Update on Cyrodiil Performance & Upcoming AOE Tests

OtarTheMad

merevie wrote: »

Limit group size in pvp.
It runs beautifully then a ball group arrives.

They are also running their own calculation addons.

Please at least test this -we players all know THIS one thing is the major problem -we can even tell when a ball group arrives and hides before they cast.

(And Darsaga is right -aside from camps ;p )

While ball groups and faction stacks are part of the problem, a major part. Reducing group size won't do much imo. With ball groups, those guys are typically in voice chat and heavily organized so they can split into two groups of 12 and still show up at the same keep and lag things out. Maybe they will test this one day and I'll be wrong and that's fine but restricting group size does not mean players won't ride together.

C0RTEX4

Also remove that stupid buggy "In combat" feature it's to frustrating not being able to mount when you are on the other side of the map away from everyone else because you took place in a fight 30 mins ago...

I know alot of pve/pvp players are gonna cry after this last one but I don't care it is a fact that addons are breaking the game,
for heavens sake use inspiration from modders to create a better stock in-game UI system and remove all addons from the game, if console can't have addons we shouldn't either...

Selot

Are you trying to turn ESO into World of Warcraft with slow and not dynamic PvP? ESO combat system isn't perfect but we like it. It makes the game unique and different from other MMO. If those changes come to live server, many people will leave the game. We don't want cooldowns.

Agalloch

Wow, ESO becoming a turn based game! Yay!

Bashev

@ZOS_RichLambert I am curious what will be the results. I think that the performance will improve.

If this fail you can try my suggestion from 6 years ago https://forums.elderscrollsonline.com/en/discussion/114541/can-we-have-a-campaign-where-all-aoe-dps-skills-are-disabled

It looks like you are getting close to it

Agalloch

Better revert all the changes you made ...and return the PVP before the Lighting patch.

Bashev

Agalloch wrote: »

Better revert all the changes you made ...and return the PVP before the Lighting patch.

You know that lighting patch reduced the fps but did not increased the server lag. It did a little because some checks were moved from client side to server side, but this was needed for security.

Tornaad

What if instead of a simple increase in casting on AOE damage you put a system that increases the cost of anything sequential abilities that can get reset by a certain number of light and heavy attacks?

Red99

I remember 3-4 years ago where there were more pvp organised guilds spamming aoe in primetime and population was higher the lag in cyro wasnt that bad..

colossalvoids

Joy_Division wrote: »

Wow.

You are really going to completely redesign the mechanics of the game

We will be very upfront, but please be aware that if these tests confirm our hypothesis, then chain-casting AOE abilities will no longer form the core of the ESO PvP experience in the way it has for the last few years. We would then go through each class and ensure that there are viable builds for each and make adjustments as necessary.

rather than upgrade servers or change the anti-bot/cheat system or try to get to the root of the faction stack AoE-fest in the first place (put objectives all over cyrodiil to spread people, un-nerf our neutered abilities so we feel we have options besides joining groups and spamming AoEs, capping group size to 12, etc)? You're going to have to redesign Templars from the ground up since just about every one of their skills worth using (Breath of Life, Ritual, Jabs, Reflective Light, Backlash, Repentance, Spear Shards) and even those not worth using (Healing Ritual, Solar Barrage, Blazing Shield, Explosive Charge) are all AoE. Seriously? If so, then it looks like the PvEers were right all along: we PvPers will be the ones who have ruined their classes, their skills, and potentially their game.

What's the plan when people are stuck with cooldowns and get run over by ball groups who stack AoE proc sets?


I want to echo a point made earlier by "Izzy" who is the head of the Dracaris guild on PC-NA. He is 1000% correct when he said that organized groups (and I'll add good players) have always been able to chain spam AoEs. Every organized group I ran with during every patch in the game has always been able to spam AoEs on every server, every campaign, CP or no CP, PvE or PvP. All the nerfs that have come down the ZOS pipe has just made things more inconvenient for organized groups (meanwhile, the ordinary players and those who do not run in organized have had to deal with the brunt of stuff ala Morrowind update). The premise that my group uses more AoEs now than they did in 2015 or even 2014 is wrong.


