At any given moment, there are literally tens of thousands of gaming servers on line, with tens of thousands of players slugging it out in their virtual worlds of choice. Counterstrike (CS), which I think sucks rocks, is probably the most familiar of all such games these days, but it's only a drop (albeit a large drop) in the bucket. Games like Medal of Honor: Allied Assault, Return to Castle Wolfenstein, Battlefield 1942, Tribes 2, Unreal Tournament 2003, Quake III Arena, and a slew of others all thrive on today's gaming servers.
So where do I get off claiming that multiplayer gaming is a failure? I admit at the outset: my claim of failure has nothing to do with the quantity. What I do claim is that the quality of the multiplayer experience is absolutely horrible in most games on most servers these days. Worse, the very designs of many of the games serve only to exacerbate the underlying problems. In this article, I will explain why I take multiplayer gaming to be a failure at present, how the various games virtually guarantee this failure, and suggest the couple of things I can come up with as ways out of the current situation. It is my hope that others will explain to me why I'm wrong, or perhaps that some entrepreneur(s) will seize what seem to me some pretty neat opportunities.
One reason I claim that multiplayer gaming is a failure is because of the quality of the experience (QoE). Take Battlefield 1942 (BF1942) for example. I was playing that earlier this evening, and I was having a pretty typically bad experience playing it. The reasons were the same reasons, really, that account for most of the bad experiences I have with multiplayer games. I was playing the "Battle of the Bulge" mission on a server to which I had a good ping, although that latter fact doesn't mean much with BF1942; i.e., its networking code really sucks compared to other, modern multiplayer games.
When I joined the game, my team had only a single control point left to it, and it had that, I suspect, only because it was a control point that cannot be captured. I selected my kit (assault), specified that control point—a question, developers: why do I have to select a control point when there is only one available? Hmm?—and got into the game within a few seconds after connecting. We (the Axis) were hemorrhaging tickets pretty badly, and I wanted to charge one of the other control points as soon as possible to help stem the tide.
Unfortunately, I spent less time playing than I did picking out my kit. The reason was simple: two enemy tanks were parked up on the hill above our remaining control point, shelling our only spawn position. Needless to say, anyone who spawned died within a handful of seconds at most if not, in fact, instantly upon spawning. This sort of thing happens in most games at one point or another, and I normally dismiss it as just part of the game. Oh, it pisses me off all right, but I realize it's part of the game, and I don't normally get too bent out of shape about it. In BF1942, however, it seems to be the most prominent tactic on every server I've visited tonight. And unlike other games, which typically have multiple spawn points in even the worst case, it rules the day in BF1942.
The net result was not surprising. Not one of my teammates managed to spawn and get off more than a single shot against the enemy tanks. The remainder of the game lasted but a few minutes, during which I spawned and died, spawned and died, spawned and died, along with all my teammates as our ticket counter nose-dived toward zero. Naturally, the enemy won the game. Gee, what a good fight they put up, eh? Gosh, what an accomplishment. What a feeling of satisfaction that must give. Sheesh.
Still, that's how it goes sometimes, and I didn't think I would find anything different on any other server—isn't that a pathetic expectation in itself?—so I stuck around for the next map. The server changed to some other ground-based map, and I quickly found out that the llamas on the server weren't confined to the other team. I spawned with an assault kit, and I had just enough time to holler for people to follow me before I died. The reason? A teammate with an anti-tank kit had fired his bazooka into the middle of the group, killing several of us instantly (me included, naturally) and wounding several others.
His name was "Embark", for sake of reference. And Embark's every spawning quickly became a matter of public knowledge, thanks to the various obscenities and calls to ban him that would spew forth over the global chat as he kept team-killing with whatever weapons he had at hand. I haven't played BF1942 long enough to know whether there is some way to kick a player by voting him off the server, but apparently that feature doesn't exist, or perhaps nobody else knew any better than I. Either way, it was essentially a game-ruining element. My team was quickly defeated, in so small part thanks to Embark, and that was all the time I was going to spend on that server (and that game for tonight at least).
