AoE Caps Discussion

ToRelax · February 2016

Then why would we let the topic die?

VincentBlanquin · February 2016

mr. Wrobel, many of your plans simply dont work or do exact the opposite. Plese stop plagueing and plagueing the game with silly things and make it simpler and playable again, like it was before. stop thinking like everything you do is your child, can you?

bowmanz607 · February 2016

Hey guys why not try out the new 8mplementations before you knock it. I mean seriously, stop complaining about changes you know nothing about. You can't say you have tested them because there is zero way to replicate the live server. There is simply not enough people on pts to test this out. If you still have complaints before dark brotherhood then fine. [snip]

[edited for baiting]

ToRelax · February 2016

bowmanz607 wrote: »

Hey guys why not try out the new 8mplementations before you knock it. I mean seriously, stop complaining about changes you know nothing about. You can't say you have tested them because there is zero way to replicate the live server. There is simply not enough people on pts to test this out. If you still have complaints before dark brotherhood then fine. [sinp]

There usually is an update about every 3 months. If we are not to complain before one, then when ever? Also I am pretty sure there is enough evidence the players aren't (if even) the only ones who know nothing about how changes will play out.

[edited to remove quote]

bowmanz607 · February 2016

ToRelax wrote: »

bowmanz607 wrote: »

Hey guys why not try out the new 8mplementations before you knock it. I mean seriously, stop complaining about changes you know nothing about. You can't say you have tested them because there is zero way to replicate the live server. There is simply not enough people on pts to test this out. If you still have complaints before dark brotherhood then fine. [snip]

There usually is an update about every 3 months. If we are not to complain before one, then when ever? Also I am pretty sure there is enough evidence the players aren't (if even) the only ones who know nothing about how changes will play out.

You have every right to complain. But how can you complain about something we have no idea about. Ure I'm skeptical myself. We can theorize all day about what may or may not happen, but that is all they are, theories. If the players PROVE it doesn't work and more has to be so be it. Spouting theories about what will happen without proof convinces no one. To be fair no aoe caps was a bit op at times in case people don't remember. Although the current system is obviously flawed. I think these cages are a step in the right direction without over stepping. Sure the players may have a better understanding of mechanics etc, but we still have zero idea what will actually happen until it does. Instead of yelling about things that have not been properly tested yet, go out, test it, then come back with wvidence.

[edited to remove quote]

VincentBlanquin · February 2016

bowmanz607 wrote: »

ToRelax wrote: »

bowmanz607 wrote: »

Hey guys why not try out the new 8mplementations before you knock it. I mean seriously, stop complaining about changes you know nothing about. You can't say you have tested them because there is zero way to replicate the live server. There is simply not enough people on pts to test this out. If you still have complaints before dark brotherhood then fine. [snip]

There usually is an update about every 3 months. If we are not to complain before one, then when ever? Also I am pretty sure there is enough evidence the players aren't (if even) the only ones who know nothing about how changes will play out.

You have every right to complain. But how can you complain about something we have no idea about. Ure I'm skeptical myself. We can theorize all day about what may or may not happen, but that is all they are, theories. If the players PROVE it doesn't work and more has to be so be it. Spouting theories about what will happen without proof convinces no one. To be fair no aoe caps was a bit op at times in case people don't remember. Although the current system is obviously flawed. I think these cages are a step in the right direction without over stepping. Sure the players may have a better understanding of mechanics etc, but we still have zero idea what will actually happen until it does. Instead of yelling about things that have not been properly tested yet, go out, test it, then come back with wvidence.

ok, i will especially wait for your feedback next month, stay tuned man

[edited to remove quote]

dylanjaygrobbelaarb16_ESO · February 2016

the reason that no aoe caps are good are as follows. 5 vs 20 the numbers have a huge advantage with lots of different tactics. single target is still relevant for the big group bucause with so many the can burst down one at a time crazy fast without needing to stack up if they chose to. the only chance smaller numbers have is if the big group chooses to stay stacked you can hit all of them at once, why have the cap at all, the big group has more dps/heals more tactical options via numbers. why make it 20v5 is an auto win why not make it so with no caps they probably will win but if they are sloppy or refuse to do anything but stack its at least possible to do full damage to all of them.
the other issue is how it chooses who gets damaged, if player x is low life but there is a bunch of players stacked so instead of him taking full damage another player does and player x takes greatly reduced damaged. this creates a situation where you get defensive benefits from all balling up together. balling up inside a big fireball or whatever it is SHOULD NOT BE A BENEFIT

apostate9 · February 2016

themdogesbite wrote: »

