PTS Update 45 - Feedback Thread for Cyrodiil Champions (Vengeance Campaign)

MincMincMinc · January 24

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.

As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

sarahthes · January 24

MincMincMinc wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.
As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

A Golden Pursuit tied specifically to doing things in cyrodiil during Vengeance would probably be a great way to deliver this.

YandereGirlfriend · January 24

sarahthes wrote: »

MincMincMinc wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.
As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

A Golden Pursuit tied specifically to doing things in cyrodiil during Vengeance would probably be a great way to deliver this.

This is the way.

MincMincMinc · January 24

sarahthes wrote: »

MincMincMinc wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.
As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

A Golden Pursuit tied specifically to doing things in cyrodiil during Vengeance would probably be a great way to deliver this.

Yea someone pointed out another good idea to advertise the rewards and the campaign as a big event when you log in. Much like how we get the featured crown store advert windows.

Khenzy · January 24

Erickson9610 wrote: »

xylena_lazarow wrote: »

sarahthes wrote: »

And I think anyone who actually cares about PvP in ESO will participate in these tests even if the classes themselves are not balanced. For the purpose of the test I don't think it matters if there's 1000 nbs and 3 necros. What matters is there's 1003 people.

They're not getting 1003 people unless they put forth a product that 1003 people are excited about. Diet Cyrodiil with an unbalanced unfun NB hell meta isn't gonna be it. They're gonna need to deal with at least that. There's no performance reason to mash together a burst heal with invisibility, didn't we already go through this with Hardened Ward?

Current Cyrodiil isn't exciting enough for 1003 people. This test, with standardized templates, is the most exciting update to Cyrodiil in a long time. A lot of people hate it, but a lot of people are interested by it, too. This will bring in many players when it hits Live.

This may be my first post in over 6 years.

Myself and approximately 20+ friends are following VERY closely the development of this test as we are very interested in being able to play in a much more curated and controlled Cyrodiil environment where the determining factors are more the skill of the player rather than the myriad factors introduced by the 'proc' sets or the myriad of passives that ultimately introduce an overdependence on these uncontrollable factors.

We are VERY pleased with the fact that they are also trying to balance the power of the immortal masses of zerglings all gathering together by limiting the areas to only 3 players.

We also see very positively the discreet introduction of cooldowns in certain skills.

Even if it's just a performance test, we are quite in favor of the vision they are trying to convey with the rebalancing of the skills of each class.

Personally, I think this is a very good approach for Cyrodiil. if they implement this as they are doing while slowly introducing very carefully curated and balanced equipment sets and other skill lines with the same more simplified balancing approach, we may be in for a very fun Cyrodiil experience, and if this is the overall intent, we will definitely be back.

PS: A small side point I would like to make as a suggestion is that it would not be a bad thing to leave aside the “buff maintanance” issue that has been dictating combat for 10 years and that acquiring an important buff to be a byrpoduct of simply having used the active ability in combat. I genuinely think important buffs should last much, much longer and skills that provide them should have more of an active machanic.

YandereGirlfriend · January 24

Khenzy wrote: »

Erickson9610 wrote: »

xylena_lazarow wrote: »

sarahthes wrote: »

And I think anyone who actually cares about PvP in ESO will participate in these tests even if the classes themselves are not balanced. For the purpose of the test I don't think it matters if there's 1000 nbs and 3 necros. What matters is there's 1003 people.

They're not getting 1003 people unless they put forth a product that 1003 people are excited about. Diet Cyrodiil with an unbalanced unfun NB hell meta isn't gonna be it. They're gonna need to deal with at least that. There's no performance reason to mash together a burst heal with invisibility, didn't we already go through this with Hardened Ward?

Current Cyrodiil isn't exciting enough for 1003 people. This test, with standardized templates, is the most exciting update to Cyrodiil in a long time. A lot of people hate it, but a lot of people are interested by it, too. This will bring in many players when it hits Live.

This may be my first post in over 6 years.

Myself and approximately 20+ friends are following VERY closely the development of this test as we are very interested in being able to play in a much more curated and controlled Cyrodiil environment where the determining factors are more the skill of the player rather than the myriad factors introduced by the 'proc' sets or the myriad of passives that ultimately introduce an overdependence on these uncontrollable factors.

