test content
Logo
What is the Arc Client?
Install Arc
Gravitar: a case study in how small steps lead to big problems -- and ways to fix it
fentonator
Posts: 48 Arc User
First off: I'm not picking on any particular dev, or even the devs in general, with this post. If it seems like I am, read again, because you missed something. I do use strong (and negative) descriptions of things that are bad; not only does that not make the folks who did them bad people, it doesn't even mean they did bad design. The whole is very much more than the sum of its parts -- but not in a positive way.
Gravitar. Just about the only folks who don't have some reaction to this content are those who aren't able to play it. A few love it, a much larger number loathe it, and the rest seem to generally just despair over it and avoid it the vast majority of the time.
To fix the problems, one first has to understand the problems. And the problems that make Gravitar such a terrible chunk of game design as a whole happened in small steps, by many different people, over a span of years. Any one of them is perhaps less than ideal, but not the end of the world. All together, they make it quite literally worthy of being a case study in how not to do this sort of content.
So a brief history of the relevant points:
There is a set of factors completely separated from any issue of "balance" which, taken together, make Gravitar into punitive gameplay that goes beyond what even the early "Nintendo-hard" or quarter-sucking video games did (for the most part). If you're not sure what I mean by "punitive gameplay", Shamus Young has an excellent discussion of it at http://www.shamusyoung.com/twentysidedtale/?p=2048 (I could have sworn I *read* this at some point, rather than it being a video, but this is what I could find readily).
What makes it so punitive? The combination of mechanics around death. I don't mean how easy it is to die -- that's a balance question. I mean how it repeatedly and brutally violates the first, last, and only tenet of game design: never force the player to stop having fun. Every time you do, you make the game less fun for them (pretty much by definition); do that enough, and they will decide that some other game is more fun. Or beating their head on a brick wall. It isn't a question of "whether", but of "when". So the best way to handle it is never to start sliding down that slope.
The first and most egregious problem: the lockout mechanic existing at all. There are ways to make something have penalties for dying that don't involve preventing play. But if my reaction to dying is "oh well, guess I'll go make a sandwich", the game has just lost me for at least that period of time, and likely for a lot longer. It doesn't take many of these to make that "forever".
The next is a lack of any warning that respawning (rather than reviving) will lock you out. In fact, if it is going to lock you out, why are you even offering the option in the first place? It is an option that nobody is going to *want* until the entire team is down, at which point you'd be better off just force-respawning the team en mass. But at the very least the game should be signalling that a lockout mechanic is in play, and that respawning is effectively the same as quitting the fight.
To add insult to (virtual) injury, the Gravitar fight respawns you with direct line of sight, so that you get to see all the fun you're not having and feel like you're in a special Hall of Shame because you didn't realize it would lock you out. Now we're not only is the game being punitive directly, but socially as well.
Finally, because of the need for a full ten-person team to take on Gravitar, it often takes a long time for the queue to fill. You literally cannot just "get back up and go at it again" unless every single person in the group decides to, or there is a pending queue -- which, after the first burst of activity when it was released, I've never seen happen.
What amazes me isn't that people dislike this content -- it is that there is anyone who is actually willing to *run* it more than a handful of times in the first place. And guess who that is? Farmers. Because there is literally no other reason to *bother* doing it repeatedly. Because 'bother' is the correct word.
Now, will all of that, the title also promised some possible solutions. See the next post.
Gravitar. Just about the only folks who don't have some reaction to this content are those who aren't able to play it. A few love it, a much larger number loathe it, and the rest seem to generally just despair over it and avoid it the vast majority of the time.
To fix the problems, one first has to understand the problems. And the problems that make Gravitar such a terrible chunk of game design as a whole happened in small steps, by many different people, over a span of years. Any one of them is perhaps less than ideal, but not the end of the world. All together, they make it quite literally worthy of being a case study in how not to do this sort of content.
So a brief history of the relevant points:
- There was no "lockout" mechanic for lair boss fights, initially. Mandragalore and Andrith demonstrate this, unless someone changed them in the past month or so. It was introduced around the time that Serpent Lantern and Therakiel's Temple came out (I think NemCon doesn't have it, but I can't remember off the top of my head).
- All of the initial uses locked a boss fight such that a respawn (rather than a revive) meant you could not re-enter the fight; however, if everyone on the team went down, the battle "unlocked" and the entire team could re-enter.