I switched next to Unreal Tournament (UT), one of my all-time favorite first-person-shooter (FPS) games. I've been having system problems with it of late, and I need to find out what's causing them. Unfortunately, I made the mistake of not checking for anti-cheat mechanisms on the server I joined, for it was a server with at least one (or more) persons using some sort of aiming bot. To be clear, I'm not some UT newbie who cries foul at every headshot. But when a single player on the opposite team has a frag count in the hundreds in the first few minutes of a game, and can spawn-snipe my teammates one after the other again and again and again at different spawn spots while racking up headshot after headshot, I start to get suspicious. Disconnecting and reconnecting as an observer confirmed it as I followed him on the map; i.e., it didn't really matter where he was aiming, somebody's head would blow apart with every shot.
I wish I could say that these kinds of problems are confined merely to BF1942 and UT, but they're not. Team-killing is featured by any game with a friendly-fire setting and even some that aren't. CS features all sorts of nifty wallhacks, bots, and other ingenious ways to cheat. Frankly, the couple of multiplayer games I've played that don't seem to have a serious problem with cheaters of these sorts are the original Tribes and its sequel Tribes 2. Those games suffer from rampant team-killing, but at least they have mechanisms to kick the offenders.
The short summary is that pretty much every game I play has at least some cheaters, team-killers, spammers, whiners, or various other sorts of llamas in it. All of these different "types" of players seem to add a single thing to the game, namely, they degrade the QoE for all others playing. Some (e.g., the spawn campers of my earlier BF1942 games) actually seem bent on pissing off people as much as possible. Others (e.g., CS wallhackers and aimbot users generally) seem interested only in coming out on the top of the scoreboard—even though it has nothing to do with their skills and everything to do with the cheats they employ. Whatever their intentions, however, the result is always the same: it helps ruin the game for everybody else.
You know the type. When I use the term 'cowboy' in this context, I refer to those persons who care only for where their names show up in the end-game screenshot. That's fine, of course, if you're playing straight-up kill-or-be-killed deathmatch, but all of the truly interesting games and game modes these days involve teamwork in one fashion or another. Some games are clearly designed to encourage teamplay to greater or lesser degrees, but even the very best cowboy fragger in a game like CS is likely to go down when faced with superior numbers. The fun of so many of these games lies not in individual accomplishment, but in team accomplishment.
At least, that's how it's supposed to work. A game like Return to Castle Wolfenstein (RtCW) is clearly designed with this objective in mind. The different classes (e.g., medic, engineer, soldier, etc.) are all individually weak but mesh perfectly when brought together as a team. Other games like Tribes 2 (T2), for example, push that even further, including weapons designed specifically to be used as part of a group, vehicles that require pairs or trios of players to operate successfully, and so forth.
The problem is that it works very rarely. When I was playing T2 last night before I went to bed, for example, I was "lucky" enough to get hooked up with what was seemingly a team stuffed to the gills with cowboys. The enemy infiltrated our base and took out our generators, yet each and every time I called for help in retaking our base I was rebuffed. My teammates were too busy team-killing, switching teams, standing around tossing various insults at each other, calling for the vehicle station to be repaired (apparently some didn't understand that no generators means no vehicles), or trying without even the remotest hope of success to take the enemy flag—individually, of course, so they could get mowed down by the nicely coordinated defenses at the enemy base.
I originally joined the game because I was hoping to practice flag-grabbing from a shrike fighter aircraft, but when I saw the need for defense I quickly switched roles. It seemed, however, that I was the only one who either saw, or perhaps cared about, that need. Whether that assumption is correct or not, I spent the rest of the game trying to dislodge the two heavies that had managed to pitch their tents in our basement with a remote inventory station to keep them well-stocked. I never succeeded, and our team lost within but a few minutes (none of my teammates seemed very interested in guarding the flag either).
Like I said, I expect this kind of every-man-for-himself thing in Quake III Arena playing deathmatch. But I can't for the life of me understand why somebody would buy a team-oriented game like T2, BF1942, etc. and then play like a cowboy, regardless of circumstances. No, that's not quite true, I do understand it as I'll get to in the next section.
There are two reasons why these things happen in my view. In roughly increasing order of influence they are: (1) game designs encourage such behavior, or at least fail to discourage it adequately in most cases, and (2) some people are just jerks. Regarding (1), if a game rewards the team-killer, the incorrigible spawn-camper, the wallhacker, the bot-user, and so forth, then some folk will always choose those courses simply out of the desire to win. To be clear, I don't think there's anything wrong with wanting to win, but wanting to win at any cost is misplaced at best.