We are VERY pleased with the fact that they are also trying to balance the power of the immortal masses of zerglings all gathering together by limiting the areas to only 3 players.

We also see very positively the discreet introduction of cooldowns in certain skills.

Even if it's just a performance test, we are quite in favor of the vision they are trying to convey with the rebalancing of the skills of each class.

Personally, I think this is a very good approach for Cyrodiil. if they implement this as they are doing while slowly introducing very carefully curated and balanced equipment sets and other skill lines with the same more simplified balancing approach, we may be in for a very fun Cyrodiil experience, and if this is the overall intent, we will definitely be back.

PS: A small side point I would like to make as a suggestion is that it would not be a bad thing to leave aside the “buff maintanance” issue that has been dictating combat for 10 years and that acquiring an important buff to be a byrpoduct of simply having used the active ability in combat. I genuinely think important buffs should last much, much longer and skills that provide them should have more of an active machanic.

If you want skill cooldowns you're playing the wrong game. Plenty of other MMOs out there that obey that convention. Most ESO players play this game specifically because it doesn't have those clunky systems. And those players will immediately dip the moment that they begin creeping into this game.

Kaelthorn_Nightbloom · January 24

MincMincMinc wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.
As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

I think this is a reasonable reply. Thank you.

My personal observation is that there is some O(n^2) processing when too many players converge on the same area. It's exacerbated by ball groups which pushes the packet queue or cpu usage very high and causes ability delay on the server. I've written about this before on the ESO forums many years ago from a Computer Science perspective and I have a background in game development and server engineering as well.

I predict that maybe a few abilities will get tweaked but nothing on a grand scale. Also, they may identify some code for the root cause and optimize to fix the issue. Like a rewrite on their server's internal event queue, network layer, etc.

Estin · January 24

sarahthes wrote: »

MincMincMinc wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.
As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

A Golden Pursuit tied specifically to doing things in cyrodiil during Vengeance would probably be a great way to deliver this.

This is actually a great idea. The rewards need to be enticing, though, and encourage players to participate on more than one class.

There could be tasks that are class specific such as kill 150 players as a nightblade which can reward a skill style for that class or 1 piece of a style set like december's golden pursuit. Milestones could reward crates, a pet, a mount, and even a furnishing, all themed around vengeance cyrodiil. This would definitely give players, especially those who don't PvP, a reason to join in for the test and help hit population cap. Since it's templated, you can hop in on a fresh character and not have to worry.

Kaelthorn_Nightbloom · January 24

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

sarahthes · January 24

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

MincMincMinc · January 24

SkaraMinoc wrote: »

MincMincMinc wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

Yeah that is great feedback for when the test actually plays out. Right now we are testing functionally will the test work? Will this setup create the data necessary? What bugs are present?

I'm sorry I think comments like "this aint it chief" do not align with the above goals. We can talk biased balance talking points after the test ends and if zos decides to go forward with a skill pass. (probably years out). Of course nobody wants to lose their power creep, but this test is to directly remove all power creep and create a baseline. We can't do a power creep removal test by keeping the power creep.
As it stands vengeance classes are very reasonably balanced. The kits are comprehensive enough to function in 1v1, 10v10, 100v100, and even 1vX scenarios. The balance of sustain, healing, tank, and damage is perfectly fine.

There are some tacked on buffs that should be removed to stay true to simplification like sorc having berserk on crit surge. Some of the dlc classes also have similar tacked on extra buffs that are unnecessary.

Incentives are not present. Ap and xp is worthless in today's game. For this test to be successful you need to effectively pay people to go through with it. Crowns, mounts, crates, Mythic leads, Furnishings, Perfected arena weapons, A unique pet, a unique title, etc. We should not be trying to get people in the test by fixing the arbitrary balance that nobody will agree on anyways and will be deleted after a week.

I think this is a reasonable reply. Thank you and I agree with most of what you said.

My personal observation is that there is some O(n^2) processing when too many players converge on the same area. It's exacerbated by ball groups which pushes the packet queue or cpu usage very high and causes ability delay on the server. I've written about this before on the ESO forums many years ago from a Computer Science perspective and I have a background in game development and server engineering as well.

