Update on Cyrodiil Performance & Upcoming AOE Tests

SpacemanSpiff1

metabLast3r wrote: »

This game has grown exponentially in the last 5 years. Easy solution, upgrade the servers to support all of this instead of pulling things back for performance. I just don't understand how hard this could be with all the money being pulled in from the crown store.

winner winner chicken dinner.

instead of reinvesting in their cash cow, they'll just reduce functionality. sad, but i'm not surprised given their track record.

Hiker

I don't quite understand. The tests will obviously work as the amount of people playing and the amount of skills being cast will be dramatically reduced. Because of the tests themselves...

frozenwattersb16_ESO

VALID ARGUMENT
Test rewards are not enough, not enough players will test these changes and will make the tests invalid.

VIABLE LONG TERM SOLUTION (CYRODIL LAG)
Create loading zones (like in Imperial City) so that load can be distributed. This would also allow the developers to bring resources to bear on zones that are experiencing more load than others. CCP did this very successfully with Eve Online.

Cheers,

TequilaFire

Most still don't even know what a ball group is but they have become the whipping boys because they heard it on the forums from people that don't even play PvP.
I got news a ball group is not a zerg.

relentless_turnip

@ZOS_RichLambert
I do really appreciate the effort to address these performance issues.

It has been very easy to assume that PvP just wasn't a concern for a long time, however seeing the lengths you guys would go to fix these issues is reassuring in a way.

I believe there are better ways to reduce the calculations that don't change the core combat of ESO and address the situations where people spam AOE's the most.

@FENGRUSH presented some good idea's and presented them very constructively.

-Reducing group sizes to 12(reduces radius check)
-Healing to only effect group members(reduces radius check)
-Only one type of each heal on a group member i.e. recasting rapids would refresh the effect rather than stacking it.

The last point also separates healers from DD's and makes healers more important and sought after in a group dynamic. Having multiple forms of healing would be essential to acquiring good healing for your group. Rather than half your group slotting a resto backbar.

@FrankonPC also suggested removing certain proc sets from PVP, those that require a radius check for example. This would also make it easier to balance them for PVE, thus making them a good substitute for low APM.

I would also suggest:

-Making all sources of damage mitigation, buffs and debuffs a flat number. This will make calculations simpler and simpler to understand.
-I also think the ability to toggle our healing priority so they only effect ourselves would also help accidental cross healing and a lot of people would selfishly switch this on
-Looking into classes and skill lines and auditing which skills could be changed to a single target ability with little to no impact would help.

I think any or all of these suggestions would have a similar/better effect on performance, whilst not changing combat. It would impact grouping, but also make it more meaningful. It would encourage a dynamic to group forming, whilst making roles more relevant.

I am all for the tests as after hearing other perspectives understand that they are in the interest of gathering data to solve the issue, rather than being the solution.
I would also like to mention @Unified_Gaming made a good point that the incentive should be better, to ensure enough people participate. A crown crate for every 200k AP earned would be fair, whilst ensuring people actually participate in PvP, as for most classes it will be a pretty rough experience.

flubber77

Remove Purge or make it purge only you. Problem solved....

Xandreia_

While a lot of us appreciate that devs are at least trying to make performance better, these proposed changes have got to be worse than when you wanted to change light attacks... all of these changes would see templar, wardens and DKS completely useless, unless you are going to rework every single class these changes would undoubtedly cause a lot of people to leave the game completely. zos seem to just want to put band aids on everything and propose silly changes that 1, probably wont do a great deal to improve performance and 2, cause more stress on the servers causing lag to be just as it was before or even worse, instead of looking into ways to make the servers stronger so they can cope with all the incoming traffic, it seems like wild ideas are just being tossed around without much thought.

ultimately if these changes do go ahead, not only would it make pvp ridiculously slow, but will kill off ball groups ( a lot of people like to watch ball group game play on YouTube and twitch so from a business point, doing this will lower the amount of new players wishing to play like this) it will also take away what most people find fun about the game, pvp is the 1 thing a lot of people stay playing ESO for.

