Is Mesh network PvP the future of cyrodil

MincMincMinc

As we saw in the livestreams zos is designing a one keep campaign. Where we have three spawn zones surrounding a single keep with a river going around it. There are bridges for bridge fights, and outside the river are resource nodes to contest. We know to solve the pvp lag issues they are simply reducing the player population again to account for the ballooning game code.

The real question is whether zos is going to take this mini keep campaign and go further with it. Theoretically they could distribute the playerbase across multiple mini zones each controlled by their own server to distribute the lag. Imagine if we took cyrodil and made a hex grid across the map. Then every hex was its own server where you could run between and swap to. Thus they could cut down the lag of a server in half or more depending how many keep fights are going on. If they were really smart they would probably disband the 3 faction concept and find ways to prevent massive singularity events where everyone in the campaign winds up at a single keep.

reazea

This would require ZOS to invest in more servers; something they've been exceedingly reluctant to do over the last decade.

Sluggy

MincMincMinc wrote: »

The real question is whether zos is going to take this mini keep campaign and go further with it. Theoretically they could distribute the playerbase across multiple mini zones each controlled by their own server to distribute the lag. Imagine if we took cyrodil and made a hex grid across the map. Then every hex was its own server where you could run between and swap to. Thus they could cut down the lag of a server in half or more depending how many keep fights are going on. If they were really smart they would probably disband the 3 faction concept and find ways to prevent massive singularity events where everyone in the campaign winds up at a single keep.

To an extent this is how pretty much how all games already work. The only difference is that they aren't running different physical servers. Yes this includes ESO. Spacial partitioning is a staple of pretty much every game since DOOM.

I've said it before, I'll say it again. This isn't a hardware issue. Even hardware from two-decades ago could handle this kind of workload. It's either a networking infrastructure issue or it's an software engineering issue. Throwing more machines at the problem without addressing the underlying networking or software architecture will absolutely only make the performance issues worse.

ceruulean

Sluggy wrote: »

I've said it before, I'll say it again. This isn't a hardware issue. Even hardware from two-decades ago could handle this kind of workload. It's either a networking infrastructure issue or it's an software engineering issue. Throwing more machines at the problem without addressing the underlying networking or software architecture will absolutely only make the performance issues worse.

Yep, let's reiterate what ZOS already said...

Two efforts immediately began to account for this - a general pass on gameplay systems focused on optimization in mass PvP environments, and the hardware upgrade...the hardware upgrade provided an immediate measurable result that was easily noticeable and overwhelmingly positive.

While the improvement was drastic, we did notice inconsistent changes over time - during non-peak hours, things remained consistently stable, but at peak hours, performance could shift drastically and frequently, just as it did before the upgrade (albeit in lower magnitude). This was somewhat expected with limited changes to gameplay, and after some investigation the result simply led us back to the secondary efforts noted above...

All of this is to say that the hardware upgrade dramatically improved the general experience on all live realms, and that remains true to this date - but adding something as simple as a new item set can shift behaviors in a way that leads to inconsistencies that we need to account for and address...

It's important to note that all of the above is in the context of Cyrodiil simply because that is where the core issues are easier to identify, but all improvements are shared across the game where applicable.

In summary, ZOS tried the hardware upgrade, new patches and updates start to cause performance issues, they started developing Vengeance.

Source: https://forums.elderscrollsonline.com/en/discussion/comment/8399676/#Comment_8399676

imPDA

It is almost impossible to cut map on hex grid, because when two players from neighbor cells meet, they would need to exchange data between two servers, and it will be thousands or millions of times slower than doing it on one server. Distributed AI between two servers works bad because any interface (internet, USB, Thunderbolt, PCI) is slower than CPU to RAM transfer, and then you have to exchange data between servers, it creates bottleneck. I was thinking about this opportunity, and the best you can do is split to 3 parts, one server for each faction. Borders will be placed where border walls between faction are now (and river instead of walls between AD and EP). That also means you will not be able to fight on border, no more EP/AD bridge fights, etc., because you will have loading screen between servers most likely. Without loading screens, it is pretty hard to implement 3 locations on different servers imo.

This can kinda solve the problem, but not necessary. If all players will come and attack one faction, it will still lag. You could also move faction starting zones to separate server, as there are many chilling or afk players usually here, but it would also remove seamless experience because of loading screen (most likely). I don't know how it works under the hood for ZOS, but with my experience in computer science, I would say it is harder than you think and it is also very limited.

They could optimize skills, adopt loadout-like system for Cyro (you still would have all skills and CP and gear, but you could change armor and skills only on base, and TP from it would lock skills and gear. This way server could just pre-calculate a lot of stats and save as constants instead of calculating everything every tick).

And second option what are they doing right now. They will just remake part of map with keep and let people play it in session manner, 30-40 minutes per siege, probably 2 factions 36-48 people per faction. That was my first predict long ago they announced this. Have they shown more about it tho?

