omf.com - rob elam's log: game design

Rob's Log - Game Design

Introduction

Here's some ideas I've personally gathered from my playing days and how they can apply to most games. It's essentially a checklist for making a good game. The document is a couple of years old, but the fundamentals are still true.

Technique

This is the most important part of a game if you want it to stand the test of time. Few games have staying power if they depend solely on their graphics, sound, bulk, flash, or whatever. What I really mean by technique is that the player can be given a simple set of rules and abilities but there's many different ways to use these abilities to accomplish their goals. Here's some examples of popular games and how this applies to them:

Tetris: On the surface this is an extremely simple game with simple controls and rules. But there are basic techniques that you learn to open access to obscured blocks or whatever. The reason I bring up this game is that it has VERY simple rules, and few of them, but a relatively high level of technique. Other good puzzle game examples are Boulderdash and Lemmings. Another thing is that when the player is given a simple set of rules and he discovers techniques and strategies completely ON HIS OWN, it's a great accomplishment. He feels like he solved a puzzle he wasn't even supposed to and since he discovered it on his own, he may think he's the only guy in the world that thought of it. It makes him feel special, smart, and in a small way, powerful.
Civilization: This has been a consistent seller for many years and because of one thing - Strategy. Civilization is technologically lame, graphics were average at best even at it's release, sound relatively poor and the interface is cumbersome on a good day, but still a BIG game. The game had fairly complex rules, but nothing compared to the strategy and techniques available to the player. You could concentrate on trade, war, science, expansion, etc. and change your government to facilitate whatever your strategy was. All techniques seemed just as valid and effective. The cool thing is that players would argue that their techniques and strategy was the best (more than one way to skin a cat). It's like real life games. Take basketball as an example, just put the ball in the hoop (and a few other rules), pretty simple. But the things that real life allows you to do makes it possible to create your own techniques. The best games, IMHO, are the ones that capture as much versatility as possible with a simple interface.
Doom: Doom has about the best technique available in 3D games. Each weapon has a high level of technique and balance, and are somewhat different from each other. This is an interesting example because that as you get new weapons they don't necessarily make the lower weapons obsolete. For instance, if you're real good a shotgun can be as effective and assuredly as gratifying as a plasma gun. IMHO, this is how a game should be. If the weapons or abilities are stepped (found in some sort of "power" order) they shouldn't really be replacements and definite successors to other weapons. They should compliment each other, not replace one another. These things, along with the other cool things in Doom are what keep people coming back. Doom could, however, go much farther in the realm of weapon and fighting technique.
Street Fighter II: You know you had to hear about this one!!! IMHO the designers of this game unwittingly (during development) stumbled upon a great concept. The combo system created a great many secrets (combos) which the designers didn't have to specifically program for (at least not much). It wasn't like in Doom when they make a secret area or secret weapons, characters, moves, etc. They made one aspect of the system loose enough so that if you were experienced and knowledgeable, you could learn these combos to more effectively and with greater style beat your opponent. This is very important to the pride filled, egocentric, highly competitive adolescent male, which is the primary customer for action games. This is why there have been so many copycat fight games and why fight games are so popular, IMHO. Not because of the side view fighting, but because you seemed to be constantly learning new things (however subtle they may be). This aspect allowed the game to "grow" far beyond what the designers intended.

Flash/Atmosphere

These things are not my primary concern, but I assuredly know how they can impact a game and how important they are to our fickle audience. This is the part of the game that initially captures the player and temporarily hooks them. After the hook stage, you have to nail them with a good game or they will move on, but we're not talking about that now. Here are some games wherein this should be taken note of:

Myst and 7th Guest: These games didn't excite me personally, but the atmosphere and graphics in both were attractive and intriguing enough to make them both huge hits. These two games were fairly well thought out with decent to good puzzles, but the flash is what sold them, IMHO, and that's why I brought them up.
Doom: Again, personally I didn't appreciate the atmosphere, but I recognized it's impact. It had a dark feel, unobtrusive and moody music, very effective and immersive sound effects, gratifying actions (shooting, getting hit, etc.) and often startling surprises (the first time through, anyway).
Command and Conquer: I'm not bringing this up for it's atmosphere (it really wasn't that remarkable) but for the flash. Every part of that game seemed to have some cool FMV going on and everything was just really highly polished and attractive. The one player game showed VERY high quality cut scenes that really drew you into the plot and drove you to finish the next mission (to see the cut scenes). The game was top-notch too, so you didn't feel that the cut scenes carried the game. Also, this game did have notably good technique, just didn't mention it before :).
Mortal Kombat 2 and 3: MK had some good graphics, dramatic moves and characters, and good sound. The overall atmosphere of the game is the best of the fighting games, IMHO. It�s super crisp and realistic graphics and dark, stylized environment have really made the MK series a big hit.