I cant see many people participating in these trials so it will be completely useless as the information gathered will be invalid.

Xandreia_

TequilaFire wrote: »

Most still don't even know what a ball group is but they have become the whipping boys because they heard it on the forums from people that don't even play PvP.
I got news a ball group is not a zerg.

THIS! ball groups and zergs are completely different!

Karmanorway

erio wrote: »

Hey streak is an aoe right? give it double ramp please

HEY!! Stamsorc is weak as it is, even compared to templar....

MipMip

NirnStorm wrote: »

IMO this test is gonna work out great, will definitely fix performance issues.
On another note, if we disabled combat altogether in PvP, that will also fix performance issues.

This test is basically "lets see if making our dynamic combat system less dynamic, will fix performance".
Of course it will. But at what cost?


Edit: I created a new video discussing this post and my personal take on it.
https://www.youtube.com/watch?v=OD67tq4rDNQ
P.S. thanks for 100 agrees!

@NirnStorm correction you have meanwhile 125 agrees - well deserved!

TequilaFire

Karmanorway wrote: »

erio wrote: »

Hey streak is an aoe right? give it double ramp please

HEY!! Stamsorc is weak as it is, even compared to templar....

Yep cast hurricane and wait 3 second to spin to win while the stamcro wastes you.

StShoot

Temeraire507 wrote: »

So based on the PTS Patch Notes v6.1.2 and these lines stated there:

These tests will ultimately make it more difficult for AOE abilities to be the only abilities used, and we’ll be adjusting cooldown, cost and regen values of all AOEs (damage and healing).

Abilities which meet the following criteria have been flagged for being subject to this test:
The ability’s initial cast requires a radius to function. This includes abilities like Breath of Life that require a radius check to fire its heal at the appropriate target.
If an ability is a single target ability that has an area component and has no cast time.


I have made a little list of abilities targeted by the tests and planned changes based on that information, of course we will only know for sure when the tests are held but the notes were relatively clear in their wording. I thought this could still add something to the ongoing discussions. This took a little time so my apologies if someone else was faster and this is not necessary anymore.

Class

Dragonknight

Ardent Flame

Dragonknight Standard and Morph
Fiery Breath and Morphs
Inferno and Morphs

Draconic Power

Dragon Leap and Morphs
Volatile Armor Morph
Dark Talons and Morphs
Inhale and Morphs

Earthen Heart

Magma Arnor and Morphs
Stonefist and Morphs
Molten Weapons and Morphs
Obsidian Shield and Morphs
Ash Cloud and Morphs

Necromancer

Bone Tyrant

Pummeling Goliath Morph and Ravenous Goliath Morph
Death Scythe and Morphs
Bitter Harvest and Morphs
Bone Totem and Morphs
Grave Grasp and Morphs

Grave Lord

Frozen Colossus and Morphs
Riccochet Skull Morph
Blastbones and Morphs
Boneyard and Morphs
Skeletal Mage and Morphs - requires a scan if you don't direct the target to attack but is a pet so I'm not sure
Shocking Syphon and Morphs

Living Death

Reanimate and Morphs
Render Flesh and Morphs
Life amid Death and Morphs
Spirit Mender and Morphs - requires a scan for sure but is a pet so I'm not sure
Restoring Tether and Morphs

Nightblade

Assassination

Lotus Fan Morph

Shadow

Consuming Darkness and Morphs
Path of Darkness and Morphs
Aspect of Terror and Morphs
Summon Shade and Morphs - requires a scan if you don't direct the target to attack but is a pet so I'm not sure

Siphoning

Soul Shred and Morphs
Strife and Funnel Health Morph
Malevolent Offering and Morphs
Drain Power and Morphs

Sorcerer

Daedric Summoning

Summon Storm Atronach and Morphs
Summon Unstable Familiar and Morphs - maybe, it is a semi-permanent Pet but obviously requires scan and the Ability (save for Clannfear) is definitely an AoE effect not sure if this counts but it might count if they really go all out on these criteria
Daedric Curse and Morphs
Summon Winged Twilight and Morphs - maybe, same as Summon Unstable Familiar safe for its activateable Ability