I predict that maybe a few abilities will get tweaked but nothing on a grand scale. Also, they may identify some code for the root cause and optimize to fix the issue. Like a rewrite on their server's internal event queue, network layer, etc.

I'm sure you can understand me wanting to push the conversation in the direction of does it work, ok, lets get results and see what happens.

I'm a MechEng that dabbles as well (testing, engine/car design, UE5), so I am sure you understand the background company politics. The manager and team are being asked what do we do for pvp. They brought up the paths, either combat changes they noted in the last tests OR hardware changes. The directors then require the managers to show evidence as to which option makes the most sense. Do we spend money on hardware again(IMO they are likely to abandon pvp in this case) OR do we prioritize this before other system reworks?

The directors will want a baseline to go off of that is representative on the current hardware, software, graphics, etc available. We are 15+ years down the tech line when the original tests were done. The only way to get a baseline is to have the pure basic essentials and disregard everything else. After the baseline you add in systems to measure performance differences. Future standardization and combat directions for the next 10 years will depend on this test.

All the while customers are screaming asking for sensitive company information that'd affect shareholder value. In this case microsoft probably doesn't want them giving out open book details.

We have all written about it one time or another. Sadly the way forums work is typically someone says a good point and then 10x people disagree. The people who agree just move on. So if you do not keep up with the 10x people this looks bad for the good point.

As much as I think proc(damage) sets are a big mistake combat design wise, they are only a drop in the bucket. Compared to a 12man x 10s hot x 12manAoE x 5 passive buffs x 10s buff timers x Smart healing x 3 proc sets x status effects......x....x.....x...x.....x. Not hard to imagine there would be issues compared to a skill that " deals 5k damage". Burst from proc sets just allows players to be more lazy skill choice wise. Instead of players choosing the simple deal 5k damage single target skill, suddenly now they can just pair it with an aoe dot pressure which does 10x to 100x the tick events.

I predict if the test is a success we will see skills and classes have their fat trimmed. Necro would lose random % buffs on everything. Sorc would lose random buffs like berserk on DD. etc. Combat wise I think the active effects are drastically under powered compared to the passive power creep that keeps being added. Players will enjoy combat more when the damage skill they press does damage they can see.

MedicInTheWild · January 24

@ZOS_Kevin

I noticed that Sturdy and Well Fitted are the only traits that seem to work, not sure if intended or not.

I know this is a test but it seems the way to win fights is to have the most people in one spot. I have tested all classes and recovery seems to be the worst part of this, taking resources that are leveled while solo sucks due to how long it takes. Thinking of what will happen when live is with running out of resources while in big fight is when zergs show up they win.

I know that is what they want during this test is big battles but I am not wanting to play mount simulator to get back to a fight because one faction is running a 36+ man zerg taking everything.

I personally think we will see the decline of solo players as no incentive to play solo seems to be showing. The biggest group taking the well so they all have 77.5K health and running with 3 full groups is going to destroy anything in its path, which seems to be the current playstyle of the biggest portion of pvp due to various reasons ( Skill issues and not knowing how to not rely on others saving you is my personal thoughts ).

My next thought is are we gonna have frequent changes implemented into the live server adding new abilities/passive/skill lines or is ZOS trying to remove identity all together? The reason I say that is most people hang on to ESO due to the wide range of build availability they have to create. If I wanted to play a game with preset skills and generic classes I would have played a game that offered that playstyle in the beginning.

Fin

Kaelthorn_Nightbloom · January 24

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

MincMincMinc · January 24

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

Yeah cant be that hard to replicate as most builds boil down to a simple macro-able rotation. Especially the problematic ball group builds. Then just make them follow a path around a keep in tick range. Unless there is more of a problem with external issues or desync issues causing hangups that have been introduced over the years.