I do not think ZOS has a very informed view on how organized groups - which are the biggest culprits here - play. For years, I and every guild I have been associated with has extended the invitation for a developer to play with us a couple of times to see first-hand what we deal with on a nightly basis, how we run, why we run the things that we do, and be able to base changes to the game on the how things actually are in Cyrodiil rather than incorrect assumptions (such as not being able to chain spam AoEs without CP, gear sets, or other additions to the game). You can't gleam that stuff looking at a data chart. Everyone would have been better off if ZoS had done this years ago, like 2014.

If you're going to take the very radical step of changing how the game's mechanics and classes function, then I think you need to take the step you did not take in Beta: guest in a (good) organized guild and experience first-hand how your game is played. That will go a long way to eliminating the excuse of "we didn't anticipate," which seems a prevalent theme in the OP.

Hard to not agree, especially when super easy stuff like 12p group size cap should be already in game by now and anti-cheat should be in works along with cp rework also as server calculations always would be the biggest offender here.

ExistingRug61

This.....is interesting. I am glad to see that there is some form of specific testing being done in relation to performance improvements.

I guess its not really a surprise that AOE abilities are being looked at with regards to performance, as the required "area check" for these abilities seem like they would be process intensive. However, I can't help but think that the proposed tests are a fairly blunt way of trying to deal with them that may not actually address the problem if AOE abilities are found to be a performance issue, and additionally would be a major upheaval to eso combat as we have known it.

Obviously I don't have an intimate understanding of how these things are coded, but there seem to be a couple of issues to this sort of approach of limiting the "spam" of AOE abilities.
Firstly, consider a ground based AoE DoT, say Wall of Elements. A single cast of this would require an area check to have to be performed every second when it ticks, which to me seems like it would be equivalent number of area checks to spamming a direct AoE, say Impulse, for the same duration. All of the proposed tests would limit the second case but do nothing for the first.
Secondly, it would seem to me that all smart heals must perform an similar AoE area check to find their target, even if the ability only effects one target. So would all single target smart heals also have to be affected by these measures?

My hope is that the proposed tests are only that - tests, where ZOS is using a simple brute force set of conditions to force players to use less AoEs to see if this helps performance.
And that if it does turn out that AoEs need to be addressed that they then come up a much more targeted and effective method of addressing these issues than simply implementing one of the test cases long term, which would likely have to be different solutions that vary based on the type of ability in question. Although even in this case it would likely be a very large and potentially radical set of changes, but I think that its best to wait on providing any feedback on such changes until we see what they may be.

I will optimistically assume this to be the case, and as such would encourage players to be in Cyrodiil to participate and provide data for the tests (I would if I could, but these tests won't be on console so I can't participate), and not get too hung up on the potential consequences of the suggested changes becoming permanent, but rather to wait until the results and then give that sort of feedback to ZOS when they propose what (if any) resulting changes may look like.

Edit:
Additionally, I would have say that the following makes a lot of sense.

Joy_Division wrote: »

If you're going to take the very radical step of changing how the game's mechanics and classes function, then I think you need to take the step you did not take in Beta: guest in a (good) organized guild and experience first-hand how your game is played. That will go a long way to eliminating the excuse of "we didn't anticipate," which seems a prevalent theme in the OP.

Organised group play will likely (and quickly) find a way around whatever limitations get put in place, so the above seems like sensible suggestion to at least try to capture some of that as it may not be present big data that will be collected from the tests.

Hämähäkki

I remember them deactivating the CP for some time because that was allegedly the problem. Now they try it with AOE and I bet it will not change a thing.

It'll be the same afterwards...

Shanan

LPCDC wrote: »

ADIOS, EP nooballers.

indeed. and ADIOS AD Nooballers.
I do like these changes a lot. Looking forward to it.

Zorgon_The_Revenged

It's like impulse and healing springs spamming was never a thing in 2014/2015. I guess if this doesn't work we could always just spread out, maybe go do other things on the map......

silver1surfer69

Ok first of all i hope there will be enough people around when you do this tests to get valid data.

To the problem:
As you stated above and as we all more or less know the problem is that there is too much load on the servers.
So moving on from here, we have 2 points to put the wrench on.

-> Load
-> Server

The first and imo most natural thing that comes to mind is server. The player number has grown, that means you earn more money. If you start a factory and the factory grows, you need to build a bigger factory. Im sure there is a specific term for that in Business Administration science for not investing enough in the own company.

-> So: can someone from ZOS please explain to me here or somewhere why is it not possible or what is the problem with building bigger and better servers (maybe the term money cant be avoided here any longer and if not please help me learn).