Gratification

This refers to the "treat" the player gets when he does something right. In the past, games could get away with a simple score bonus, but that doesn�t cut it in todays market (many traditional score driven games like platform and fighting games now often don�t even have a score). Now you need some good graphics, sound, something to reward players for doing something right. Examples:

Mortal Kombat 3: This game just feels good when you pull off a hard combo or an uppercut. Even after doing the same combo 50 times it still feels good getting a player one more time. Killer Instinct and Doom also had high gratification levels for typical game actions.
Mega Man: I only played 2 of these games a great deal (MM2 and MM3 on the NES, I believe) but one thing I remember about it is that after you beat a boss, you generally got to use his weapon. This is a big bonus. Now, I�ll admit that many games these days give you weapons after beating tough opponents, but few let you see the weapon in use before you get them. For its time, I thought this was really cool.

Control

This is an area which many companies miss the boat, making a great game with either poor or overcomplicated control. I'll talk about some examples of good and bad control in fighting games:

Street Fighter Alpha: Excellent control. A fairly simple interface and you seldom find yourself trying to do a special move and it simply not working.
Mortal Kombat 3: Fairly responsive, but they added a button called RUN. All this button did was allow you to charge forward (not even back). The move was useful, but just as easily done from a controller movement. This was an unnecessary button, thus an overly complex input system.
Rise of the Robots: Humph, well, let's just talk about control problems (I can't list all the problems, not enough RAM to write them :) ). They made an action game (they called it a fighting game!) wherein you hold a button down then release for all moves, taking up to a full second to charge a move. An action game that requires this much time to execute a move is not an action game.
Crusader: No Remorse: Just played this game and it's good. Biggest problem with the game is it's control. It's not really responsive and overly complicated. They have a mode of walking wherein left and right turn your guy and forward and back move you forward and back, basically like Asteroids, and it doesn't work in an isometric view. The other mode is a more conventional directional key walking method but when you switch to firing mode your guy rarely targets the appropriate enemy unless you're lined up with him. I found myself constantly switching from keyboard to mouse to aim my gun. This, and many other obvious control problems deeply hurt what was an otherwise excellent game.
Doom: Great control. Only problem I have with it is there's seemingly no reason for the run key (who doesn't use it?) and no way to toggle it.

Learning Curve / Out-of-Box Experience

Some games have an exceptionally high learning curve, like flight simulators, intense puzzle games, RPGs, that kind of thing. In the past they were accompanied with manuals you were to read before playing. IMHO, this day and age if I ever have to read a manual to play a game, my opinion of the game immediately drops. There's no reason to need a manual and I think the general public feels the same way. Here's some good and bad examples:

Flight Sims: The last flight sim I played a great deal of on the PC was Strike Commander (unless you count WC series). It required a manual and had a fairly large number of keys to remember. This is not necessary. In a game like this, do a tutorial. With CDs and FMV, you can do a really nice one that the player will not only appreciate but enjoy. Have on-line help. In early levels of a flight sim you can have a wingman or co-pilot tell you what to do and what to look for (which I love). These things may seem like a lot of work, and they are, but it makes the player much happier.
Super Metroid: This was a very well designed game and they did a great job of showing how a new weapon or ability was used. Without text, voice, or a manual, I easily figured out how to use the weapons and abilities I acquired. They'd either give you clues about it's use (having a creature use same ability) or immediately require you to use the weapon or ability to get out of the place you got it. Not only did this work really well, it was a little mini-puzzle for you to solve.
Crusader: No Remorse: I had to read the manual for an action game!!! That's all I have to say about that.
Strategy Games: Strategy games often require manuals because of the complex things that you need to know about in the game. The really good ones overcome this by either showing cutscene demonstrations of the weapon/unit/ability (like in Command and Conquer) or simply having good in-game documentation or help (even Civilizations had a pretty good in-game manual). With CD-ROM and Windows Help functions at your disposal, one of these things can and should be done.

Expandability

With the PC, players and hackers have a very valuable asset when it comes to games; changeable and transferable files. This means that a good "hacker" can alter game files and continuously change the way the game operates. Most importantly, this is done without the intervention (work) of the games design team. "Hackability" can greatly extend a games life and enhance its money making capabilities. The most obvious example is Doom. ID really didn�t do much to specifically make the game hackable, IMHO, they just designed a system that best fit what they wanted with a centralized file system and the game was so popular that people hacked it apart and made editors for it. Now at the least, I think if the game type permits, the designers should make the game files as easy and as hackable as possible, if not releasing a scaled down editor for the game. This gets the public working for you to prolong your game�s life. People might argue that it makes it harder to sell add-ons with this scenario, but I think the effects would be negligible. Even if it made a big difference, it�d be offset by the new customers that�d be generated by the longer life and increased market penetration caused by the hacked levels, or whatever. Also it�s my opinion that a well designed system should allow practically any game element to be changed, from maps to the look/animation/action of a single bullet (I think that very little of a game�s playability has to be hard-coded).