Many patches ago with the 100% mit desync bug we could see that it is possible to halt all healing calculations and then make them go through at once. Or atleast be corrected by the server in that one instance. I don't know how their backend works, but maybe this is why they need live external test subjects.

malistorr · January 24

This issue is pretty simple IMO. If the game (and PVP specifically) is important to people (it appears to be because of all the feedback) then ZOS is asking you to participate in a test for a week. The reward is possibly having PVP in this game improved so that server performance in the future is hopefully at least acceptable instead of really bad. You're either willing to spend some of your time helping them (and indirectly you) out or you're not. Considering there has been almost no changes to PVP in Cyrodiil for the history of this game, I'm personally happy that ZOS is doing anything at all here.

ZOS is not even going to release chapters for this game any longer! So it's clear they're spending less dev time on the game moving forward and I wouldn't be surprised if devs have left the game out of boredom or just better opportunities. ZOS is also likely moving people to other upcoming projects/games. So the fact that they're dedicating any significant time to PVP this year is honestly astonishing to me. Isn't it better than the "nothing" we usually get?

I'm willing to deal with a nerfed-down toon and unfair class balance for a week if it means the game willl perform better in the future. We're talking about 1-week guys! For people who claim to love the game or even just play it regularly I don't think this is a big deal personally. Did you really expect ZOS to spend weeks or months of dev time to fine tune a week-long test? If so, I don't think you have any clue how a game is made at all or how much time or effort it takes to make changes. I really don't mean to insult anyone here at all but if a 1-week test is going to cause people major stress in any way, when it will hopefully improve something you use/play, I think some need to reevaluate their life priorities. If you don't want to participate, go work out, spend time with friends and family, go touch grass and come back a week later. There is life outside of this game. Those who participate in the test, even if they have to die a lot more for a week, will hopefully be helping to improve something they use/play/like/love for a long time to come. I don't remember any death counter in the game and you lose nothing for deaths in PVP. If you're that worried about comments from guildmates then again, life priorities...

Kaelthorn_Nightbloom · January 24

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

Some takeaways.

It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.
The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.
It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.
The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.
They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.
They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

MincMincMinc · January 24

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

Right all of which I have been saying. It'd make no sense to solidify the current vengeance as the only pvp. Why stop half your playerbase from ever needing new content? Why would a pvper ever buy eso plus or dlc if the gear or progression plays no role. Again why i've been saying there is no point to complain about sets and skills missing. It was clearly just doomer drama bait, as it makes no sense by all accounts.

At most skills may be reworked or fixed in the next big pass to be more performative and failsafe. Maybe dk leap wont have to be so buggy? Maybe streak wont have so many components to it. etc. If they can up the player cap, they could probably make cyrodil one single campaign. Be nice to be able to just port into cyrodil from the map again.

sans-culottes · January 24

YandereGirlfriend wrote: »

Khenzy wrote: »

Erickson9610 wrote: »

xylena_lazarow wrote: »

sarahthes wrote: »

And I think anyone who actually cares about PvP in ESO will participate in these tests even if the classes themselves are not balanced. For the purpose of the test I don't think it matters if there's 1000 nbs and 3 necros. What matters is there's 1003 people.

They're not getting 1003 people unless they put forth a product that 1003 people are excited about. Diet Cyrodiil with an unbalanced unfun NB hell meta isn't gonna be it. They're gonna need to deal with at least that. There's no performance reason to mash together a burst heal with invisibility, didn't we already go through this with Hardened Ward?

Current Cyrodiil isn't exciting enough for 1003 people. This test, with standardized templates, is the most exciting update to Cyrodiil in a long time. A lot of people hate it, but a lot of people are interested by it, too. This will bring in many players when it hits Live.

This may be my first post in over 6 years.

Myself and approximately 20+ friends are following VERY closely the development of this test as we are very interested in being able to play in a much more curated and controlled Cyrodiil environment where the determining factors are more the skill of the player rather than the myriad factors introduced by the 'proc' sets or the myriad of passives that ultimately introduce an overdependence on these uncontrollable factors.

We are VERY pleased with the fact that they are also trying to balance the power of the immortal masses of zerglings all gathering together by limiting the areas to only 3 players.

We also see very positively the discreet introduction of cooldowns in certain skills.

Even if it's just a performance test, we are quite in favor of the vision they are trying to convey with the rebalancing of the skills of each class.