I guess you never saw PvP pre 1.6 Wrobel. 5 man groups could put up very strong fights against 20 people.. IF you will change healing you need to chnage it in a way that isnt punishing smaller sized groups. Keep heal cap on 6 but completely REMOVE the damage cap, that solves all problems. Numbers shouldnt matter, Skill howeever, should.

The reason healing can outscale damage is AOE caps and turteling.......

Numbers shouldn't matter? Right......

RobbaYaga · February 2016

@Wrobel

The simplest fix from a game stand point and probably the harder fix from a coding stand point is to bring some reality into the game:

Players (and horses) should take up space and block LoS and AoE abilities should be affected by LoS.
(Goodness knows NPCs can certainly take up space and block LoS.)

Zergs are less of a problem when they cannot stand on top of each other anymore.

I can hide behind a tree and not get hit with a single target ability, the same should be true of an AoE.

If I am standing behind someone when a grenade goes off on the other side of them, I will have some protection. If an AoE goes off it should be mitigated (or eliminated) by the person in front of me. The same is true for healing. If that magic beam of healing goodness can't get to me because someone is between me and it, then the healing I receive should be mitigated (or eliminated).

This change also would make "Tanks" relevant in PvP as they could block entryways (think 300) and protect those behind them.

Skills that operate differently may not be affected by this LoS change. Caltrops should affect everyone in the AoE equally, but only when they attempt to move. Fire Rune (and it's morphs) seem to blast upward, so anyone standing on it should probably get hit the same. There are probably other examples.

I really hate Proxy Detonation, but if you want it to be a Zerg buster, then maybe it damages all enemies equally, but is mitigated by allies in between me and the enemies (to prevent 5 guys running through a crowd and vaporizing them, though again, if they CANNOT run THROUGH the enemy players, only the people in the front get hit).

Adjusting AoEs to be affected by LoS may be challenging, but making players take up space is the easier piece to this and I think everyone would be pleasantly surprised how well a tiny dose of reality fixes a lot of things.

hrothbern · February 2016

RobbaYaga wrote: »

@Wrobel

The simplest fix from a game stand point and probably the harder fix from a coding stand point is to bring some reality into the game:

Players (and horses) should take up space and block LoS and AoE abilities should be affected by LoS.
(Goodness knows NPCs can certainly take up space and block LoS.)

Zergs are less of a problem when they cannot stand on top of each other anymore.

I can hide behind a tree and not get hit with a single target ability, the same should be true of an AoE.

If I am standing behind someone when a grenade goes off on the other side of them, I will have some protection. If an AoE goes off it should be mitigated (or eliminated) by the person in front of me. The same is true for healing. If that magic beam of healing goodness can't get to me because someone is between me and it, then the healing I receive should be mitigated (or eliminated).

This change also would make "Tanks" relevant in PvP as they could block entryways (think 300) and protect those behind them.

Skills that operate differently may not be affected by this LoS change. Caltrops should affect everyone in the AoE equally, but only when they attempt to move. Fire Rune (and it's morphs) seem to blast upward, so anyone standing on it should probably get hit the same. There are probably other examples.

I really hate Proxy Detonation, but if you want it to be a Zerg buster, then maybe it damages all enemies equally, but is mitigated by allies in between me and the enemies (to prevent 5 guys running through a crowd and vaporizing them, though again, if they CANNOT run THROUGH the enemy players, only the people in the front get hit).