Second, if all the lean management which you are doing in improving the processes and making it leaner and more efficient is not enough, you need to restrict the number of players (here comes the term money to my mind again, but maybe im wrong). So maybe restrict number pf players per campaign or make mor campaigns with smaller max player number. I know this is a hard thing and we all will suffer from that, but if nothing else proofs possible, you have to.

Pheeenixxx

1. Who causes lag? A 12 people ball group or the 30-70 players chasing them? If those 30-70 players are to lazy to use discord/ts and to create an own group setup and to run coordinated together its not the ballgroups fail. Good and organised groups will be always better then a bigger number of unorganised people and thats the point of groupplay and an mmorpg. Besides that i would distinguish between ballgroups and a zerg (24 people +). Only a ballgroup can stop a zerg from just rolling over everything. Just play ad non cp eu and face ep. 24-70 ep players together. You think you can kill them without coordination?
2. Remove the Hammer of Lag (Volendrung)-biggest Lag creator in game. So obvious and if you dont recognize maybe run a 1 week test.
3. Midyear mayhem server performance was (really) good - so for me it is obvious that ZOS just have to pay for better servers with our money- its not a question of aoe it is the server quality which makes the difference. This aoe discussion is ridicolous and its just distracting from the potato servers.
4. Besides that improve the value and meaning of guilds
-Claim a keep, resource, town or outpost in Cyrodiil for your guild
possible benefits for the guild: small pvp buffs for guild members in this area or for a certain (short) amount of time, gold / hour, recources / hour for the guild (example: Your guild claims the lumber_ the guild gets 1 k wood/hour)--> might help against all players at one spot (LAG), reward players for pvp instead of forcing them to do pve content.

other guild ideas:
https://forums.elderscrollsonline.com/en/discussion/433784/improving-the-value-of-guilds-a-big-wish-for-the-next-eso-chapter-and-dlcs#latest

5. Create better content instead of digging in the dirt for 20-40 €.

Rhaegar75

I know it's a typical silly answer...but...I cancelled my subscription which was due to be renewed on 3rd August. I'll be on the fence and resub as soon as I'm sure that PvP is not going to get wrecked.

I'd advise more players to cancel their sub now so that we can send a true clear message

nml

- Please do not blame player behavior for performance issues.
- In a massive world with three alliances fighting for control over castles - you don't want large scale "faction-stacked" battles? Those are the best battles. Rather than trying to prevent them, Zenimax should:

1. Write more efficient code.
2. Revisit/upgrade ESO's network and server architecture to ensure it is taking advantage of modern services provided by cloud providers like AKAMAI/AWS/IBM/Microsoft to reduce latency and increase processing capacity.
3. Remain true to the vision of Cyrodiil PVP.

[snip]

-NML

[Edited to remove Bashing]

Luke_Flamesword

Who causes lag? A 12 people ball group or the 30-70 players chasing them?

If chasing players plays normally, than still 12-people ballgroup Non-ball grouped people don't have infinite sustain to cast AOEs all the time so this 12 people still can make more damage to server than rest of population. Problem is not with may fighting players at once, but with people exploiting system to cast infinite AOEs like crazy. It makes me sad, because we have really dynamic and fun fighting system, but of course there are always people who can't just have a good time, but they have to win at all cost. This is really fun to you, running like crazy chickens in group and smash instantly one button? I blame toxic ballgroups, because they know that they making lag with theirs crazy spamming, but instead of playing normally (you can still coordinate, but not like this!) they still choose to do it, because winning for them is more important. Now we all will pay for this, as always - both PvE and PvP always suffers because of elitist, exploiters and freaking winning-machines. You always must exploit every freaking mechanic and loophole just to have more DPS. Yeah, I know that ZOS should have better servers, better game code and better reactions for all this. They react now and I don't blame them that they delayed this decision for years - it's not popular and not fun, but players makes it. Similar situation was with BGS - discord premades was using they advantage to smash PUGS like crazy and that was fun for them too.

It's so egoistic, thinking "if game allows something, than it can be used and it's others problem if they don't use it". In real life you also make everything to achieve your goal without empathy and watching for other people? You think "it's only a game, it doesn't matter" but it''s not a true. It's a living environment nad mostly created by players. ZOS don't have all control - they make some systems, but players are using them. And they can used them moraly or they can exploit every loophole to push more and dominate others. if players don't want to behave healthy and play nicely, ZOS don't have a choice - they must limit us. ZOS are not saint in this, but still I blame players - especially all this constantly whining about performance and and at the same time making everything to kill it.

