Why the Mapserver Event Needs to Change

[mechahamham]

The Mapserver event is... not for me. I don't begrudge the people who do enjoy it. However, it's problematic outside of its content. It needs to be reworked in a few different ways.

 

Lag

 

Sitting around waiting for your powers to recharge is no fun. Sitting around waiting for your powers to recharge when the entire zone has slowed to a crawl is pretty miserable. Your powers LOOK like they should have recharged, but haven't really. The real mapserver simply can't keep up with a full league, everything else happening in the zone, and people constantly trying to get into the possibly full zone and/or trying to float a second league on top of the first one.

 

This isn't just a problem with the Mapserver event, but also plagues the Halloween event, especially in Peregrine Island and Kallisti Wharf. It's less noticeable for Mothership raids and Hamidon raids simply because those zones hard-cap with a fairly small populations. Even then, it's still VERY noticeable during those events. Regardless, it's effectively impossible to have more than one (full) league active for those events at a time.

 

Worse, the lag for the Mapserver event seems to spill over to the entire rest of the server, even after the patch early in the week of the 1st. My taskforce teams noticeably bogged down and glitched often.

 

Even if you're not participating in the event, heaven help you if you happen to have a mission in the same zone, especially if that zone hasn't managed to spawn a second instance.

 

How can we make this less problematic?

 

First and foremost, the single biggest problem here is the event causing lag and other issues for those not participating. This is going to require at least some profiling and debugging effort to rectify. Are the events causing resource usage to top out on the servers? How much information does the event send to clients when compared to other invasion events? If the event can be optimized, it should.

 

If the perception is, 'well, we don't need to optimize it because it's a very limited event we only run once a year,' then that ties into the next point.

 

To make it kinder to those not participating, I'd also suggest that the event always takes place in a second instance of the zones it affects. For example, if Talos Island gets an invasion, the invasion should spin up a Talos Island 2 instance, even if Talos 1 is nowhere near to full. Those interested in the invasion can zone-hop. If Talos 2 hasn't emptied and despawned by the next time the invasion happens, it should default to that instance rather than the first instance.

 

"The Mapserver of Peregrine Island has gained sentience and gone rogue. (Peregrine Island 2)"

 

Loading Screen Tip #587: "If you need to travel between instances of the same City Zone, you can do so at the Tram, Ferry, Portal, or other zone entrances."

 

Honestly, I think Excelsior's population is at the point where we probably NEED two instances of our most populous zones up at all times. Blue-side, this would be Atlas, Steel, Talos, Peregrine, Pocket D, and RWZ. The Hive and The Abyss are weird cases since there's nothing that happens there BESIDES the Hamidon event (and fighting GMs to harvest EoEs and start the event).

 

A more controversial approach to this is to have the event prefer zones on the less populated servers rather than Excelsior.

 

Length

 

Last night, at the time of this writing, I was participating in a Rikti Invasion event in Peregrine Island. I heard the following from one of the other players: "Man, I've been spoiled by the MS invasions. This experience feels like nothing."

 

Ah, I thought to myself. The event lasted long enough that the experience gain from it started to feel 'normal'. It wasn't a special treat any more after more than a week.

 

In fact, there are players all over the server in withdrawal. When players who've been on Homecoming for a while related that the event usually only runs once a year, there were complaints.

 

My feeling is that, to maintain a sense of 'specialness' to the event, it should probably last, at most, a single weekend. And if it does become more common, rebalance and optimization becomes VASTLY more necessary.

 

For those uninterested in the event, the length of it was still a problem. It very noticeably drew players away from any other content. Even other invasions were ignored entirely for the Mapserver events. Worst of all, the lag created by the event for the rest of the server was more and more obnoxious as the week wore on.

 

Balance

 

I say 'Balance' here, and I think most people would immediately leap to the conclusion that I'm complaining about the experience gain. Instead, I'd like to focus on other issues.