Dark Magic

Negate Magic and Morphs
Encase and Morphs
Daedric Mines and Morphs

Storm Calling

Overload and Morphs
Mages Fury and Morphs
Lightning Form and Morphs
Lightning Splash and Morphs
Power Surge Morph
Bolt Escape and Morphs - Ball of Lightning is debateable, the ball it places might count as secondary area component though

Templar

Aedric Spear

Radial Sweep and Morphs
Puncturing Strikes and Morphs
Explosive Charge Morph
Spear Shards and Morphs
Sun Shield and Morphs

Dawn's Wrath

Nova and Morphs
Reflective Light Morph
Solar Barrage Morph
Purifying Light Morph
Unstable Core Morph

Restoring Light

Rite of Passage and Morphs
Rushed Ceremony and Morphs
Healing Ritual and Morphs
Restoring Aura and Morphs
Cleansing Ritual and Morphs
Rune Focus and Morphs

Warden

Animal Companions

Feral Guardian and Morphs - maybe, this is a semi-permanent pet but it probably needs to scan an area for its targeting if you don't command it
Scorch and Morphs
Growing Swarm Morph

Green Balance

Secluded Grove and Morphs
Fungal Growth and Morphs
Healing Seed and Morphs
Living Vines and Morphs
Lotus Flower and Morphs

Winter's Embrace

Sleet Storm and Morphs
Frost Cloak and Morphs
Impaling Shards and Morphs
Arctic Wind and Morphs
Frozen Gate and Morphs

Weapon

Two Handed

Stampede Morph
Cleave and Morphs
Reverse Slice Morph

One Hand and Shield

Deep Slash Morph

Dual Wield

Lacerate and Morphs
Whirlwind and Morphs
Blade Cloak
Shrouded Daggers Morph

Bow

Volley and Morphs
Arrow Spray and Morphs

Destruction Staff

Elemental Storm and Morphs
Force pulse Morph
Wall of Elements and Morphs
Shock Clench Morph
Impulse and Morphs

Restoration Staff

Panacea and Morphs
Grand Healing and Morphs
Regeneration and Morphs
Blessing of Protection and Morphs
Steadfast Ward and Morphs

World

Soul Magic

Shatter Soul Morph
Soul Splitting Trap Morph

Werewolf

Pack Leader Morph and Werwolf Berserker Morph - maybe, since the Direwolfs require area scan and the berserker heavy attacks are technically an area component
Brutal Pounce Morph
Roar and Morphs
Infectious Claws and Morphs

Vampire

Swarming Scion Morph
Mesmerize and Morphs
Blood Mist Morph

Guild

Fighters Guild

Dawnbreaker and Morphs
Silver Shards Morph
Circle of Protection and Morphs
Expert Hunter and Morphs
Trap beast and Morphs

Mages Guild

Meteor and Morphs
Magelight and Morphs
Fire Rune and Morphs

Psijic Order

Time Stop and Morphs

Undaunted

Blood Altar and Morphs
Trapping Webs and Morphs
Necrotic Orb and Morphs

Alliance War

Assault

War Horn and Morphs
Vigor and Echoing Vigor Morph
Rapid Maneuver and Morphs
Caltrops and Morphs
Magicka Detonation and Morphs

Support

Barrier and Morphs
Siege Shield and Morphs
Purge and Morphs
Revealing Flare

Volendrung - probably also a target as it would be odd if it bypassed the rules for AoE's

Ruinous Cyclone
Rourken's Rebuke
Pariah's Resolve

As you can see this does hit everyone not only or mostly templars as many seem to believe probably based on Punncturing Sweep being used as an example in the original post. We will have to see the implications of this during and after testing even if I have to say these changes for all these abilities sound pretty harsh for some playstyles and especially with redesigns it will be hard to keep this only inside cyrodiil as intended.

you should make a own entry for it so it doesnt get overseen in this thread

Nagastani

Foto1 wrote: »

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

Huh

So what you’re saying is, we just take the servers and upgrade them. It’s like here’s some money (from our Crown store loot) and there are the server upgrades and we just upgrade the servers? That’s... never been tried before, it’s experimental.