Adjusting AoEs to be affected by LoS may be challenging, but making players take up space is the easier piece to this and I think everyone would be pleasantly surprised how well a tiny dose of reality fixes a lot of things.

reality wise..... giving every object, including players, a physical size, add collision and shadow effects (line of sight & line of damage/heals for certain abilities)....

reality wise.... it would be awesome indeed

But that will cause a completely new game mechanics..... and also current Server capabilities will not be able to handle that at all !

So... I guess we have to have some patience there... one or two decades or so.

bowmanz607 · February 2016

RobbaYaga wrote: »

@Wrobel

The simplest fix from a game stand point and probably the harder fix from a coding stand point is to bring some reality into the game:

Players (and horses) should take up space and block LoS and AoE abilities should be affected by LoS.
(Goodness knows NPCs can certainly take up space and block LoS.)

Zergs are less of a problem when they cannot stand on top of each other anymore.

I can hide behind a tree and not get hit with a single target ability, the same should be true of an AoE.

If I am standing behind someone when a grenade goes off on the other side of them, I will have some protection. If an AoE goes off it should be mitigated (or eliminated) by the person in front of me. The same is true for healing. If that magic beam of healing goodness can't get to me because someone is between me and it, then the healing I receive should be mitigated (or eliminated).

This change also would make "Tanks" relevant in PvP as they could block entryways (think 300) and protect those behind them.

Skills that operate differently may not be affected by this LoS change. Caltrops should affect everyone in the AoE equally, but only when they attempt to move. Fire Rune (and it's morphs) seem to blast upward, so anyone standing on it should probably get hit the same. There are probably other examples.

I really hate Proxy Detonation, but if you want it to be a Zerg buster, then maybe it damages all enemies equally, but is mitigated by allies in between me and the enemies (to prevent 5 guys running through a crowd and vaporizing them, though again, if they CANNOT run THROUGH the enemy players, only the people in the front get hit).

Adjusting AoEs to be affected by LoS may be challenging, but making players take up space is the easier piece to this and I think everyone would be pleasantly surprised how well a tiny dose of reality fixes a lot of things.

They gave many reasons why they don't want to add collision some time ago. Think about this. Ad is fighting roe. The inner 8s down. Ad goes to push in,but before they do a group of DC that rolled ad forms a line at the breach. Now the keep is protected and there is nothing you can do about it cause you can't kill your own alliance.

jbcrocks · March 2016

Any news? @Wrobel

Aektann · March 2016

@ZOS just ignoring this topic. @Wrobel remove AOE caps!
Or try to name one adequate reason for them to leave , at least one?

DaybreakerMusic · March 2016

Aektann wrote: »

@ZOS just ignoring this topic. @Wrobel remove AOE caps!
Or try to name one adequate reason for them to leave , at least one?

zos always does.
They are ignoring the complete playerbase... otherwise they wouldnt *** things up so hard like this recent patch
dont try to deny it. everyone knows this is the truth
sadly

dylanjaygrobbelaarb16_ESO · March 2016

get rid of it gets rid of calculations. the pop locked lag is at times unplayable. while your at it reduce the max pop again. your servers cant handle it

borchan · March 2016

The aoe cap should be removed, it was a huge mistake implanting it in the first place.

jbcrocks · March 2016

Tho if aoe caps are removed. Skills like prox det and sets like vicious death should be removed too.