 

Invasion events in general have some pretty serious mechanical flaws. When enemies appear in any invasion, they often have a brief period of time in which they're untargettable or invulnerable, except by pets for some reason. In fact, if you participate in any invasion event, you often hear 'target through a pet'.

 

This needs to be fixed one way or another. Either enemies should be invulnerable to ANY attack until their 'spawn in' completes, or they should be vulnerable from the moment they're visible. Currently, this creates a bias in events for players who have a pet of some kind.

 

(Additionally, in some situations, you often see 'desirable' enemies spawn in and then immediately disappear such as Rikti Heavy Assault Suits during this same period. That's annoying, but it's not as biasing.)

 

Chaining onto the odd 'spawn in' period, players with damaging AOEs can often reduce most of the enemies into a fine paste. This creates a bias in events for players who have strong AOE attacks.

 

After the 'spawn in' period, buffs and debuffs become much more important to keep those participating alive and doing damage to the enemies that remain. This creates a bias in events towards strong AOE buff, and to a somewhat lesser degree, debuff effects.

 

For the most part, these biases cancel out, even if you exclude the whole 'target through a pet' thing. Most characters are capable of doing at least SOMETHING to aid everyone else participating in the event. Characters that focus on single-target damage, however, tend to become a bit more excluded.

 

There is, however, a class of character that gets to 'triple dip'. Incarnate-level players have Lore pets, Judgement blasts, and Destiny buffs.

Not being able to contribute is worse than being unfun. It's actively demoralizing. You don't even have the fallback of 'I'll learn the event this time and contribute next time.' There's not much TO learn. It's 'fire all your abilities off as soon as they're recharged' and maybe 'focus fire on the giant monster when it appears'.

 

Nothing makes me want to avoid content more than not being able to contribute.

 

This really needs to change for ALL invasion events, and not just the Mapserver event.

 

The usual counter to arguments like this is something like 'why should higher-level players be punished for being higher-level?', and that's correct. Those players should not be punished. But just like those players should not be punished, neither should lower-level players be punished simply for being lower level.

 

One of the guiding principles of CoH has historically been that both high- and low-level players should be able to contribute to the same content. One player's enjoyment of the game should never come at the expense of another's. The recent scaling changes we've seen for giant monsters should be an example of what we need to do going forward for invasion events. In general, scaling and level compression should be used whenever possible to ensure that as many players can contribute to the event's success as want to.

 

In addition, new event mechanics that incentivize separation of duties should be considered. These could be enemies who are more vulnerable to different forms of attacks, just like Hamidon's organelles. Imagine having enemies appear in invasions that incentivize being controlled, or attacked with single target effects. These not only allow more opportunity for different players to contribute, but also add more to learn about the event structure.

 

Imagine hearing a callout like "Lowbies, please focus on taking out the Rikti Portal Technicians! Dominators and controllers, please lock down the Monkey tamers!" That's the kind of thing that would excite me about an event.

[Rudra]

9 minutes ago, mechahamham said:

A more controversial approach to this is to have the event prefer zones on the less populated servers rather than Excelsior.

Each server gets its own events. And if you limit events to less populated servers, then players on more populated servers won't see the events to partake in. So this does not work on multiple counts.

 

10 minutes ago, mechahamham said:

First and foremost, the single biggest problem here is the event causing lag and other issues for those not participating. This is going to require at least some profiling and debugging effort to rectify. Are the events causing resource usage to top out on the servers? How much information does the event send to clients when compared to other invasion events? If the event can be optimized, it should.

What causes the lag isn't the event itself. What causes the lag is the number of players participating in it, the number of mobs spawned for it, and all the powers and attached power effects bogging down the server's capabilities.

 

12 minutes ago, mechahamham said:

Last night, at the time of this writing, I was participating in a Rikti Invasion event in Peregrine Island. I heard the following from one of the other players: "Man, I've been spoiled by the MS invasions. This experience feels like nothing."

 

Ah, I thought to myself. The event lasted long enough that the experience gain from it started to feel 'normal'. It wasn't a special treat any more after more than a week.