- The Monster Island and Lemuria crisis missions have had a queue since the beginning; for a long time, however, the first person joining the queue triggered a timer, and if that timer expired before you had a full team of five, the mission would start with whomever you *did* have, even if it was only one person.
- At some point, this appears to have changed for the Lemuria crisis, or possibly just broke; Monster Island still has the behavior, but I've heard comments about it having been broken for a while.
- The four Cosmic mobs in the game (Grond, Kigatilik, Teleiosaurus, and Qwyjibo) each have unique behaviors and nearly-unique powers, and require different tactics *and* strategies to beat. While you can bring as many people to the fight as you want, folks generally don't bring more than a couple of teams since the loot drops only happen for a single team.
- On Alert introduced the idea of the Rampage, a ten-person alert that is a straight-up fight against a new Cosmic (so far, Gravitar).
There is a set of factors completely separated from any issue of "balance" which, taken together, make Gravitar into punitive gameplay that goes beyond what even the early "Nintendo-hard" or quarter-sucking video games did (for the most part). If you're not sure what I mean by "punitive gameplay", Shamus Young has an excellent discussion of it at http://www.shamusyoung.com/twentysidedtale/?p=2048 (I could have sworn I *read* this at some point, rather than it being a video, but this is what I could find readily).
What makes it so punitive? The combination of mechanics around death. I don't mean how easy it is to die -- that's a balance question. I mean how it repeatedly and brutally violates the first, last, and only tenet of game design: never force the player to stop having fun. Every time you do, you make the game less fun for them (pretty much by definition); do that enough, and they will decide that some other game is more fun. Or beating their head on a brick wall. It isn't a question of "whether", but of "when". So the best way to handle it is never to start sliding down that slope.
The first and most egregious problem: the lockout mechanic existing at all. There are ways to make something have penalties for dying that don't involve preventing play. But if my reaction to dying is "oh well, guess I'll go make a sandwich", the game has just lost me for at least that period of time, and likely for a lot longer. It doesn't take many of these to make that "forever".
The next is a lack of any warning that respawning (rather than reviving) will lock you out. In fact, if it is going to lock you out, why are you even offering the option in the first place? It is an option that nobody is going to *want* until the entire team is down, at which point you'd be better off just force-respawning the team en mass. But at the very least the game should be signalling that a lockout mechanic is in play, and that respawning is effectively the same as quitting the fight.
To add insult to (virtual) injury, the Gravitar fight respawns you with direct line of sight, so that you get to see all the fun you're not having and feel like you're in a special Hall of Shame because you didn't realize it would lock you out. Now we're not only is the game being punitive directly, but socially as well.
Finally, because of the need for a full ten-person team to take on Gravitar, it often takes a long time for the queue to fill. You literally cannot just "get back up and go at it again" unless every single person in the group decides to, or there is a pending queue -- which, after the first burst of activity when it was released, I've never seen happen.
What amazes me isn't that people dislike this content -- it is that there is anyone who is actually willing to *run* it more than a handful of times in the first place. And guess who that is? Farmers. Because there is literally no other reason to *bother* doing it repeatedly. Because 'bother' is the correct word.
Now, will all of that, the title also promised some possible solutions. See the next post.
Post edited by fentonator on
0
Comments
The ability to respawn is there in case you use it toward the end of the fight, (with the assumption that you contributed during the fight), and don't mind waiting until the others can finish the fight.
Now don't get me wrong - I too would have preferred that a clear warning was given that respawning would lock you out. I would also like to see rez-other powers buffed in such a way as to make them more desirable for a person to take - like a buff to them and the one they rez, the prevention of the loss of stars, etc.
The big question is *what* could be implemented to deter so-called "zerg" tactics, where players just go in, deal some damage, die, then respawn just to repeat the cycle...
"Is it better to be feared or respected? I say, is it too much to ask for both?" -Tony Stark
Official NW_Legit_Community Forums
For what its worth I am not defending the Gravitar fight. I find it to be rather boring. I merely wished to point out that other people enjoy things that you (and I) do not. This particular piece of content was not intended to be enjoyed by the majority of CO's players.
Bosses with good self heals or some sort of mechanic where the boss gains health (or some other buff) when he defeats a player.
'Caine, miss you bud. Fly high.
Gravitar is a Cosmic. She's supposed to be a hard fight. Painfully so, if you don't come prepared. That is not, itself, a problem; it is a reasonable design goal, and properly done it is, as the saying goes, "what it says on the tin".
