Online Multi-Player Games

ON-LINE MULTI-PLAYER GAMES, by Dani Bunten Berry, 1996

What new opportunities and challenges impact the design and development of on-line versus traditional computer games? People. Although there are other features of the on-line world that will have a part in future games (such as massive databases and easy distribution), nothing will have the impact as designing for people to play together.

The Potential … The thing orientation (as opposed to a people orientation) implicit in computer games has for years excluded all but a small demographic subset of the population from playing. "No one on their deathbed says 'I wish I had spent more time alone with my computer'!" This is an attention grabber I have used for years when giving talks to game designers about the potential of multi-player games. Although it's a somewhat fatuous statement, it points out that all of us value people in our lives. Even the stereotypic "geeks" that have been the mainstay of the PC games business for years have ranked multi-player ability second, just below originality, in surveys asking "What features would you like to see in future games?" When you include the more mainstream, casual players who are currently coming into the PC market, it's evident that products that have a people orientation will become the growth area for the industry in general. On-line multi-player games are positioned to take advantage of two major trends - the desire for personal freedom and the desire to share time with other people. With millions of new users signing on-line daily and looking for neat things to do, the potential of on-line entertainment is boundless.

The Challenge … What about the challenges? We don't know how to design good on-line games. That's the long and the short of our challenge. There may be other problems that will surface once we have learned to build really great products (such as: how to market them, how to charge for them) but first we need to know what they are. To date little has been learned about designing multi-player games for the computer since very few attempts have been made. However, despite this paucity of data, the history of multi-player games in both on-line and "off-line" (floppies and CD's) environments should be examined. Studying what was tried, what succeeded and what failed might offer us some insights for the future.

Off-Line Games … In the off-line field, multi-player games have been developed for various platforms (personal computers, game consoles, arcade machines) but compared to the whole industry the number produced is minuscule. Even when multi-player capability is offered it's simply one feature among many and not the main focus. The conventional wisdom has been that multi-player games don't sell well relative to solo games and whether this was a self-fulfilling prophecy or true data point is irrelevant. Very little experience has been gained. Of the attempts made, the overwhelming majority of multi-player products have been sports games (using multiple joysticks for player interaction). With older computers that supported more than one joystick (C64 and Atari) football, baseball, basketball and hockey were quite successful and conceptually easy to implement. Olympic events (Decathlon, Summer Games, etc.) and golf were also popular and worked with a single input device (via passing the joystick or taking turns at the keyboard called "hot-seat" play). On the cartridge and arcade machines more aggressive sports like boxing, wrestling and karate predominated. When these types of games were done on the IBM PC the typical input technique used a split keyboard (with each player using definable cursor keys). Although other types of games were also tried (such as trivia games, game-shows and original designs) none of them attained much market penetration. And even the sports games success did not rely solely on their multi-player aspect since all of them also offered solo-play capability.

Why They Failed… Several rationales have been offered to explain the lack of success of multi-player games in the off-line games business. I consider the most persuasive explanation to be simply the logistics. In the case of computer implementations, the computer is most often in isolated location in homes which makes gathering a group of people together to play difficult. Even the console/cartridge machines which are frequently in the family room have problems too. Extra joysticks and adapters are usually required over the base machine. The cables are usually short and the displays have poor resolution which requires close seating (on the floor instead of the couch). Finally, arranging for people to get together at all requires invitations and certain accommodations such as straightening the house and preparing snacks to entertain them properly. (These constraints, by the way, do not apply to on-line play and once again point out the potential on-line entertainment has - painless play!)

On-line Games … The history of multi-player games attempted for the on-line environment has been much shorter and more casual. From the start it was apparent to all proprietary on-line services (Genie, CompuServe and AOL) that multi-player games were the only type that clients wanted. In fact, the intense loyalty of their early adopters may have stifled innovation because of their willingness to pay to play games that weren't "state of the art" or even original. Only now are the proprietary on-line systems beginning to bring their development budgets up to the level of the off-line market. Innovation is still not a priority for them. Nonetheless, some products have succeeded enough to warrant continuing upgrades and expansion (Air Warrior and Gemstone) and even the spawning of games-only on-line services (ImagiNation Network and various MUDs).
The successful products that have been offered so far seem to fall into three general categories: conversational systems (with various amounts of added paraphernalia), action simulations and conventional games. Another only somewhat facetious way to describe the first two groups would be: "stories with props" and "playgrounds with toy guns". This latter definition of the categories points to the two main themes that resonate with the participants by somewhat mimicking childhood experiences (and as any good game designer knows, what type of play resonates with a given player is utterly dependent on this formative background).