Nagastani

TequilaFire wrote: »

Karmanorway wrote: »

erio wrote: »

Hey streak is an aoe right? give it double ramp please

HEY!! Stamsorc is weak as it is, even compared to templar....

Yep cast hurricane and wait 3 second to spin to win while the stamcro wastes you.

Or the NB ganks you.

Czekoludek

@ZOS_RichLambert Please, someone from ZOS should watch @FENGRUSH response. Great vid with so many valid points and propositions that will definietely hurt game less then what we see in OP.

https://www.youtube.com/watch?v=nO-NLBjSO1U

I thought about these proposed tests and think that Zeni take the wrong approach here, looking at wrong direction.

Nagastani

Xandreia_ wrote: »

While a lot of us appreciate that devs are at least trying to make performance better, these proposed changes have got to be worse than when you wanted to change light attacks... all of these changes would see templar, wardens and DKS completely useless, unless you are going to rework every single class these changes would undoubtedly cause a lot of people to leave the game completely. zos seem to just want to put band aids on everything and propose silly changes that 1, probably wont do a great deal to improve performance and 2, cause more stress on the servers causing lag to be just as it was before or even worse, instead of looking into ways to make the servers stronger so they can cope with all the incoming traffic, it seems like wild ideas are just being tossed around without much thought.

ultimately if these changes do go ahead, not only would it make pvp ridiculously slow, but will kill off ball groups ( a lot of people like to watch ball group game play on YouTube and twitch so from a business point, doing this will lower the amount of new players wishing to play like this) it will also take away what most people find fun about the game, pvp is the 1 thing a lot of people stay playing ESO for.

I cant see many people participating in these trials so it will be completely useless as the information gathered will be invalid.

Yeahhh

So why wasn’t this done during mid year mayhem event(s)?

Sanguinor2

Hey @ZOS_RichLambert
Can you (or any other Zos employee) please publish the percentages of different classes played during the tests? Really curious about how the numbers will look like.

Sheuib

The more I think about it the more I like the idea of less AOE's in cyrodiil. Fighting will be a lot more fun if it is mostly single target abilities. It will take a lot more coordination for groups to play effectively. Zerg on Zerg fights will become massive individual fights instead of AOE free for alls.

I think being forced to use more single target abilities will make the game more dynamic and fast passed. People will be forced to get skilled at the game vs just spamming one ability while following the crown and eating Doritos. Everyone will have to learn how to fight an individual fight.

However, I do not think that healing should be limited to groups. I should be able to help a random person on the battlefield by healing them. Otherwise having factions is pointless.

Icaruzs

DCZergNoob wrote: »

Rezdayn wrote: »

Please dont put this into pve....

PVE has it paradise compared to PVPers, dont even try and pull that card

of course, every damn patch zos balances the game around pvp, nerfing pve players in sequence and you say we have a paradise.... sure

Look at mag sorc that is all balanced around not being OP in pvp, that *** most of the class personality in pve

Tigor

With all the upcoming changes, I'm looking forward to more open field battles due to shorter sieges.

In general, I expect taking keeps/ fortresses / castles to be more fun and less stessful, without dominating AOE groups.

I think everyone can then play more as he/she wants, and use the gear he/she likes, without being handicaped by lag and that is also a good thing. So more diverse and less boring.

Increase of quality....

Nagastani

Tigor wrote: »

With all the upcoming changes, I'm looking forward to more open field battles due to shorter sieges.

In general, I expect taking keeps/ fortresses / castles to be more fun and less stessful, without dominating AOE groups.

I think everyone can then play more as he/she wants, and use the gear he/she likes, without being handicaped by lag and that is also a good thing. So more diverse and less boring.

Increase of quality....

