The Essential Scatter

As fun as I had last time around in my guild’s massive Gloamwood event, I noticed a flaw. Or rather, I saw the flaw in another form. It’s a unique flaw that has been appearing more in the age of public grouping. Let us call it “the zerg.” The zerg is a group of overwhelming force of otherwise unimpressive individuals, and a zerg in an event usually emits a strong gravitational pull entrapping other players. It’s not a unique thing, as its been for as long as there has been open world PvP (if not longer). Yet, it comes across as something different, possibly fouler, when the zerg’s opponent is the system.

Near the end of the Gloamwood crusade, I was starting to get bored. I was thoroughly enjoying all the camaraderie, but the game was being distilled down to merely following the herd and firing off as many spam skills as I could before whatever was targeted inevitably popped. The system was stretching to the outer limits of its “balancing.” Yet, there were far too many players for it to respond in a useful way. This is when the system needs for players to scatter.

The simplest solution is also the most un-fun. Simply stop the activity. When players get bored, they will disperse back to the capital city, or quests, or crafting, or slightly active zones. There will be an exponential bleed off rate, and the trick for the developer should be finding that sweet spot between downtime duration and zerg weakening.

Having other activities helps. For example, in Rift there are always quests. Most of them are fairly standard and/or boring, but it passes the time between events. With enough downtime, a player will begin to think of the wealth and experience she could be amassing being active. These other activities lower the pain threshold to bleed because there are other carrots to chase after.

Yet, other carrots are hard to design. In Warhammer Online’s early age, a zerg in an RvR lake became an unstoppable force as it rampaged across the tier. As long as the opponents continued to poke and prod in defiance, the zerg would stay on the move as one. The other carrots were bland in comparison to the ongoing RvR event even in the RvR lakes. So what if the piddly opposition took a few small RvR objectives when we were about to fell a keep. There was simply no reason for a task force to peel off. The rewards gained by zerg action far outweighed nearly every other activity.

Another way is to be heavy-handed with event completion. Tell the zerg they have won in a super-significant manner. The more pronounced the reward, the more people are likely to call the zerg quits because “they done good.” With the final mega-event in Gloamwood the other night, getting a purple token was a great time for me to call it a night. It was unlikely I would get a better reward that night, and it caused me recollect on the fact that I had fun.

The hardest solution is to split or kill the zerg. Sure, it is the easiest solution to implement. Just create an unbeatable opponent that one-shots every player or endless mobs until everybody hits the inevitable out of energy. Yet, it’s hardest to handle elegantly. Designers want their game to be fun, and “rocks fall, everybody dies” is not fun. It’s also pretty hard to get individual groups to peel off. Many players are incommunicado and will simply follow the herd. In the Gloamwood mega-event, the final stage is eruption of mobs surrounding the central city, Gloamwood Pines. Theoretically, this was designed to split the zerg, yet I saw it was much easier just to trample each node in turn. If one node fell, it actually became easier for us to defend as we only had to keep up one defense point.

In review, I think Warhammer Online did a fairly poor job. The motivation to keep the zerg rolling overcame everything except non-action. That’s the way I remember nearly every zerg dying, there was simply nothing to do. Rift is doing a fairly good job at using win-points and other activities as ways to disperse at least parts of a zerg. I am interested to see how things progress further as the events become more complicated and more opposed by the enemy faction. Finally, Guild Wars 2 is going to be an interesting data point. On one hand, everything in the open world is an event. Can a zerg just faceroll a zone, or are the zones designs clever enough to continually eat away at the zerg from behind? The biggest unknown in Guild Wars 2 is how map warping will affect event zergs. Will it increase the growth rate of a zerg to a ridiculous degree as news spreads, or will it cause players a very low-pain means to jump out of the zerg for a break. I’m not even sure ArenaNet can know that answer until some sort of large beta.


19 thoughts on “The Essential Scatter”

  1. I’d like to see them give people reasons NOT to bunch up. Right now, in a zone event, everyone wants a piece of the boss for various reasons like the achievement, so if you’re not in that mess, you miss out.

    Instead, make all things equal in some way. I’m sure a lot of players would rather be in smaller, discreet groups anyway. That’s why you may see LARGE groups during the run-up to the event boss, taking down the number of rifts or tackling invasions, but you don’t see everyone massing until the focus shifts to that one location in the zone.

    1. I am hoping patch 1.1 in Rift makes the boss fight less crucial to getting good reward tokens. I mean if you close 4 rifts and miss out on the boss fight spawning on the other side of the zone, you still had significant contribution.

      1. Right, right. I feel that I HAD been getting some DECENT rewards without hitting the boss (in Freemarch, he’s always somewhere I’m not!), but maybe I could be getting better stuff.