Imaginary Play … The chat/story is like imaginary play where kids assert an idea verbally that other players witness and embellish. This type more than any other appeals across gender lines. It's the one area of on-line entertainment that has a significant female presence. This type of product offers considerable flexibility in the way players present themselves as well. Some anecdotal evidence indicates that many people use this area an outlet for their alter-ego. In the case of adolescents at least, this has led at times to experimentation with different persona types that later get incorporated into the person's character. This is a significant accomplishment for products whose main intent is entertainment and if the finding holds up, it makes designing good products both a worthy achievement and a somewhat daunting responsibility.

Playgrounds and Guns … The playground/toys type product is more structured with an environment drawn from "reality" and is more oriented to the equipment provided in that environment. This group contains the current flock of "shoot-em-ups" that are primarily played over LAN's (Doom and it's ilk). Air Warrior was probably the first successful product of this type but it had a number of technical problems. (Planes would "warp" past each other because of latency). However, as processors, graphics cards and modems all get faster the technical issues will be less significant. There are a number of points in favor of products like this on-line. They are easy to play and typically have high audio-visual production values. They also connect well with the males who currently predominate in the on-line entertainment market. However, there is some early indication that products that were fun when played over LAN's with coworkers won't necessarily transfer to an anonymous on-line environment. More "content" will evidently be needed to maintain interest. I expect the playground/toys products to add strategic elements and become more like games than just simple simulators and shoot-em-ups.

Old Time Games … The final category of products offered on-line is "games" (with rules and winners) but to date very little original design has been tried of this type. Most of the current on-line crop of true games (such as checkers, backgammon, etc. as offered on ImagiNation Network) could reasonable be seen as extensions of "chat" since the game itself matters little. However, I believe this category will become one of the largest in terms of revenue per product in the on-line entertainment field. In the off-line field strategy games have had the "longest legs". Games like Empire, Civilization and my own M.U.L.E. have shelf-lives of years and are played over similar spans of time -- especially the multi-player ones. Also, games that include and enhance social interactions existed long before computers and thus have a cultural role to play that can be re-established in cyberspace. Such products could vary widely and still qualify as games as long as they include a beginning, middle and end and have a winner (or winners). These games could be played by groups as small as 2 to those as large as thousands (much as games were played historically: at family sizes up to whole villages). Such products could be "head to head" competition with each player for himself or herself, they could be "team" competition or they could be a "hybrids" including both competition and cooperation. The thematic content could be sports, fantasy, economic, military, political, video kinesis (pure video game) or a combination of any of these.

Designing for the Future … Taken together these three product types offer endless possibilities. What's needed are designers with a particular vision and companies willing to let them try out ideas. Play-testing by trial and error will be the only way to uncover what will work with real players. Here are a few specific considerations that can help any game work better as a multi-player experience.

For most products, provide for alternating phases of high intensity and low intensity. Continuous action can become overwhelming and eventually causes burn out. Slower pace segments in the game allow players to reflect on their progress (an important part of letting players glean a positive image even when they're not winning), plan for the future (anticipation makes all experiences more compelling) and to chat with their playmates (it's why they chose to play on-line rather than just "solo").
When possible, end the game after a certain amount of time. Time limits let players schedule their involvement to fit their schedules and their budgets. Games that drag on and on will let players lose interest and drop out (a disruption to the players) or they finish this game but don't play again (a disaster to the designer). It's better to have players finish a game and ask for another play than to go away exhausted and not come back. Games that last 30 minutes is a good goal for a single play and 2 hours should be considered a maximum.
Allow for player directed play-balancing that handicaps participants so they can play head to head even when they have disparate skill levels. If this seems problematic then find some way to protect new users from being beaten up by experienced players. Nothing can kill a new player's interest quicker than losing!
Include some way for players to "save face". In most games there will be more losers than winners. Find a way to make the losers feel better. The addition of randomness and chance to a game can let players rationalize their loss.
Offer "game replay films" that let players re-view their performance. Since almost all on-line games are minimizing the amount of data sent between players by sending only the 's (delta's) it's usually possible to store the stream of data sent and received and to re-play the game by simply running that data back through. This gives player a chance to study their performance and that of their opponents. This can let players explain away their loss or savor their victory. More than anything it lets players turn a fleeting and ephemeral experience into a story they can share or even a legend they enshrine.
Whenever possible extend the middle area where players do not know who will win and who will lose. This will keep players from becoming "spoilers" or worse still from dropping out. If you have random events you can balance them so winners get only bad luck and losers get only good luck.
Create incentives so players stay till the end of the game.
Allow spectators to watch games being played and even include the option for some spectators to be play-by-play commentators.
Test it, test it, and re-test it with real users to see if it works!

-Go Back Home-