We could try it however judging by some of the replies on here it looks like this could get ugly. For my part I am spending less and less time in Cyrodiil due to ball groups and problems with move speed. In fact I did not even try midyear’s events this year due to how horrifically bad the lag and server drops was from last year.

The Uninvited

What everyone should realize, is that what this actually boils down to, is in fact creating artificial lag (3 second cooldown) in order to reduce lag.

Elsonso

silver1surfer69 wrote: »

-> 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).

I am not from ZOS, and would love to have ZOS answer this. Then, I would not have to guess. They won't, so I continue to guess.

There is a limit to computational power for a given platform, in this case very likely to be Intel Xeon based servers running Linux. It is possible for player behavior to require server computational power that simply moving up a generation of CPU cannot address.

I do not think it comes down to money to upgrade for hardware Cyrodiil, at least not within the platform that they are on, as much as people might think. At least, not within the platform and architecture that they have chosen. Each megaserver is huge, consisting of many individual servers (per pictures). Compared to the whole megaserver, each Cyrodiil campaign is pretty small. The term "drop in a bucket" may apply here.

nml wrote: »

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.

For the problem at hand, that being constant use of spammable AoE skills, this is not the answer. I like cloud as much as the next geek, but no, cloud is not the answer, here.

relentless_turnip wrote: »

-Reducing group sizes to 12(reduces radius check)
-Healing to only effect group members(reduces radius check)
-Only one type of each heal on a group member i.e. recasting rapids would refresh the effect rather than stacking it.

I dunno. Honestly, this sounds like the same band-aid solutions that got us to this point in the first place. Ultimately, this sort of tweaking is the same thing that got us here. I think that this will have us back here talking about performance before too long.

relentless_turnip

Sanguinor2 wrote: »

Hey @ZOS_RichLambert
Can you (or any other Zos empl

Elsonso wrote: »

silver1surfer69 wrote: »

-> 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).

I am not from ZOS, and would love to have ZOS answer this. Then, I would not have to guess. They won't, so I continue to guess.

There is a limit to computational power for a given platform, in this case very likely to be Intel Xeon based servers running Linux. It is possible for player behavior to require server computational power that simply moving up a generation of CPU cannot address.

I do not think it comes down to money to upgrade for hardware Cyrodiil, at least not within the platform that they are on, as much as people might think. At least, not within the platform and architecture that they have chosen. Each megaserver is huge, consisting of many individual servers (per pictures). Compared to the whole megaserver, each Cyrodiil campaign is pretty small. The term "drop in a bucket" may apply here.

nml wrote: »

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.

For the problem at hand, that being constant use of spammable AoE skills, this is not the answer. I like cloud as much as the next geek, but no, cloud is not the answer, here.

relentless_turnip wrote: »

-Reducing group sizes to 12(reduces radius check)
-Healing to only effect group members(reduces radius check)
-Only one type of each heal on a group member i.e. recasting rapids would refresh the effect rather than stacking it.

I dunno. Honestly, this sounds like the same band-aid solutions that got us to this point in the first place. Ultimately, this sort of tweaking is the same thing that got us here. I think that this will have us back here talking about performance before too long.

Do you feel it is anymore of a band aid fix than putting a cooldown on 60% of skills in the game?

I'm not sure, but I know which makes the gameplay feel worse.

silver1surfer69

When reading some of the posts i want to add something:

I dont think zos cares enough about pvpers to do this effort justfor the pvp community. Pvp is surely the minority, but pvp is like the testing ground for the whole system of combat, abilites and everything. In pvp unbalanced and not working intentional stuff comes to daylight faster and clearer. Only if the whole thing works in pvp u can know that you have a coherent and working system in itself. By recognizing this you can also see that pvp and pve have a relationand are tied together and depend on one another.

thegreatme

Alienoutlaw wrote: »

the main issue for me at least is groups exploiting the lag and building to specifically produce that lag

Yeah see and this is the main issue and honestly why I can't take a lot of the complaints from people about AOEs seriously.