      2. Making the boss fight less crucial to get good rewards, that’s the key. If the other events rewarded purple (X) sourcestones as well… well, I won’t say we wouldn’t see the zerg, but we’d see it less often.

        I know it goes against the grain of everything we’ve been conditioned to expect all these years, namely time+group size=purple, but cakes, having, eating, etc.

        1. As far as I can tell, the boss fight not only isn’t crucial to getting good rewards (achievement aside), it’s slmost irrelevant (except insofar as participating counts toward overall event participation).

          I’ve gotten purple rewards without getting closer than half the zone away to the end boss and blues without even being in the same zone when the boss spawned. The key to the purple stones seems to be defense (Wardstones, planar capacitors, whatever) and keeping them defended even after the boss has spawned. No official word on how this works, naturally, but plenty of anecdotal evidence pointing that way.

    2. Right now during some of the tougher invasion events, my guild group defends one wardstone while the zone zerg moves around knocking out rifts and such. Good times (and rewards) for us, plus we stay away from the zerg if we want.

      I’m hoping as events get more complex, this will only increase.

  2. If the zerg is a problem, your game is too easy. Ramp up the difficulty by using aoe attacks so that many players need to sit back, heal, and raise other players, and those players are out of the fight for a bit. FFXI had world events long before Rift with Campaign, and the zerg wasn’t a problem at all with it because the mobs could lay waste to you.

    I’m betting the soul system is letting people tankmage way too easily, hence the zerg.

  3. If the boss fight is too trivial for the amount of people in the zone why not have more than 1 boss spawn at the same time in different locations and they need to be killed within certain limited time of each other or they spawn a new boss to replace the one that has already been killed. More people more boss’s, obviously they would need to be far enough apart that the zerg just can’t run from one to the other within the time limit. This would require the zerg up. The down side is that it would require more communication, if thats a down side at all.

  4. Different context, but same problem.

    In GW1, there’s a PvP arena called Alliance Battles. 2 teams of three squads of 4 players. You get points for capturing flags (which are defended by NPCs), and points for killing enemy PCs.

    To cap a flag, you need to have more friendlies within range of the flag than enemies. This shifts a control meter in your favour. When the meter goes to zero, the flag becomes unclaimed. When it hits the other end, it becomes yours.

    Maximum speed is achieved with 4 characters. Any more than 4 is wasted.

    Optimal strategy is to keep your 3 groups of 4 separated and run around the map, steamrolling flags. Avoid enemy PCs unless you’re sure you can steamroll them as well.

    Losing strategy is to form a mob. Sure, the mob can steamroll one or even 4 enemies, but smart enemies avoid the mob, and you quickly lose out on flag points. The only time mobbing wins is when the enemy chooses to fight it in dribs and drabs (“feeding the mob”), rather than avoiding it and cap flags instead.

    A similar technique might work for un-zerging. Make multiple targets that have to be fought *simultaneously* – as long as any one is active, the others will keep respawning, possibly with increased strength. This requires a degree of PC co-ordination (in that having the whole zerg rush to another site fails), but does cause people to split up across the map.

    Guild Wars uses this to good effect in several missions (eg, and I don’t see why it wouldn’t work with an ad-hoc mob.

  5. Hopefully the waypoints in Guild Wars 2 are spread enough to prevent these Zerg issues. If not I think it will be worked out eventually through beta testing and such.

  6. Having played a bit of GW2 at PAX East, it looked like they were aiming to have a number of smaller events in one area, each event being a significant amount of foot distance from each other with no or perhaps one nearby warp point. Then there’s the really big events like fighting the dragons which will probably have a cool down, or a series of events that reset that the zerg mob will have to warp all the way back to start over and win them all in succession again.

    One solution presented in a rather MMO-lite game, Monster Hunter, is to have the enemy AI consistently use attacks that force the players to run around, not bunch out, and therefore not combine firepower very easily. The game is essentially a number of monsters that act like bosses that then do things like charges and tail flips to constantly split up the players. At the same time, the game benefits from having 90% of the weaponry being close combat, which means the players are forced to move much more easily. While this solution doesn’t necessarily cause the fight to end, it works to keep the players on their toes, so even if they’re doing the same fight again and again, it removes an amount of boredom.

  7. This is an interesting problem. AB isn’t very good at solving the problem, while there are multiple points to go and capture the mob will clear out the posts quicker than the smaller groups and move faster, eating up players. Also the ability to gain points by killing enemies is problematic, and encourages you to kill people instead of playing the format.