vontariel · March 2016

If you really want to add more tactical aspect to PvP, then you should consider some of it. There are just examples, but they are clearly show direction on what I understand as a "tactical" combat:
-You should never have best offensive and defensive capabilities at same time. Just never. If you ar geared for dealing tons of damage, you should receive tons of damage, plain and simple.
Some skills can be used to increase your overall survival but never should be the case, that player in robe can last longer than player in heavy, and especially when player in robe is capable of dealing enormous damage at the same time. Not only it creates unhealthy and unfriendly environment, it also forces players to "dps race" meta, because they would have enough survival anyway. Obviously it also strikes tanks, glaciers and bruisiers from being valid choice.
-Buff heavy armor defensive capabilites (not so much), resource managment (for quite margin).
Resolve numbers could be slightly increased.
Constitution should be changed to be 5 heavy bonus, and 8 second cooldown removed.
Bracing should now scale with how many parts you have, giving you 2% block redu per heavy part.
Rapid mending should increase any resource income, not only health.
-Add tanks valid role in cyrodiil. Apart one armor set they can't intercept attacks. Maybe make unstopable basic verion of immovable, and change unstopable to: "additional effect: you can intercept all single target projectile attacks against your allies. Intercepted attacks damage deal only 50% damage.
-AoE caps from ground-based AoE should be removed. Anybody can just walk out from it, so there should be no reason for them to be affected from cap. Also damage for some (most) ground-based AoE could be increased.
-All nova-type or Circle projectile-type AoE should have 2(or even 3 for big ones) zones. Smaller full damage middle area, and reduced damage outer ring. It'd be much better than dynamic scaling from center both for players (easier to predict what damage could be done) and server (less computing, just one IF more in code).
-Change aim for arrow spray to be reticle based. And raise damage of it.
-Raise damage on Cleave, and widen a little affected area. Maybe change carve for empower-type mechanism (the more enemies are hit, the more damage it will deal to each one).
-CC immunity is too long. While it's easiest mechanism to implement against CC spam, much better approach (for both sides) would be if consecutive cc of that category (ie. slow) would be 50% shorter than one before (to a 0.5s cap), and after 8?10? seconds of no cc of that type, DR will be reset. CC break will break current cc's, give some immunity time, but it will reset DR timer. You can't reapply CC of that type if one is currently in effect. This would prevent for constant spam-refresh CC which would burn down CC timer quickly.
-Currently there are 3 skills that reflects attack to enemy, 2 of them are class specific (eclipse, reflective scales), one weapon specific (Defensive posture). That means, only one reflective mechanic is available to all classes, and it's worst one. Reflective scales have no real downsides (apart 4 projectiles threshold), and it can't be countered in any reliable way. Also RC is only reflect to affect bow.
-All defensive spells (shields, reflects, immunities, etc. NOT heals) should have DR mechanic, which would disable option of spamming it. It should work in a same way as DR for CC i mentioned above.
-If for some reason shorter timer can't be apllied, double cost for each consecutive usage. Resources usage would rise geometrically which will discourage spamming it.
-Disable block-cast. You should never have best offensive and defensive capabilities at same time.
-Disable magicka regen when blocking. Only health regen should be allowed when blocking.
-If there is a shield buff it's power should always scale from HP, and never from your damage resources. You should never have best offensive and defensive capabilities at same time.
-Add AoE "flashbang".Very Low DMG, weak slow, disorient. Major maim, minor fracture, minor breach, minor mangle. last 8 seconds.
-Add acid bomb stamina skill, which will work in a same way as green circle from trolls on Sanctum Ophidia, and apply minor defile.
-Add "dark bond" skill. Creates tether between player and target. All enemies who touch beam will receive moderate-big damage. Beam lasts 8 seconds, and will break if players are further than 25m. One morph can be: cast on enemy and slow him for 50% for whole beam duration, this CC is never reduced, but after it target would receive slow immunity for 4 seconds. Maximum range of beam is increased to 30m. Second morph: Stamina based. Cast on friendly target for both players to receive major expedition for whole beam duration.
-Add AoE (small-medium area) "silence zone" skill which will prevent from casting magicka abilities. One of morphs could be "pacifism" which would disable ALL abilities inside of zone. Other morph could enlarge "silence zone".
-Change blood altar: All single target non-DoT attacks of allies inside zone restores 10-15% of their hp. Altar is targetable by single target spells and abilities (it can have similar/less HP as explosive scamp portal in IC). Destroying altar would shield enemy player for 100% of his HP. After destruction/expiration ground is defiled and no altar can be placed for 20s. Altar cannot be over defiled area. In PvE altar is highest target priority except for taunt, but monsters will never receive HP shield.
-make quick cloak basic skill and change quick morph to:"trickster cloak": if there is another player or monster between you and attacker all ranged single target abilities are intercepted by him. Yes this will allow semi-friendly fire, hiding from damage behind allies and monsters and encourage much more careful positioning from attacker. Or make it as separate new skill not tied to DW. However i think dual wield NEEDS some love ATM, and this would be great buff for both PvP and PvE (imagine hiding behind tank during trials/dungeons).