PvPers are the core group of ESO that KNOWS AOE spam exploits vulnerabilities in server performance and uses it maliciously against other players, both inside PVP and outside of it.

Are you really that great of a pvper if you have to FORCE the game to lag and hit load screens so only those who can afford super good computers and super good internet can win by proxy of "Oh look, the other side hit a load wall?" And these aren't just "small ball groups" I've seen do this. These are massive zergs. DC seems to be especially guilty of this in my experience.

Anyone else remember when ZOS canceled Midyear Mayhem one time because of broken Queue problems for Battlegrounds, replaced it with a PvE event (I think it was Thieves and DB event?), and a bunch of outraged PVPers went "Well since you canceled our ONLY PVP EVENT, we're going to protest by going to PvE zones and spamming AOEs to make the PvE event people suffer and cause mass lag and mass crashing on purpose"? Because I sure remember it.

Not to mention if you involve yourself in the public town RP circles, then you know about the problem of PVP trolls showing up in groups spamming AOE light shows causing mass crashes for all the nearby players. Or they follow you around harassing you for PVP duels, spamming every AOE they have until you either agree to duel them or crash because of their AOE spam. And the PVPers who do this know they are causing mass-crashes by doing so.

While I do agree that server performance on the back end hardware would be a great step, we as the players don't actually know what they have or haven't done on the back end to remedy these issues.

We don't know for sure that they haven't upgraded or maintenanced their hardware, we don't know the actual skill level of the people they hired on to re-code the game for performance, we're only GUESSING at incompetence and patting ourselves on the back for "knowing better" when we lack that knowledge into what actually goes on at ZOS and what steps they've taken to remedy the problems of the game (I'm not saying they have taken those steps, I'm just saying we don't have that knowledge and if they have taken those steps and it hasn't fixed the issues, it just makes everyone who pulls that card look really stupid lol)

You know what we do know for certain? That AOE spam causes server strain. That PVPers know it causes server strain and exploits it. So personally you'll see no complaints from me if AOE spam in PVP goes away.

While we're at it I wouldn't mind AOE be disabled outside of combat/having a target in general so people can stop randomly spamming it everywhere they go just to disrupt other people.

silver1surfer69

I mean lets bring the facts to the table:

Its a cashcow and the cow is ridden until it dies. Feeding and investing into the cow is considered a waste.

We as players fill an environment created for us with life and we bring it to life, but this environmet is not seen (not by the creators but by the guys who decide and have the ressources in the back) as someting beautifull and healthy and preserve worthy.

SpiderKnight

Czekoludek wrote: »

@ZOS_RichLambert Please, someone from ZOS should watch @FENGRUSH response. Great vid with so many valid points and propositions that will definietely hurt game less then what we see in OP.

https://www.youtube.com/watch?v=nO-NLBjSO1U

I thought about these proposed tests and think that Zeni take the wrong approach here, looking at wrong direction.

Does he spend the entire 1 hour 40 minutes talking about it or is there a certain point that's more important? I stopped watching his vids because they are ridiculously too long. (6 hour patch note videos)

npuk

I think a lot of the lag etc is due to packet loss and high latency.

For example, 2am this morning BST, on XBox EU there were a total of approx 14 DC and 30 AD, zero bars showing on all alliances, the mini load screens were still triggering, even withing southern high rock. In a 2v2 it took 3-5 button presses for leap to go off.

Any skill that requires a 'lock on' say heavy attack with a lightning staff or toppling charge wont go off without a light attack first, the only ability that consistently land are AOE, if this problem was fixed there would be less AOE's used.

One thing I find very strange is when something is changed to see if it fixes an issue when it doesn't the change is left in place, this is a very odd debugging process as you will never know if the subsequent change would have been effective if it wasn't for a previous change being left in.

Casandra

'The code for these changes will be going into today’s PTS patch, and we will be running some basic tests throughout the day on Tuesday in Cyrodiil to ensure we’re able to make the above changes without requiring any maintenance or downtime.'

is Tuesday today. (7/28/2020).. or Tuesday before the patch update???