The first problem: "prepared" means something very different than what it does for the four other Cosmics. For them, you generally have one of two situations: either one or two teams who are organized, geared, and usually practiced are going after the mob (a directed hunt), *or* a mob is up and someone manages to herd enough cats -- er, heroes -- in that general direction that the mob is facing a tremendous amount of combined firepower, even if few of the individuals are particularly heavy hitters.
But for Gravitar, you effectively combine the limitations of the first group -- at most, two teams -- with the limitations of the second, in terms of the group being a semi-random mash of people where the only guarantee is that they have at least the minimum required level. They also generally don't have some of the other advantages, such as having everyone set up on a hands-free mic *and a common 'feel' for how conversation needs to flow to be effective*. So, fundamentally, that needs to be taken into account when considering "how hard is 'Cosmic' hard?"
Maybe it was, but my personal impression from the fight is that *if* it was, the design is still several iterations off of being reasonably balanced. Now, for all I know that's simply a matter of time; I can hope so. But really, the only reason this is an issue is that it makes all of the *rest* worse, by ensuring that a lot of folks spend a fair amount of time face-down, even if they aren't locked out.
Also: yes, there are folks with builds that can solo her. *I* can solo Kigatilik without significant problems, using my old build which is significantly less effective than my current one, which Gravitar can splatter in seconds. One person being able to solo does not *automatically* mean a PUG of ten is facing something of an appropriate difficulty.
But on to the second issue: the lockout mechanic. This is the biggest single problem, and if you take it away almost all of the rest become annoyances at worst. I can hear the cries now: "but then people will just zerg-rush her and it doesn't take any skill at all!" Well, first, I'm not convinced that's actually a *problem*, since folks can do that in every *other* alert. But purely for the sake of argument, let's say that it is: there are still multiple ways that it can be addressed. Ways that are in fact already in use elsewhere in the game.
Example 1: Kigatilik, with the impressively obnoxious Health-on-Kill. This one actually works fairly well, except for one significant drawback: the first rule of Kigatilik is NO PETS, and the second rule of Kigatilik is NO PETS. Make pet death not count, or count for the fraction-of-a-player that a pet is really worth, and it would not overly penalize pet-based builds. But even with that drawback, this general approach has a lot of promise. We'll come back to it. Failure case: Kigatilik at full health, any hounds not killed at full health.
Example 2: Grond, who you have to stay two steps ahead of in order to succeed. You can afford to have a couple of folks go down, but due to the required timing and distance to the spawn points, you *must* either have enough spare folks, or excellent coordination. Or, ideally, just not have people dying. Failure case: Grond is at full, and stays on his path as usual.
Example 3: Andrith / Vikorin, where you can re-enter the battle but the respawn is all the way back at the start of the lair, a couple of minutes away. If there is ever a point when everyone is outside the battle, it resets. Failure case: Vikorin at full, significant time on the craftable protection used up.
Example 4: Mandragalore / Bronze King: basically the same as Andrith, except that there is no timed consumable... but the mobs respawn along your path, and are sufficiently deadly that individual players who aren't strong / skilled enough may have trouble making it back, and certainly won't do so quickly. Failure case: Bronze King at full? (Never failed this one to the point of total party wipe, so not sure)
There are probably others, but this should make it fairly clear that there are a variety of approaches that are already proven to work reasonably well *specifically* for fights involving very powerful bosses and the need to prevent zerg-rush wins. And are implemented in working code, right now, today.
Next post: my suggestions for the Gravitar Rampage, specifically
Fixing the problem with it being behind a queuing system: the simple answer is "you don't fail the alert unless and until you give up". It works for almost every other mission in Champions *and* nearly every team-based mission in any other MMO I've seen. Because, quite simply, at some point either you're going to get good enough to win, or you're going to give up. Making you wait in a queue until you have more folks, just to let you turn around and try again, is purely (if probably unintentionally) punitive. It prevents learning, discourages experimentation, and drastically reduces both the desire *and* the ability of a marginally-skilled player to develop those skills.