Merlight · March 2016

vontariel wrote: »

(less computing, just one IF more in code)

That may have been better before Pentium Pro, but on today's processors, branching is often more expensive than a straight burst of arithmetic instructions.

vontariel · March 2016

Merlight wrote: »

vontariel wrote: »

(less computing, just one IF more in code)

That may have been better before Pentium Pro, but on today's processors, branching is often more expensive than a straight burst of arithmetic instructions.

Branching problem was more or less "solved" (diminished to nearly irrelevant state), by introducing branch predictors like 20 years+ ago. And modern processors, have little-to-none problem with branching.
Also I don't think, that potential benefit of 5 (at most) lost cycles is relevant enough to use less clear design. I'd say even that code optimalization is nowhere close to point for such things to be relevant.

But thank you for rising technical point in discussion

hrothbern · March 2016

vontariel wrote: »

Merlight wrote: »

vontariel wrote: »

(less computing, just one IF more in code)

That may have been better before Pentium Pro, but on today's processors, branching is often more expensive than a straight burst of arithmetic instructions.

Branching problem was more or less "solved" (diminished to nearly irrelevant state), by introducing branch predictors like 20 years+ ago. And modern processors, have little-to-none problem with branching.
Also I don't think, that potential benefit of 5 (at most) lost cycles is relevant enough to use less clear design. I'd say even that code optimalization is nowhere close to point for such things to be relevant.

But thank you for rising technical point in discussion

I was 35 years ago heavily involved in Computer chess Artificial Intelligence on not that powerfull 8 bit microprocessors, programming in machine laguage, and having smart mathematical/game theory solutions to handle the branching was pretty much fundamental to get some decent chess out of your program.

The technology development of the processors was tremendous in the past decades allowing for more and more "brute force" handling in the last decades.

@vontariel , I am certainly not up to date anymore, but the amount of branches per player is quite terrific, also because of buffs, debuffs from other players.

Is the current state of the art really so that predictive brancing does work effectively ?

Merlight · March 2016

vontariel wrote: »

Branching problem was more or less "solved" (diminished to nearly irrelevant state), by introducing branch predictors like 20 years+ ago.

Sure, but for the suggested "IF distance < X", I don't think there'd be a predictable pattern (depends on X as well as player behaviour), so the prediction might fail pretty often. Anyway, I just felt like nitpicking, as the post was a bit too extensive to react to as a whole so early in the morning

Totally agree with the rest of your reply.

vontariel · March 2016

Merlight wrote: »

vontariel wrote: »

Branching problem was more or less "solved" (diminished to nearly irrelevant state), by introducing branch predictors like 20 years+ ago.

Sure, but for the suggested "IF distance < X", I don't think there'd be a predictable pattern (depends on X as well as player behaviour), so the prediction might fail pretty often. Anyway, I just felt like nitpicking, as the post was a bit too extensive to react to as a whole so early in the morning Totally agree with the rest of your reply.

It all comes in implementation.
All of this values are checked (position) already, so for damage computing other method/function could be executed, based on this values. Current branch solutions are smart enough to solve that situation, and with properly written code there would be no decrease in performance at all.
You are technically you're correct (the best one type of being correct), that predication will fail (but while still high, fail rate, would be far less than 50%), when code is not properly optimalised, and there are constant calls to compute some data, rather than storing that values. In that case there will be some strain, but it would be still insignificant from practical point of view.

Currently there are much bigger problems with algorithms from design point, which we can easily observe on Cyrodiil.

hrothbern · March 2016

BTW

If ESO can crack the lag fundamentally in a smart way with still:

enabling a multitude players close to each other
complex features from sets, abilities and passives
high feeling of responsiveness
working rotations including weapon swapping

ESO will own.... will have a very important competitive edge

I don't know if the scale size of the staff of ESO will allow to invest for some time in a small A.I. team to crack it.
Or cooperate with a leading University on this, picking the right brains to develop that IP.