Personally, I think this is a very good approach for Cyrodiil. if they implement this as they are doing while slowly introducing very carefully curated and balanced equipment sets and other skill lines with the same more simplified balancing approach, we may be in for a very fun Cyrodiil experience, and if this is the overall intent, we will definitely be back.

PS: A small side point I would like to make as a suggestion is that it would not be a bad thing to leave aside the “buff maintanance” issue that has been dictating combat for 10 years and that acquiring an important buff to be a byrpoduct of simply having used the active ability in combat. I genuinely think important buffs should last much, much longer and skills that provide them should have more of an active machanic.

If you want skill cooldowns you're playing the wrong game. Plenty of other MMOs out there that obey that convention. Most ESO players play this game specifically because it doesn't have those clunky systems. And those players will immediately dip the moment that they begin creeping into this game.

Cooldowns or their absence have no bearing on my playing ESO. Many people on other forums have noted the combat system being what drove or kept them away. I think this game is fun, but I don’t by any stretch think the combat system is gaming’s Platonic ideal. It’s great that you’re having fun and seem to value this part of the experience more than other areas, but that’s arguably part of what has helped keep ESO from appealing to a broader audience.

React · January 24

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

One other interesting thing that was mentioned by Brian was that the "bogging down" we see server wide is the result of the global actions being limited by what can fit within the server frames. Brian said that the same thing happens on an individual basis client side when you're performing high APM actions, i.e animation cancelling/weaving/block cancelling/bash weaving, etc.

So that phenomena high APM players have been describing for years where they feel like they need to stop performing actions to let the game "catch up", or where things become unresponsive when they're operating at high APM, is actually the reality of how things work now.

Hopefully they're able to solve that problem with whatever results they get from this testing. It's extremely frustrating to be punished for playing well.

MincMincMinc · January 24

React wrote: »

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

One other interesting thing that was mentioned by Brian was that the "bogging down" we see server wide is the result of the global actions being limited by what can fit within the server frames. Brian said that the same thing happens on an individual basis client side when you're performing high APM actions, i.e animation cancelling/weaving/block cancelling/bash weaving, etc.

So that phenomena high APM players have been describing for years where they feel like they need to stop performing actions to let the game "catch up", or where things become unresponsive when they're operating at high APM, is actually the reality of how things work now.

Hopefully they're able to solve that problem with whatever results they get from this testing. It's extremely frustrating to be punished for playing well.

Yeah the first thing that breaks down is the input buffer system. If you press too quickly the game wont keep up with the order. This is what causes the missed cast feeling people get. When it lags you need more and more perfect inputs or else your previous input gets overriden/forgotten. This then gets worse as the server itself lags and you need to essentially guess when you can put in the input. Basically guessing what delay you should put between keystrokes.

xylena_lazarow · January 25

It's fair for players to be skeptical, but I appreciate seeing the devs invest in PvP.

Soraka · January 25

ZOS_Kevin wrote: »

Hi all, Just wanted to follow up with some info. We will have a live stream today at 1pm ET with Gina, Rich, and Brian to chat about Cyrodiil Champions. It will be on the shorter end of things, but solely focused on the test. Info is below.

ZOS_GinaBruno wrote: »

Hi everyone!

We've been seeing a lot of questions about the Cyrodiil Champions test currently on the PTS. To help clarify a few points, we will be hosting a very brief (15-20 minute) livestream today at 1pm EST on twitch.tv/bethesda to answer some common questions including what the goals of this test are, how this current test is different from what will be on Live, what success looks like from our end, and more. This is not taking the place of the written Q&A that is in progress and we do not plan to cover any other topics.

Looking forward to seeing you all in a few hours!

Thank you Kevin and team! I really appreciate the effort I've seen in additional communication about this from everyone. I also appreciate the time taken to see and address some issues.

While there are still some things I worry about, this does go a long way in helping me "trust the process" compared to the past. I've been participating and intend to continue to participate in testing on live, but this helps with the doubts about future direction some so I can focus on the present tests and not feel dread. Additional info like explaining the server things also helped give more context into the how and why (I understand that a lot must remain a mystery for us players, but little stuff like that helps quell the doubts.)