I've been on my share of missions that were hours long to get to the final fight, and then just couldn't overcome the last boss. Frustrating, but it happens, that's life. But there are two little secrets that make all the difference in the world. First, because the players are in control of when they say it isn't worth it anymore, they are far less likely to view the content itself in a negative light unless it is just ridiculously out of balance, even if they fail utterly. And second, there are few things that feel more rewarding, or that build social bonds out of a group of random people who don't know each other stronger or more rapidly, than succeeding together against a difficult challenge. But that word 'together' matters, it doesn't work if it is nine folks watching one person carry the rest because they can't do anything but lay face down. And if you doubt the value of such things, ask one simple question: why do 'raiding guilds' even exist, and more importantly, why do they frequently persist across games and years?
As a minor sub-point of this: provide a "safe zone" where people can go when they decide "we need to regroup and change tactics, this isn't working". Don't make them leave and re-queue; among other things, it is almost guaranteed to end up with a different group of folks, which kills the entire point of "regrouping". The respawn area would actually be perfect for this if the respawn *mechanics* change, and even better, it would be the obvious and natural choice for where to do a regroup.
Visibility of the fight and the negative emotional and social reinforcement? Goes away entirely if the lockout mechanic is gone, because the only reason for someone to be standing there is either "folks are regrouping", "they are bringing up forms/pets/etc." (something that is also punitive: several common build types have severe penalties because they take several seconds each to 'fire up' their critical survival abilities), or "slacker"; the first two are generally obvious, the latter is behavior that legitimately deserves the negative social reinforcement.
Lack of warning about respawn locking you out? While it would still be good to have a more explicit notice about the *cost* of respawning, it becomes vastly less critical if that cost is something that is fundamentally fixable by the player investing effort after they respawn.
So, the last but really most important piece? The lockout. There can be a *price* to respawn without it being "stop having fun *and* screw over the rest of the group". Granted, the mechanic used with Grond wouldn't really work with the confined space of an alert. The extended distances of the two early lairs wouldn't work *as such*, but having a "penalty box" mechanic where the lockout is for a limited (and visible!) time, while still somewhat punitive, is vastly less problematic on multiple fronts. But the simplest, most direct, and probably most effective method is to adapt the existing health-on-kill mechanic that Kigatilik uses into one that is a bit more balanced.
Though honestly, even just adapting it wholesale, with all its warts, would be infinitely preferable to the existing situation. If it is easy to adopt it now but will take time to balance it better? Adopt it now and say that a better solution is in the works but will take a while.
I'll go into details on the adapted version I would like to see, and respond to a couple of posts that have shown up so far, a bit later (time to go play for a bit, right now).
The adaptation that I think would work somewhat better than reusing the current Kigatilik mechanic purely as-is: rather than "health on kill", which penalizes *dying*, and which has significant problems with making pet builds extremely counter-productive due to how easy it is to kill a pet relative to a player, make it "health on respawn". That way a rez does not trigger it -- encouraging the use of them the same way the lockout mechanic clearly intends to do -- but leaves an option open for the times when that simply isn't workable.
It means you have to work harder and be more effective to overcome the handicap of coming without sufficient rez-abilities, but it also means that there are more viable *approaches* available to overcoming the challenge. As a bonus, it also means that pure bad luck cannot hose an otherwise well-prepared team by having three or four rez-capable folks die in a span of a few seconds because of the RNG. It'll set you back a good ways in either available rez capability or damage dealt, but it removes the potential for a short run of particularly bad luck to make it *impossible* to win. This addresses one of the other "tricks" to having folks enjoy what is nearly the exact same content a great deal more, by leaving luck a major *but fundamentally recoverable* factor. It causes people to say "never give up until the fat lady sings" rather than "*censored* this, I have better things to spend my time (and money) on".
While I suspect this wouldn't work for Kigatilik (because it seems very unlikely the engine tracks *what* killed you to a level that respawning could award health to that mob), because Gravitar is in the context of a single-mob alert it actually *would* work. Because the only thing that could have killed you is her, as long as there is enough logic that the game can invoke special handling when you respawn (which it clearly can, by locking you out), it should be possible to have that same type of logic award her health instead, up to her maximum.
There are several details this doesn't address, such as "fixed amount or variable?" and "if it varies, what does it vary based on?" because those are balance issues for the fight overall and are primarily of concern for tuning the difficulty, rather than the fundamental mechanics of the fight. Personally I'd love to see it happen based on some comparative mechanic that actively discourages *careless* zerging without being quite so negative if someone has a bad moment, but even a flat "heal 1/10th of total HP for each respawn" (number selected purely arbitrarily) wouldn't be problematic to cope with, it would just change the details of the effective strategies a bit.