But it would be tremendous

vontariel · March 2016

hrothbern wrote: »

BTW

If ESO can crack the lag fundamentally in a smart way with still:
enabling a multitude players close to each other

complex features from sets, abilities and passives

high feeling of responsiveness

working rotations including weapon swapping

ESO will own.... will have a very important competitive edge

I don't know if the scale size of the staff of ESO will allow to invest for some time in a small A.I. team to crack it.
Or cooperate with a leading University on this, picking the right brains to edevelop that IP.

But it would be tremendous

Point 1,3,4 can be easily* achived by thick(er) client, thin(er) server approach.
Main problem is with proper data security and validation (i.e. anti-cheat algorithms) both on server and client side. And potential anti-cheat solution with more than 99.99% success rate (it means 1 per 10 000 players will be falsely accused of cheating, or cheater would not register as such) should have much less strain on server than current counting logic. But it can be achieved.
Point 2 is to achieve by different team, and optimistically it should be only amount of time, by analizing raw data and taking to account player feedback (not only top, but also bottom of ladder).

*It's easy, but only from design/theory point. There are many difficult and non-trival errors, cases and issues based on architecture, code structure, chosen tools and so on, to take into account, which is not EASY task, and it's TONS of workhours. In fact i'm more than 99% sure that this solution would never be done, because it's too expensive and requires quite team of experienced developers solely dedicated to this issue for really good time. Realistically if ZoS would start such work in april (not counting recruiting new people), we would have chances to see it on halloween. 3 months of development 1 month of inner QA, up to 0.5 month of fixing, 0.5-1 open beta, rest of time to polish. Maybe shorter dev time, but not less than 2 months.

hrothbern · March 2016

vontariel wrote: »

hrothbern wrote: »

BTW

If ESO can crack the lag fundamentally in a smart way with still:
enabling a multitude players close to each other

complex features from sets, abilities and passives

high feeling of responsiveness

working rotations including weapon swapping

ESO will own.... will have a very important competitive edge

I don't know if the scale size of the staff of ESO will allow to invest for some time in a small A.I. team to crack it.
Or cooperate with a leading University on this, picking the right brains to edevelop that IP.

But it would be tremendous

Point 1,3,4 can be easily* achived by thick(er) client, thin(er) server approach.
Main problem is with proper data security and validation (i.e. anti-cheat algorithms) both on server and client side. And potential anti-cheat solution with more than 99.99% success rate (it means 1 per 10 000 players will be falsely accused of cheating, or cheater would not register as such) should have much less strain on server than current counting logic. But it can be achieved.
Point 2 is to achieve by different team, and optimistically it should be only amount of time, by analizing raw data and taking to account player feedback (not only top, but also bottom of ladder).

*It's easy, but only from design/theory point. There are many difficult and non-trival errors, cases and issues based on architecture, code structure, chosen tools and so on, to take into account, which is not EASY task, and it's TONS of workhours. In fact i'm more than 99% sure that this solution would never be done, because it's too expensive and requires quite team of experienced developers solely dedicated to this issue for really good time. Realistically if ZoS would start such work in april (not counting recruiting new people), we would have chances to see it on halloween. 3 months of development 1 month of inner QA, up to 0.5 month of fixing, 0.5-1 open beta, rest of time to polish. Maybe shorter dev time, but not less than 2 months.

Thanks @vontariel , for your explanation

You say also:
"Main problem is with proper data security and validation (i.e. anti-cheat algorithms) both on server and client side. And potential anti-cheat solution with more than 99.99% success rate (it means 1 per 10 000 players will be falsely accused of cheating, or cheater would not register as such) should have much less strain on server than current counting logic".

IF, I don't know,.... but IF ESO does a 100% check on cheating by some algorithm....
My first target to adress would be that 100% check....

It is like having a police officer per citizen to control no crime is committed.

Why not a 1% check per global cooldown of 1 second and cycle around all players in a 100 second cycle !
Once you spotted a player, you tag him for 100% control and if abuse is certain you ban.
This would lessen the server load to close to 1% compared to actual.

There is no real harm done when a player can abuse for 100 seconds only.

