTD, both of which are semi-clones of the 90s MicroProse game Transport Tycoon Deluxe. Now, I never played TTD itself, but did spend hours with the Linux version of one of its predecessors, Railroad Tycoon II, another Sid Meier title. (What is it with me and Meier's games?) RT2 had actual scenarios you could play, with goals and victory conditions. They were usually pretty long-term, but still, you could do them if you tried. Connect Cairo and Johannesburg by 1935, for example. Lots of railroad history, not a subject I find particularly interesting, but fun nonetheless.

You could also just play it open-ended, with no real goal except to "be good at it". Simutrans and OpenTTD have only this latter mode, from what I can tell, which is fun as far as it goes, but eventually loses its lustre. Same deal with the free clones of SimCity such as Lincity-NG and OpenCity. You could play SimCity open-ended, but it also had scenarios, like "bring 1950s Paris under control" or "save this city beset by natural disasters" or whatever. As I recall, SimCity didn't enforce victory conditions itself; you had to judge for yourself whether you'd met them. But at least it had goals.

So I guess that's another criterion: a game should have a goal, at least optionally. I can play an open-ended game, and in some cases prefer that mode, but I also like some competition against a fixed measurement.

Another positive aspect is the ability for the user community to contribute to the game itself. Oolite is probably the prime example of this: The base game is okay, but really gets old quickly unless you're a die-hard old-school Elite fan. Yes, it has a rating system (again, open-ended, though there are milestones you can achieve), and a few built-in "missions", which have clear goals and computer-enforced victory conditions, but the real shine of oolite is OXPs, oolite expansion packs. These are user-contributed add-ons that enhance the game, with new ships, new equipment for ships, new missions, new environment to fly in, just a whole host of extensions that can turn the game into a whole new experienc.

Wesnoth has user-contributed campaigns, though the ones I've tried haven't been up to the quality standards of the "mainline" campaigns. Simutrans and OpenTTD also have sites with lots of user-contributed content, all of which is nice but sorta pointless unless you're a transport buff.

The culmination of the sadness of open-ended games, I think, is the FlightGear Flight Simulator, a testimonial to how good free software can actually be if dedicated professionals love it and work on it. It has gorgeous graphics, a huge and accurate database of terrain, very realistic flight models, and tons of user-contributed aircraft. As much as I loved flight sims in the 90s, you'd think I'd be spending all my time with FGFS, and yet I don't. That's because, as gorgeous as it is visually, and as accurate as its flying and environment are, it has no point, no goal. You take off, fly around, and then land, or crash, depending on your skill. Oh yay, tie me down. I cannot say enough good things about FGFS as a software application and as a flight simulator, but that's all it is—a simulator. It's not really a game.

One topic that's very hot in games these days is online play, that is, against other humans. I have no objection to that, and it's really almost essential for any game these days, but one thing I demand in a game is the ability to play solo, either against the game's difficulty or against AI opponents. In my fantasies of game design, I envision a single interface that can be used either by AIs or by UIs, allowing you to "plug in" any number of human and computer opponents to suit your taste. Freeciv does that, so I know it's possible.

Finally, let's talk themes. Wesnoth and DCSS are firmly in the fantasy world, swords and sorcery, dungeons and dragons, elves and orcs. Fine stuff for those who prefer it. Freecol and the transport clones are more in the historical genre, sometimes accurate, sometimes not so much. Freeciv is as much about technology as it is about any other aspect of civilization, starting with the bronze age and moving into futuristic things like robotic laboratories. Now me, I'm a fan of science fiction, what has more recently been called speculative fiction. I read Asimov and Heinlein and Niven and Spinrad and Anderson and Bear and Brin and that whole bunch, and keep a weather eye on TV in hopes that another Babylon 5 or Firefly will come along someday. Some SF films are semi-okay, though the pickings tend to be slim even there. What the world thinks of as "SF" is not what I think of, but eh, there are people out there who share my tastes, and some of them continue to write books and short stories, so I'm still happy enough.

Same story with computer games: SF is a huge genre, and I've owned and played a number of SF-themed titles, from Lightspeed to Wing Commander, Star Wars Rebel Assault to Frontier. And yet, most of them are more in the "popular" SF universe rather than "my" SF universe. And most surprising to me, the SF-themed titles among libre software games are appallingly few. Of the ones I mentioned above, only oolite is SF; all the others are, well, not. Where are the good SF games on Linux?!

Finally, on the "theme" topic, you may notice that most of the games I mention above involve at least some amount of combat between opposing parties. Now I'm no pacifist; I believe the best defense is a good offense, eye for an eye, tit for tat, nuke 'em from orbit and all that. Nuke 'em til they glow, then shoot 'em in the dark! That being said, I prefer games that are not just all combat all the time.

That's one reason I don't go for the hugely popular first-person shooter genre. I played Castle Wolfenstein and even a bit of Quake, but quickly grew bored with the constant unending stream of no-mind enemies I had to explode, perforate, dismember, or otherwise destroy. I played the free non-combat game Search and Rescue and found it modestly diverting, but even though it has clear missions and semi-nice graphics, it just didn't light my fire somehow. So combat will probably be part of any game I enjoy, but will not be the entire game. There has to be some planning involved too, strategy in addition to tactics.

Obviously, if I want:

Then I'm going to have to write them myself. And when I write software of any noticeable complexity, I prefer to use Ada. So far, the GUIs available for Ada have not been up to my standards, so that's what I'm working on right now: A GUI for Ada. After that's working, then I'll start on a game or two. I'll let you know!

Posted at: 00:00 | category: /Software | Comments (0)