    Here are a few things I think might act as solutions. 1) Spread out the objectives. The ability for a zerg to quickly take something down is diminished when they all have to travel from point A to B, slapping a time constraint should help this when dealing with PvE. The distance of objectives should be enough for PvP as it gives the smaller groups plenty of room to maneuver around the larger single group. 2) Assigned objectives. People like to follow those symbols on their compass and do things that they are told. The zerg seems to act like that, all these people going in one direction, to the objectives. I think having an AI recommending what players should do could help alleviate the zerging problem. We have these behaviors learned, lets use that to an advantage. I think Brink might have a system like that, and its an FPS. 3) Make everything objective based. Points for killing enemies is fine and all, but should be a means to an end, especially for PvP. For PvE giving more points for the secondary or multi-primary objectives should facilitate this process. I think thats all I’ve got atm.
    /end armchairdeveloper

  8. Wouldn’t the best trick be to prevent the zerg in the first place?

    If GW2 pans out as they are planning, there will be so much happening across the world at any given time that it is unlikely to attract all the playerbase to one location at one time. (Well, once the game settles down. I’m not about to call anything for the first month or three!)

    With event scaling, GW2 has a basic zerg resistance anyway, but the sheer variety the world offers, coupled with its sidekicking preventing the playerbase all being shepherded to one place, will hopefully prevent that critical mass being reached outside mobs deliberately planned to overwhelm the system, which people will tire of eventually anyway due to the numbers required.

    1. It sounds like event scaling and enemy skill-unlocks will prevent zerging from staying together, while multiple events that all give the same reward prevent it from forming.

      Then again, a great way to split up the zerg is for a Dev to summon an Elder Dragon, thereby one-shotting or taking players temporarily out of the fight, making them run around reviving, freeing players, seeking siege weapons, and frantically searching/updating the wiki. With no death penalty/durability on gear, dying isn’t that big of a deal.

  9. In general I’m in favour of the zerg, or as I prefer to think of it the non-capped raid.

    Consider a 40 man Vanilla WoW raid: you had to have a few people who took the game very seriously but you had enough people that it was ok for some people to be individually unimpressive.

    Over time WoW has morphed into what is basically a 10 man raiding game with a few diehards holding out as 25s. Everyone on the team is brutally scrutinised, before, during and after a boss fight and it’s become a game of Ann Robinson’s The Weakest Link.

    Now if from those 40 man raids instead of condensing down WOW had opened up you don’t need to bully people over their gearscore. There’s no need to kick anyone. Instead if you are trying but narrowly failing you need to seek positives – enthuse people, motivate them, get them excited.

    And the more friends you have the better as even some dopey afk autoshotting hunter is another 2k dps on the boss you’re getting to 1% then wiping.

    I think the genre went the wrong direction – I think big spontaneous tough fights are awesome.

    The only problem I have with Rift is it’s usually too easy – you know you will win it’s just a matter of spamming your way up the contribution meter. I would like to see Rift have a big random factor to the bosses (like Diablo 2) where you kill 5 easy bosses then holy crapola this one is MultiShot Lightning Enchanted!

  10. I thought you were going to suggest that their be a literal scatter mechanic. Imagine if a mob boss knocks you and 30 other players back into another zone or to edge of the event, this would thin the zerg pile. Kinda like Sheamoor Broodmother being mobile and knocking players back with it’s tail. Of course that knockback wasn’t as far as I’m suggesting for bigger bosses.

  11. We often see maps with towers or cannons that can be destroyed. Once gone, it stays gone. This is what allows a zerg to bum rush one location to the next — the finality of it all. Once a layer of the onion is peeled, it stays peeled.

    To get people to scatter in a battleground or RvR scenario, you need a steady stream of events all over the map that cannot be destroyed for the remainder of the scenario. There has to be a pressing need at nearly all times to be in many places at once. Quickly dispatching one location at a time will result in a never-ending game of bop-a-mole that will deplete the zerg’s resources and result in defeat.

    Imagine troop transports landing at various locations and deploying troops. If the detachment is left unfettered, they eventually setup artillery and bunker in. If they come under suppression fire, they have to defend and can’t build (or as quickly). Yes, they can be zerged, but other transports are constantly streaming in. The last detachment to be zerged had enough time to entrench itself and fortify. If the zerg spends all its time destroying the detachments, then they allocate next to no troops to the final goal before the clock runs out, and they lose.

    Imagine an airfield that periodically launches a bomber to do massive AoE vs. PCs. If it comes under suppression fire, the pilots can’t launch. Enemy NPCs constantly spawn to keep the suppressors busy. If the suppressors leave or are overwhelmed by the airfield guardians, the takeoffs continue. It is impossible to permanently disable/destroy the airfield without perpetual commitment of PCs against it. If a giant zerg does nothing but suppress the airfield, again, not enough troops are committed to the final goal.

Comments are closed.