Cireous

I'm dying inside, dramatically so.

You will be creating an entirely different game with all of these changes. Please leave any new, potential performance-improving, game identity destroying changes in PVP, where you are having all the problems.

Please finally create a separate rule set for skills and sets when it comes to PVP and leave us PVEer's alone.

Bashev

Pheeenixxx wrote: »

1. Who causes lag? A 12 people ball group or the 30-70 players chasing them? .

The 12 ppl cause way more lag than the 30-70 chasing them.
These 12 ppl put same amount of calculations on the server as the 50 (average) who chase them. The problem is if the group is good they dont die easily and they attack more and more ppl. And the lag increase.

technohic

carlos424 wrote: »

Who is actually going to play during this time? They could give out 5x AP. Lol. Can’t this be organized on the PTS? Recruit people to play, and give some incentives, such as a free month of eso+, etc.

I'd recommend everyone who doesnt like the idea to play. Otherwise; lag will be better and the changes will look good. So come up with builds and group comps and come try to prove them wrong.

technohic

For the group size reduction idea; it would only work if heals and buffs were limited to the group so you couldnt have 4 dedicated purge and heal spammers. I would allow 24 man groups, but only in Cyrodiil would I then create "squads" inside the group to where all heals, support, and buffs are limited. I'd limit these to 8 people myself as 12 allows enough for good groups to continue their behavior that's causing issues. 8 is plenty to cover basis if. Ot too much. Youd then have squad leaders, maybe with a banner on their back (think dragonstar arena NPCs) and a different colored chevron and crown each squad. Bigger crown for the squad leader thats also acting crown.

XNativexPrideX

During midyear the lag isn't nearly as bad. If you invested into your servers with all the money players invest into you (ZoS) we could be going down a better road. Instead you'll kill ball groups. Which yes, many zergling players are happy about. But why punish players for being good? If you ever played pvp you'd know that some faction zergs would steam roll an entire map with no one but ball groups standing in their way. Example EP on PS4 NA for a few months now they have been faction zerging keeps away. 70+ EP players will show up. And thankfully coordinated groups can help slow them down, if not stop them. But now a 70+ zerg will just face roll the map. Pvp combat will slow way down to a point where a 5-10 minute fight will turn into a 30+ minute fight as zergs try to single target everyone to death.

Taraezor

Thank you ZoS for looking into this. Ball groups are ruining the experience for me and everyone else who is not a part of a ball group. Thank you for exploring options and giving us a heads up on the difficulties of finding a solution.

Rex-Umbra

If they do any of these changes they should really increase AOE damage to be par with single target to make up for them being on a 3 second cooldown or 5 second cost increase.

Delsskia

This just sounds like a recipe to make combat even clunkier and skill casting less reliable. And the server will still be doing a ton of calculations to figure out when abilities ARE allowed. You know we'll still be hitting the AOE buttons so that we can get one off the moment it's available, so the server will be flooded figuring out when that can happen. If it grays out our skill buttons after a cast, that is a VERY slippery slope. We already have to deal with that after a roll dodge or break free.

I believe that the simplest and most elegant solution is to use Battle Spirit to just disable ALL proc sets in Cyrodiil. That would effectively take about 1/2 of the calculations out of the zone without affecting PVE in any way. Let's quit screwing up actual skills and look at the mindless procs that require zero skill.

PureEnvelope35

JamieAubrey wrote: »

I'm reading this and all I get is RIP BALL ZERGS

YES YES 10000000 TIMES YES

You truly think this is gonna stop ball groups? I promise you they're gonna be even worse if these changes go live permanently...

Foto1

or maybe it is better to test the increase in server power:
week 1+ 10% power
Week 2 + 20% power...
so until the lags disappear.
we don't even need double AP

Paske

CTO: It is offical, AOE-s are killing the performance. We need stronger hardware.

CFO: Hold on, let me crunch some numbers .... Yep just as I thought. It will cost money. No can do ...

Third guy: But what if we simply limp the combat system. Like with a 3 seconds cooldown. I mean, they will hardly notice.

The whole room: GENIUS !

Blizzard hearing of this: GENIUS !

Players: Why do you hate your player base. Is the money we are giving you somehow offensive and you decide not to invest this offensive money into hardware infrastructure.