BF1942 absolutely rewards spawn camping. As I've mentioned in my review of the game, the smallish maps frequently become nothing more than extended spawn-killing fests in my experience. And it's not even a case of the victor being determined by which team can set up camp first, for those maps also happen to be the maps on which one side starts out with all but one control point. The team with all the control points has a huge advantage, simply because of the geography of the situation. If they all spawn at the control points closest to the lone enemy spawn point and then make a mad dash to it, the enemy is essentially screwed permanently. The game is effectively over within seconds.
BF1942 doesn't really reward team-killing insofar as the team-killer hurts his team's chances for winning. He may also attract the ire of his teammates. Nevertheless, it does nothing to discourage it either. Like I said, I haven't even found any mechanism by which the players can vote to kick someone. I assume that a server admin could do so, but I've yet to see a server admin available on any server, so the point seems largely moot.
I've done a little work in the game development industry, and I'm not at all ignorant of software development generally. There's no reason the game engine couldn't track the last few kills made by a player and auto-kick him if he's consistently killing teammates. Similarly, several games have built-in voting features that allow the players to kick a team-killer, change the map, etc. Where spawn camping is concerned, any design that leaves a team with only one spawn point is practically begging for this kind of abuse. That's just plain stupid in my book.
Of all the myriad games I've played, the very best design with respect to QoE is found in T2. Some of the weapons are best used in groups, the targeting laser guarantees pin-point accuracy to those willing to work together, the vehicles will be largely useless without good cooperation, a voting system exists for kicking jerks and other such tasks (even nominating a player to the position of admin), and the design of the game is so well thought out that I'm not aware of a single bot or cheat for the game. In short, I've yet to run into anybody cheating in T2, and facilities exist for handling most QoE-degrading behavior. The only serious problem in T2, in fact, is team-killing, and that can usually be dealt with by voting.
Of course, even the very best game designs can still be exploited. I was playing a game of T2 just a few days ago in which my team was up by a couple of dozen points as time was running out. It was a hard-fought game indeed, and I was really looking forward to the victory I thought we had earned. With mere seconds remaining, however, a player on the enemy team was voted admin, and his first act was to extend the length of the game to avoid defeat. In the confusion surrounding why the game didn't end, the enemy team managed to break in and out of our base with the flag, and as soon as they scored, they initiated a vote to change the map and claim victory. The vote didn't pass, but the damage was already done. Our victory was pre-empted essentially by administrative fiat. I left the game in utter disgust at the llamas who pulled that move.
In short, I think that even the best designs I've seen to date can be abused. What is certain, however, is that the majority of designs are utterly pathetic in the way they encourage QoE-degrading behavior and fail to provide facilities to put a stop to it. None of the changes I've mentioned would be particularly difficult to code, any more than any of the known mechanisms are difficult to code. There is just no reason for such poor attention to QoE. Developers can and should be doing more.
Jerks are part and parcel of human existence; i.e., as long as there are human beings, there will be jerks. I know that's not very politically correct amidst a population so sympathetic to secular humanism, but it's what the empirical data suggests. Whatever else you may think about Christianity, I think the notion of original sin and man being a fundamentally twisted thing from birth enjoys overwhelming support from common-sense observations. This much I take as certain: such a view is clearly closer to the truth than the gosh-we're-all-basically-good pabulum that seems to rule the day. If you doubt this, watch the evening news for a while.
At any rate, the jerks of the gaming world seem to enjoy nothing more than degrading the QoE for other players. This is alien to my nature, so I do not fathom what gratification they must obtain from team-killing, exploiting map bugs and other design flaws, using bots, and so forth. Nevertheless, it clearly holds some allure, for there are many who engage in it. I can make but two suggestions to developers regarding the handling of such persons.
First, I think all games should assign unique IDs to players. It is a great strength of games like T2, Half-Life, etc. that they have warrior IDs/WON IDs/etc., which allow the banning of a player completely. Many games (e.g., UT) supply only banning by IP address, hostmask, or similarly mutable features, and this simply will not do. Admins desperately need the ability to ban a player, regardless of what name, IP address, or other parameters under which he might connect.