I particularly appreciate hearing you intend to keep the whole spirit of the game for PvP players, feel like PvP is part of the whole game (not some outcasts 😂), and acknowledge that classless playstyle wouldn't go over well. I feel like there is possibly more understanding of what the pvp experience and desires are than I originally thought.

silky_soft · January 25

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

I rewatched it, where do they say the server frametime is 16.667ms(60hz)? Or is this previous information?

From that, anyone on 16ms or less latency they are essentially playing 'live' with the server and good luck anyone else, because the server will prioritise their actions and drop all your useless information. Then your client will revert your character back to the server timeline locking it up unable to skill, (because animation or gcd), while your face is being rolled by these extremely low latency players?

So if that assumption is correct in how it works. Then why not set bots up externally routed via vpn in different locations to get a more broad spectrum of response times to the server? Also why not increase the frame time to 32ms or 64ms?

Then another path opens up why have any instant skills in the first place? Try putting 100ms cast time on every skill that currently instant. Let less information be spammed towards the sever.

Also is this a stat, the server frametime, they would allow to be displayed beside the current fps/ping counters?

I'd like to see them live streaming this event. Even just once for 30mins during a prime time session at thier convenience.

Turtle_Bot · January 25

React wrote: »

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

One other interesting thing that was mentioned by Brian was that the "bogging down" we see server wide is the result of the global actions being limited by what can fit within the server frames. Brian said that the same thing happens on an individual basis client side when you're performing high APM actions, i.e animation cancelling/weaving/block cancelling/bash weaving, etc.

So that phenomena high APM players have been describing for years where they feel like they need to stop performing actions to let the game "catch up", or where things become unresponsive when they're operating at high APM, is actually the reality of how things work now.

Hopefully they're able to solve that problem with whatever results they get from this testing. It's extremely frustrating to be punished for playing well.

Very, very, interesting indeed, all but confirms what the majority of players have thought about the impact of ball groups on the servers over the years.

YandereGirlfriend · January 25

Turtle_Bot wrote: »

React wrote: »

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

One other interesting thing that was mentioned by Brian was that the "bogging down" we see server wide is the result of the global actions being limited by what can fit within the server frames. Brian said that the same thing happens on an individual basis client side when you're performing high APM actions, i.e animation cancelling/weaving/block cancelling/bash weaving, etc.

So that phenomena high APM players have been describing for years where they feel like they need to stop performing actions to let the game "catch up", or where things become unresponsive when they're operating at high APM, is actually the reality of how things work now.

Hopefully they're able to solve that problem with whatever results they get from this testing. It's extremely frustrating to be punished for playing well.

Very, very, interesting indeed, all but confirms what the majority of players have thought about the impact of ball groups on the servers over the years.

Pretty sure that describes the impact of any large concentration of players.

Turtle_Bot · January 25

YandereGirlfriend wrote: »

Turtle_Bot wrote: »

React wrote: »

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

One other interesting thing that was mentioned by Brian was that the "bogging down" we see server wide is the result of the global actions being limited by what can fit within the server frames. Brian said that the same thing happens on an individual basis client side when you're performing high APM actions, i.e animation cancelling/weaving/block cancelling/bash weaving, etc.

So that phenomena high APM players have been describing for years where they feel like they need to stop performing actions to let the game "catch up", or where things become unresponsive when they're operating at high APM, is actually the reality of how things work now.

Hopefully they're able to solve that problem with whatever results they get from this testing. It's extremely frustrating to be punished for playing well.

Very, very, interesting indeed, all but confirms what the majority of players have thought about the impact of ball groups on the servers over the years.

Pretty sure that describes the impact of any large concentration of players.

The difference being the number of players required to reach that concentration of commands/inputs being sent to the servers.

With ball groups it's just the 12 members of the ball group, with unorganized zergs/PuGs it requires the entire faction (and usually all 3 factions stacked like this in 1 area) to have that sort of impact on server performance.

Avran_Sylt · January 25

If you can, set skill loadout presets that make sense for each class, so players in the upcoming March test don't need to think about too critically what skills to slot and get right on into the action (assuming an influx of new non-PvP players ("What's this?" regarding a changed ability slot, vs "How do I position everything to be the best combination?")