Now as for the part that only farmers will put themselves through the pain I don't agree with that but understand why you think that is the case. I run Gravitar, a LOT and it's not for farming or not exclusively. Sure if she didn't drop Questionite as a reward I would probably stop the same way I don't go looking for Cosmics anymore. But I do like the challenge her Alert provides. Making a build that is good enough to fight that alert without being an inconvenience to others is pretty much the only end content we have. Also being a level 35+ alert you don't have to deal with people that lack the power set or experience like the rest of the alerts do. Now don't get me wrong, I don't agree with changing the level requirements of Smashes/Bursts/Grabs, I've done a couple of alerts with my guildmates with a team full of level 6 and we've won them. So veteran players shouldn't be punished because people lack common sense and go for alerts being new to the game.
I still like to fight the cosmics but since their reward is crap that usually means soloing them, and their respawn time is unknown to me. My favorite Open Mission is Mega D in downtown. Might have crappy loot but the resources are good if you rank first. On top of that it is fun and a bit of a challenge too. It might be a lvl 24 alert (wish it had the scaling mechanics the Lemurian enemies have) but you can still faceplant during the last phase before Mega D shows up. So I do like to challenge my builds, get to know them better, see what they can and can't do or rather what I can or can't do.
That same challenge is provided by Gravitar on an alert. So I can go in queue for a couple of times and leave as I have other things to do. It's perfect. When Nemcon gets the Carrier mission treatment (as in you being able to PUG it rather than hoping you can find 4 other people willing to play it or that even know it exists) I will queue for that a lot too. It is short and the reward isn't that bad plus it has some costume unlocks. Can't say the same about Therakiel's Temple. It is pretty long and with PUGs that never have done it, it'll be a nightmare. It's too bad the recover feature is global, if it weren't there would be no need for a lock out zone in Rampages. You die you keep eating dirt until you get rezzed or you log out 'cause you don't see the team winning. Which btw you can do with Socrates without logging out so there's really no need for the lock out area if Rampages didn't have the recover button. As a side effect Gravitar has added a lot of names to my ignore list, not from something personal but from idiots that like to spam the help me button.
I don't want to put a plug here but if you're playing champions online through Steam, I uploaded my guide on how to fight Gravitar. If anyone is having trouble on figuring some stuff out please have a read, if you have some input or questions you can PM me. I think that alert can be fun once you get the learning curve out of the way. There's also the Q4Gravitar channel that has a lot of experienced players on every role.
Quite true, but it begs the question: why should you have to learn by being punished with unrecoverable failure when you use something that the entire rest of the game encourages you to treat as a momentary setback?
Sorry, I can't buy into that. Standing around doing nothing, instead of laying there face down doing nothing. What advantage or benefit does that actually give? If it were locked out until the fight were actually over, I could see it as a simple coding approach to dealing with dead folks, but otherwise?
Also, while it wasn't posted yet when you wrote this, I believe the alternative mechanic addresses this better anyway; if the fight is over and you respawn, there's no mob to heal, so you still have the option of waiting but being able to stand up when all is said and done.
I think the reason a lot of folks lack rez-other powers isn't that they aren't good enough (they may or may not be; certainly some comparable powers in a certain other game had the joke "wait, it also rezzes?" told many a time), but rather that they simply aren't actually *useful* enough outside of Cosmic or Lair fights with the lockout mechanic for most people even realize they are significant.
As for anti-zerging, the overly-simple but fundamental answer is "make respawning cost more than it benefits you if you are doing it more than once every <X> amount of time", where the notion of 'cost' and the value of X both vary not just by game, but by fight.
If you load in slowly, you can get locked inside without ever getting a chance to leave. And no, the crime computer doesn't work, so you have to either log out for several minutes or sit there uselessly until the team wins or dies. I never really liked the Gravitar alert, but that completely killed it for me.
Up until the last sentence there, I'll plead that my biases were showing. There are indeed some folks who like the fight as a fight. However, I take issue with the last sentence, and with the notion that there are many of these people, based on one simple observation: if there were enough of them that at least ten (or even close to that many) were on during the peak hours any given evening, I wouldn't see regular pleas to "Queue for Grav" on the MC zonechat.