Second, tools need to be provided for dealing with such people even when an admin isn't around. Some games allow voting a player to be admin, and this is a good start. What would be even better is a feature that logs all kicks/bans/etc. for review by server admins, or better still a feature that would send such information to an email address for review. I always applaud when a team-killer gets kicked, but the joy is typically short-lived because he generally just shows up again and continues under a different name. I suspect that if the logs of such events were brought to the attention of server admins, then more of these jerks would get filtered out right off the bat.
In truth, I'm not sure there is a way out. As long as any jackass can plunk down his $40 for a new game and get connected to play, we're all going to have to live with jerks. Developers can do a very great deal more than they are doing to help improve the overall QoE in multiplayer gaming, however, and I hope that they will. All it requires is for consumers to hold their feet to the fire. Beyond what I've said already, however, I think that one of the most promising ways out of the current mess is to bring the community together against the idiots.
I can see this happening in a couple of ways. A good first step, I think, would be for some kind soul to build a web site dedicated to the exposing of jerks, along with the dissemination of information necessary to avoid them. For many games, of course, this isn't possible. But for any game with unique player IDs, it should be a straightforward enough thing to build a community around such a thing. As it stands, all the very best gamers—both in terms of their skills and their level of maturity—are in clans (in my experience). To an extent, the various competitive ladders provide the best multiplayer QoE I've seen to date, but they are limited to official ladder events.
I'm talking about a site that is dedicated to the larger gaming community. My vision is of a site that lets an admin easily download the current ban list for the games he hosts. Perhaps someone could take the time and trouble to develop a utility that will do this automatically, downloading the list, parsing the details, and updating any server configuration files on the local drive. Such a toolkit would both provide a strong disincentive for being a jerk and make it much easier for server admins to band together to exclude known llamas.
There are a number of non-trivial issues to be addressed, of course, in such an undertaking. What is the standard of evidence for accusations of cheating? I've been accused of cheating several times in UT, for example, simply because I'm a good player. It wouldn't be right to allow someone to be black-listed on the mere word of another. Further, what process (if any) should exist for jerks who have "seen the light" to reintegrate into the community? Does a name come off the list in a month? Six months? Never? I believe that simple methods exist for addressing all of these concerns, but that's a subject for a much larger essay.
An even better solution, and a more likely one in my view, would be for some entrepreneur to start his own gaming network. And by this I don't mean a network of servers, rather, I mean a network that provides server lists. For sake of example, call it the Nojerk Network (NN). The way I see it, NN should be free to all admins, so that any admin who wants to his server listed on NN should be able to get it listed with very little trouble. In so doing, however, he agrees to abide by NN rules and regulations governing the banning of jerks.
Players wanting to play on the NN network pay a monthly fee, and in exchange for their money they get to point their clients at the NN master servers. By this simple subscription mechanism alone, I would bet that the vast majority of jerks simply disappear. The fact that it costs nothing to connect to pretty much any old multiplayer server plays no small role, I suspect, in the great abundance of jerks in gaming. If they had to pay money to be a jerk, I suspect that many jerks would be filtered out immediately.
Since some people are big enough jerks to pay money to piss others off—note well the popularity of "dunking" booths at local carnivals, for example—NN could feature services much like the community described previously. If a player is caught being an ass then several options are available, from simply reprimanding him publicly and revoking his privileges temporarily, all the way up to permanently banning him from the service and listing his unique ID in all the server ban lists (which all admins agree to as a condition of their server being listed on NN). I believe such means would be very effective in curbing the on-line jerks, and I, for one, would be downright eager to pay any reasonable monthly fee (I'm thinking a handful of dollars at most) to have access to such a network.
To be realistic, nothing is ever going to produce a gaming utopia; i.e., there will never be a gaming community bereft of team-killers, wallhackers, bot-users, and so forth. But I do believe that the average multiplayer QoE can be raised dramatically if only developers would add a few simple features to their games. I further believe that bringing market forces to bear, or perhaps the stigma generated by a thriving community, can go further still in rendering multiplayer gaming a true joy more often than not.
Let's face it folks, time is the one resource that we do not have in abundance. Each of us has 86,400 seconds in each day, and the days of our lives are just as numbered. When it's time to kick back, relax, and play some video games, I'm looking to have a good time playing hard with a good team. I'm not interested in cheaters and other sorts of jerks. Such things degrade the QoE for all of us, and that's something we should all work toward changing. I hope this essay will help make it happen.
12/09/2002