 

In fact, there are players all over the server in withdrawal. When players who've been on Homecoming for a while related that the event usually only runs once a year, there were complaints.

 

My feeling is that, to maintain a sense of 'specialness' to the event, it should probably last, at most, a single weekend. And if it does become more common, rebalance and optimization becomes VASTLY more necessary.

Many players have limited time to play. Some have odd schedules. Their available time to play isn't going to line up with others. If the event is limited to a single weekend, it will be over before many players even have a chance to find out it ever even started. To a lesser extent, this applies to limiting the event to a single week as well. For the Halloween event for instance, I have friends that aren't able to get on the game for the entire first week of the event.

 

15 minutes ago, mechahamham said:

Invasion events in general have some pretty serious mechanical flaws. When enemies appear in any invasion, they often have a brief period of time in which they're untargettable or invulnerable, except by pets for some reason. In fact, if you participate in any invasion event, you often hear 'target through a pet'.

 

This needs to be fixed one way or another. Either enemies should be invulnerable to ANY attack until their 'spawn in' completes, or they should be vulnerable from the moment they're visible. Currently, this creates a bias in events for players who have a pet of some kind.

This is not an invasion problem. All spawns get a set amount of time to animate arriving before being able to be attacked. You will see this most often with Freakshow when they rez. As they animate getting up, they can't be targeted by players. Non-player entities however do not have that limitation imposed upon on them. So pets can attack even rezzing Freakshow and kill them before they finish animating rezzing and being able to fight back.

 

18 minutes ago, mechahamham said:

For the most part, these biases cancel out, even if you exclude the whole 'target through a pet' thing. Most characters are capable of doing at least SOMETHING to aid everyone else participating in the event. Characters that focus on single-target damage, however, tend to become a bit more excluded.

Move slightly away from the mob of players. The mobs on the fringes of the fight are more stable for targeting by players.

 

[Saiyajinzoningen]

I play daily on the most popular shard (excelsior)

I have not encountered any of the mentioned issues

in fact the chaotic copious amount of players gives me the "live vibe" 

 

Check with your isp for throttling or if you can upgrade

what are the specs on your rig? ram is a pretty cheap way to alleviate lag/stuttering/etc...

are you on wifi? if so switch to cat5 or cat6 for a huge performance increase

try a less populated shard

you can adjust one of the chats to inform you when and where events are happening in real time

lastly turn down some of the graphics settings.

 

[CrusaderDroid]

/noparticles 1

 

Or /noparticles 0, I forget which one.

 

Either way, by turning off everyone else's stuff, it becomes much more graphically stable.

[Duckbutler]

35 minutes ago, CrusaderDroid said:

/noparticles 1

 

Or /noparticles 0, I forget which one.

 

Either way, by turning off everyone else's stuff, it becomes much more graphically stable.

 

This was helpful, and unfortunately I didn't discover this until the last day of the event. It should be included as a graphics option in the menus if it isn't already. I can say that I participated in and took advantage of the event. I cannot say that I enjoyed the event. After a few rounds it became "put a PBAoE on autocast and watch TV" and even then, the autocast didn't work great.

[JasperStone]

There are a ton of other slash commands on the wiki.

https://homecoming.wiki/wiki/List_of_Slash_Commands

[Haijinx]

I /jranger all suggestions that won't fit in a college blue book on general principle.

 

"Please reduce the frequency of high lag events"

 

That might be reasonable.

[Calculator]

The two biggest boosts to performance for me were '/noparticles 1' and reducing my settings to their lowest settings (slider full left)..but, that did not reduce the message traffic lag nor the timer issues with powers..and let's not even talk about targeting...

 

I even disabled camera shake.

 

I can tell from how the battle ends that the lag is not from my computer..or, at least the hardware. (I have a ROG Strix laptop and a steady 385Mbps download stream.)

 

It usually takes a minute or two for all of the 'battle chatter' to die off and people to start leaving.

 

How often do devs reply to these threads? I'd be curious to know how their packet/buffer system works in the client and on the server~