As for the last sentence, if you have a pointer to a dev saying that, great, but otherwise it makes no sense: every other deliberately hard mission is tucked away where the average 40 won't find it unless they're at least vaguely looking for it. Sticking something out there on the main screen of *every* 35+ player, *all* the time, for content which is intended for a population which is unlikely to be above the single-digit percentages? Doesn't make a lot of sense as a good use of mind-share.
That said, it is a Cosmic fight, it shouldn't be a cakewalk even for an above-average player. The problem is that there isn't a "learning curve" to it, it is a sheer vertical *cliff*, a step function: either you're already good enough, or you cannot straightforwardly learn to be good enough without making it extremely likely everyone else in the instance will fail because you consumed important resources in learning. There is no 'middle ground' where you can have up to, say, five of the folks in the fight be 'passable but not particularly great' at it as long as the other five are exceptionally strong, and have it just take somewhat longer. As it stands now it is strictly pass/fail, both individually and for the team.
If it wasn't intended to, then it shouldn't be where the majority of them are going to see it every time they look at their screen, plain and simple. I have my doubts that was the intent, but I didn't design it so I can't say if the issue is with the content or the advertising, in that sense. But the content could be adjusted so that it still gave a very high challenge, without being actively punitive -- which makes it no easier to win, but a whole lot easier to learn... and for most humans, basic psychology says that will make it more enjoyable on the whole, even when they don't win.
For the record, I don't think what I proposed is the only approach to fixing it, either, and anyone who has alternatives is certainly welcome to fly them in the thread. It is just the one that seemed to have the best return for the least dev effort (and the least major balance changes, meaning the lowest chance of having severely broken/out-of-whack content after such a change). And that's based entirely on guesses as to difficulty from "outside the box" (unable to see the actual source code), so it could well be wrong, but the principles of what needs to change to fix the problem don't change.
2. Do away with lockouts
3. Respawn works with an ever increasing timer in rampages, to make players with resses useful still. 30s, 60s, 90s, +30s each death, cannot to be too short to invalidate resses. Loss of stars still inherent. (But still advisable to pack res devices.)
4. Due to diminishing rewards that kills the queue, have a special version of the rampage for 15mins every hour (can change the color of the icon) that grants fixed amount of questionite, and higher rewards, random green, blue, special purple gear kit or fun devices like those from nighthawk or warlord or even higher chances for a costume drop.
Whether everyone agreed to everything that another person had said, I feel that the game could only stand to benefit if such suggestions were implemented. However, it will be even better if they were already put in place before release to live and I feel that more people need to be PTS testing. It is nearly impossible to queue test Gravitar sufficiently to give feedback for the devs.
Fortunately, I could test out the newest mission that will be implemented for the mega event because it was recommended for but not compulsory for 2.
Johhota raises some valid points, though I will note that the respawn timer for the four "big" cosmics are rated in "days" rather than "hours", which is its own separate issue. Of course, when I first read "when NemCon gets the Carrier Mission treatment" my reaction was "nooooooooooooooooooo! Don't ruin another one!" Then I read the rest of his sentence.
I can completely understand the challenge of coming up with a Gravi-build, but I think that it shouldn't be a *required* thing in order to have some chance of doing the alert. After all, ATs *are* supposed to be able to succeed at it. But my issue really isn't with the difficulty; if dying isn't the end of the world, then dying quite so easily isn't nearly as problematic. And while guides are an excellent thing, and folks should take advantage of any info they can get, good game design dictates that a generous portion of the target audience be able to figure it out reasonably through trial and error within the game itself. Obviously a few folks have, but just as obviously a lot of folks gave up because it felt like a useless exercise in futility until the top 1% of folks, or whatever, figured it out.
I'm ambivalent on whether respawns should have an increasing timeout, but functionally that is the same as the "penalty box", and it absolutely does provide for a few slips while still keeping zerg-rush from being a viable tactic. I think that if I had my preference I'd argue in favor of a 'decaying' timeout: the first time is "free", but each time you die the 'base" timeout goes up by X seconds... but to counteract that, if you stay alive (or maybe "hold aggro" / "heal" / whatever other thing non-damage behavior should be encouraged) that reduces the timer back towards zero. Do it enough and your next respawn is, once again, free -- but only because you've contributed enough to the overall effort that you clearly aren't zerging. The main difference is that if you have a total party wipe, or even a nearly-everyone wipe, the "always add X" with no counter will build up to the point that while you aren't *technically* unable to finish the encounter, the cost of a mistake becomes prohibitively high and gets back into "punitive". Doing damage should not count toward this, for two reasons: first, it is its own reward in terms of progress toward the goal, and second, it opens the system to min-maxing where folks try to zerg *just* slow enough that they get their timer down, by doing enough damage before dying. It is a lot harder to 'zerg' heals or tanking.