You could even use several algorithms with differing costs (in time) and effectiveness to optimise.
have some heuristic approach on odd performances as well to escalate

Or are ZOS already doing something like this ?

vontariel · March 2016

hrothbern wrote: »

vontariel wrote: »

hrothbern wrote: »

BTW

If ESO can crack the lag fundamentally in a smart way with still:
enabling a multitude players close to each other

complex features from sets, abilities and passives

high feeling of responsiveness

working rotations including weapon swapping

ESO will own.... will have a very important competitive edge

I don't know if the scale size of the staff of ESO will allow to invest for some time in a small A.I. team to crack it.
Or cooperate with a leading University on this, picking the right brains to edevelop that IP.

But it would be tremendous

Point 1,3,4 can be easily* achived by thick(er) client, thin(er) server approach.
Main problem is with proper data security and validation (i.e. anti-cheat algorithms) both on server and client side. And potential anti-cheat solution with more than 99.99% success rate (it means 1 per 10 000 players will be falsely accused of cheating, or cheater would not register as such) should have much less strain on server than current counting logic. But it can be achieved.
Point 2 is to achieve by different team, and optimistically it should be only amount of time, by analizing raw data and taking to account player feedback (not only top, but also bottom of ladder).

*It's easy, but only from design/theory point. There are many difficult and non-trival errors, cases and issues based on architecture, code structure, chosen tools and so on, to take into account, which is not EASY task, and it's TONS of workhours. In fact i'm more than 99% sure that this solution would never be done, because it's too expensive and requires quite team of experienced developers solely dedicated to this issue for really good time. Realistically if ZoS would start such work in april (not counting recruiting new people), we would have chances to see it on halloween. 3 months of development 1 month of inner QA, up to 0.5 month of fixing, 0.5-1 open beta, rest of time to polish. Maybe shorter dev time, but not less than 2 months.

Thanks @vontariel , for your explanation

You say also:
"Main problem is with proper data security and validation (i.e. anti-cheat algorithms) both on server and client side. And potential anti-cheat solution with more than 99.99% success rate (it means 1 per 10 000 players will be falsely accused of cheating, or cheater would not register as such) should have much less strain on server than current counting logic".

IF, I don't know,.... but IF ESO does a 100% check on cheating by some algorithm....
My first target to adress would be that 100% check....

It is like having a police officer per citizen to control no crime is committed.

Why not a 1% check per global cooldown of 1 second and cycle around all players in a 100 second cycle !
Once you spotted a player, you tag him for 100% control and if abuse is certain you ban.
This would lessen the server load to close to 1% compared to actual.

There is no real harm done when a player can abuse for 100 seconds only.

You could even use several algorithms with differing costs (in time) and effectiveness to optimise.
have some heuristic approach on odd performances as well to escalate

Or are ZOS already doing something like this ?

It's impossible to reach 100%. Every software bugs, glitches and so on can cause false alarms even for legit players.

I even proposed in other thread for similar "policeman" idea:
- Firstly, client and server generate map with potential output for that player (considering his skill and item loadout and all possible buffs), such map is used to valid player outputs (one comparison per action). Secondly, per map, random seed is generated, that is used to generate all pseudo-random values for that player for current loadout.
- Now we have mechanism to check if sent data is proper. But because we don't want to compare all of time against seed (to save computing power), good watchdog should be implemented.
-Watchdog should monitor (ie checking every number posted by client) in several cases:
-any number exceeded one in map for that particular skill.
- random check
- Something involved memory violation or netprotocol.
- other.
And once we have good watchdog we can have good filtering which are false alarm, and are potential cheats.
From this point you get the idea.

I'm more than sure that there is some sort of that even in current implementation. However we can see easily in cyrodiil that for some reason it's not good enough solution.

KenaPKK · April 2016

If we let this thread die, they'll never change AoE caps.

ToRelax · April 2016

KenaPKK wrote: »

If we let this thread die, they'll never change AoE caps.

Then we better don't let that happen.

wigglesgaming26 · April 2016

It doesn't matter if we "let it die" or not, this discussion is already dead to Zenimax. You're just beating a dead horse at this time guys