I don't know if you have data on typical usage of skill across all PvE/PvP, but that might help with skills and placement on bars.

Give a hefty PvE incentive to play this test mode. Ideally not all at once, neither for consistent play for the duration.

Kaelthorn_Nightbloom · January 25

silky_soft wrote: »

SkaraMinoc wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

sarahthes wrote: »

SkaraMinoc wrote: »

MincMincMinc wrote: »

Comparing vengeance balance or numbers to live to persuade players to hate it before trying is not constructive or helpful in anyway.

Yeah this is flat out wrong. Players are allowed to voice their opinion on the "test" whether you like it or not. Just because you disagree doesn't mean it's not constructive and helpful. In fact it's extremely important that ZOS understands how players feel about the test. So get out of the way and stop pressuring players and putting them down.

The takeaway should be "don't test anything, just fix it without testing?"

I suppose that is feedback, but I strongly hope it's the minority.

That shouldn't be the takeaway.

It seems odd that ZOS can't reproduce the performance issue(s) on a private dev server with bots. It's generally bad practice to do tests like this on a Live production system. Yes I understand they are trying to reproduce the exact controlled environment but that doesn't change the fact that they should be able to do a controlled test internally. Or maybe they have already and didn't find anything 🤔

I wrote an MMO server back in 2010 and the very first thing I did was create 50k bots for testing performance. We also had bots on our Ultima Online servers back in 2004-2006 for testing performance. So I don't know why ZOS can't do that.

They addressed why their bot testing isn't adequate in the livestream today. They do extensive bot testing of all of their content. It provides certain types of data, but it boils down to bots don't act like people, and are supplemental to real people testing (in numbers) but not a replacement.

Oh, I didn't know there was a stream. Ok that's funny 🤣 I'm watching it now.

https://www.twitch.tv/videos/2362374320

Some takeaways.
It's just a test for performance and there no plans to make Vengeance a permanent thing. It's not a test to measure player engagement but they do want as many players as possible to join the test to get more data.

The intention for Cyrodiil is to still be designed for progression and customization. So they want players to have gear, skill choice, CP, etc.

It sounds like they're focusing on the server frame metrics. The server frame was described as a window of time where abilities, movement, etc are processed. At a high level this is the same concept as a server tick/pulse where all the incoming packet messages are dequeued and the combat calculations are done. Usually this is a very small window of time like 16.67 milliseconds (60 tick rate/hz). Then once the calculations are done, outgoing packets are sent back to the clients.

The server frame metrics elevate and/or jitter during large fights and they're trying to get those metrics down with the very distilled Vengeance combat. From there, they will decide what to do next.

They can't get the performance data they need with bots since there are many ways players can engage in PvP and it's not feasible to program all those PvP scenarios. So they need to have real players.

They need players to help test on PTS.

There should no longer be a conspiracy that Cyrodiil combat is changing permanently. They intend for this to be temporary to help address Cyrodiil performance which has been a very longstanding issue and complaint.

One thing that's probably obvious is if players do end up really liking the Vengeance campaign then ZOS might pivot and make it a full feature. They never said that on stream but it's a safe assumption to have. That's my take at least.

I rewatched it, where do they say the server frametime is 16.667ms(60hz)? Or is this previous information?

They didn't cover that in the stream. I don't think ZOS has ever made those details public. 60 tick rate is just an example. It can also be variable and adjust based on server load, or they might use something similar to Valve's sub-tick updates.

Zeraw86 · January 25

I also looked at all the classes on the PTS. The only classes that would be useful to me are the NB or the DK and the Templar; all the other classes aren't really my thing. Nevertheless, I won't be entering Cyrodiil during the test week because it's an absolute disgrace! I'm sure 90% of the community will agree with me!

BlakMarket · January 25

Hate to say it, but this is where a lead developer very familiar with pvp/combat mechanics would shine, they would be able to ensure bare bones server load to ensure testing goals but be able to balance classes to be playable/fun during testing, most people will now just skip the week instead of participating,

We pay for ESO+, then expect us to test your game AGAIN like literal employees testing a barely playable beta, at least make it fun or heavily incentivized for people testing.