I have thoughts on the diminishing return problem, but I haven't sorted through all of them yet. I agree it is part of the reason that the queue is not drawing more, but I also agree with the sentiment that just cycle-running that one fight with *no* variety isn't a great answer either. My hind-brain is telling me that this is actually a problem with the larger 'grouping' mechanics, beyond even alerts, but I'm not sure yet.
I would certainly hope that any such major change would go through the PTS first. Which isn't the same as believing it will, if it ever happens, but I can hope, darn it. I suspect that two things would make a huge difference in terms of PTS-testing Gravitar in particular:
The mechanic that Cryptic already has and is able to implement in very specific and seperate alerts only context:
I would also like to add, there is no way to stop zerg rushers. Be it those unwilling or due to the nature of alerts, unable to learn because they can't tell or understand their level of contribution compared to another. Then there are the Q farmers, who can zerg rush into melee range with a range build and faceplant pretending to have contributed, or respawn pretending its the lockout's fault, or simply afk at the spawn point. We can't force upon anyone the same level of participation. There are even eb users that put up a great pretence performance by consistently heal, block and move (I caught one on video incidentally), but even with all that effort, refuse to use his attack powers for the whole fight (hard to understand the mind of a leecher).
But a penalty for death is a learning mechanism itself, because it's clear cut, that you can't contribute when you can't stay alive. It's then up to the player to better himself with what's available to him.
I'm somewhat less concerned about preserving the five minute time-frame, because frankly I think it is silly if you can't use something you purchased pretty much whenever you want, but I also agree that 15 seconds is really barely enough to *notice* someone is down if the battle just went pear-shaped and the healer(s) are focused on not dying *themselves*, first, much less to get to the fallen toon and rez them. Frankly, even two minutes has issues because of the refresh speed on rezzes, but I would be more inclined to address that by lowering their cooldown rather than going past the two minute mark. I actually think that is fairly significant because it is a number that appears in several other contexts (back, once again, to the 'penalty box', for example), and thus has some cultural resonance for at least a reasonable number of players.
All of that said, while I would *like* to see a system that was slightly less of a drastic cutoff than "<5m, no help, >5m, bam everything" because step functions are almost always less than ideal, the fact that it exists and is working in the engine already means it gets the same vote from me that Kigatilik's mechanic does: "triage" it and fix the worst of the problem by using an easy, proven solution, and worry about making it fancier when you get the resources to address that.
Aside: please, please let them get the resources to get that far down the list!
Tuning the initial delay and the increment are things that are, well, tuning. If the devs write it with the assumption that it *will* need to be adjusted (in fact, not just immediately but over the long term as frameworks, expectations, play-styles, and player desires evolve over time), then making adjustments as necessary is a fairly minor matter of tuning, not a major production.
Every remotely reasonable player on every MMO I've ever personally seen was entirely willing to accept "hey, this seems to work okay on the PTS but be aware we may need to adjust it once we see how it works out in real play" as a valid reason for difficulty and timers and damage and all sorts of other things to be up in the air for several patches after the introduction of a new thing. It is only when it *stays* completely chaotic, or when it changes without warning after six months or a year, that they usually get cranky.
[ I had a "protip" here about MMO player-base psychology, but it seemed crass, so I'll leave it as an exercise to the reader ]
This is more for the benefit of people who support the excuse that the coding can be too complicated or beyond the engine's capability.
Not wanting to go into too specific details is my way of being open to suggestions. Hopefully, it's not taken the wrong way.
I am however concern when suggestions circumvent in game powers' mechanics, or rather how it would change the builds landscaping. Because builds with cooldown reductions or through gear is a fan favorite for many these days. A 5 mins CD can be almost halved with just gear or in some combination with INT and specialisation. Adjusting it without voiding ressers is more a matter of balance than how we can improve Gravitar as an alert. Having no penalty for rebirth or respawn devices simply makes dying meaningless.
But to keep on track, there are issues suggested that are more urgent before we need to look at this balance though.
I think we're mostly in vehement agreement here. Now if we can just get those who can cause it to happen to gaze upon the thread and Find It Good...