MincMincMinc

Sluggy wrote: »

MincMincMinc wrote: »

The real question is whether zos is going to take this mini keep campaign and go further with it. Theoretically they could distribute the playerbase across multiple mini zones each controlled by their own server to distribute the lag. Imagine if we took cyrodil and made a hex grid across the map. Then every hex was its own server where you could run between and swap to. Thus they could cut down the lag of a server in half or more depending how many keep fights are going on. If they were really smart they would probably disband the 3 faction concept and find ways to prevent massive singularity events where everyone in the campaign winds up at a single keep.

To an extent this is how pretty much how all games already work. The only difference is that they aren't running different physical servers. Yes this includes ESO. Spacial partitioning is a staple of pretty much every game since DOOM.

I've said it before, I'll say it again. This isn't a hardware issue. Even hardware from two-decades ago could handle this kind of workload. It's either a networking infrastructure issue or it's an software engineering issue. Throwing more machines at the problem without addressing the underlying networking or software architecture will absolutely only make the performance issues worse.

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

I am just really disappointed the first test showed this, yet zos is going ahead and reworking skills proven to the an issue without finishing testing first. Crazy considering it seems like they even want to ramp up pve events with things like the wall.

imPDA wrote: »

It is almost impossible to cut map on hex grid..............

And second option what are they doing right now. They will just remake part of map with keep and let people play it in session manner, 30-40 minutes per siege, probably 2 factions 36-48 people per faction. That was my first predict long ago they announced this. Have they shown more about it tho?

Well there is also the Albion approach where each hex grid is its own server, but you need to load into the server as you cross the border. Its not like the old days where ganking is really viable anyways. We hardly see combat exactly in the center of transitus lines anyways. Much like IC you would give immunity until a few seconds after moving on as you cross into the next server.

Yes technically all the players could all go to one keep and bog the server down. Well then this is really up to zos to design the gamemode in a way to disincentivize this. LIKE MAYBE NOT HAVING A VOLENDRUNG??? Im still amazed thats in the game. (The game lags too much when players all go to one spot.......OH I have an idea, lets make an item that causes everyone to group up in one spot!)

Rohamad_Ali

First, What is Mesh network PvP?

Sluggy

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

MincMincMinc wrote: »

I am just really disappointed the first test showed this, yet zos is going ahead and reworking skills proven to the an issue without finishing testing first. Crazy considering it seems like they even want to ramp up pve events with things like the wall.

I'm right there with ya. I'm pretty sure both of us were big proponents of the potential we saw.I wanted to see more tests, rapid-fire, pushing the limits as features were brought back. I'm certain we could have at least restored most of the sets that didn't have ridiculous proc conditions and loads of data to track and certainly most, if not all, of the passives. But the community didn't want that. I'm disappointed not just in ZoS but in the PvP community as well. As much as I've heard cries for separation of PvP mechs from PvE mechs, they sure didn't respond well when they tried it. People say they'd jump ship at the first viable alternative game. But as soon as that option shows up, they get cold feet. Some of it was of course skepticism in ZoS's ability to follow through. And I had a healthy amount of that myself. But I was willing to at least let them try because as far as I'm concerned, it's a dead-gamemode-walking otherwise.

MincMincMinc

Sluggy wrote: »

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

No i mean that hardware change itself had no real impact. The only notable performance benefits since 2016 have been them reducing the cyrodil population over time. Which has gotten us from the 2014-2016 era 900 players down to the current 300 players. Ive argued with countless people on this topic and not only zos denied it at the time, but even those people who argued hardware changes had their own statements back then saying it helped, then a day later saying it was lagging the same. Even some streamers at the time did the same thing where they ate the placebo and a day or two later said there was the same issues again............What I assume happened was during the shutdown alot of normal coordinated groups didnt play for that following day or two. So much like MyM the player population was diluted with casual or solo players. Then a day or two later got diluted again with coordinated spam groups.

Its alarming how clear it is when the server can lag from ballgroups yet zos doesn't break it up. Zos did say they can look at the logs and see whats occurring during the spikes, so you'd think they would see mostly the ballgroup spam or a list of player names being hit all at once. I still wouldnt be surprised if it has to do with smartheals or similar AoE check systems that have always had issues......of which groups are big fans of spamming.

Otherwise, yeah it is rather disappointing how many PvP players had rather watch the ship sink instead of taking over vengeance threads and leading zos in a better direction. If people had just spam asked for basic item sets back and some basic nonproc passives we could have seen a whole different test path happen. Instead it was rejected and zos is replacing GH with this mini campaign, then veng will inevitably be the only cyrodil.

ToddIngram

MincMincMinc wrote: »

Sluggy wrote: »

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

No i mean that hardware change itself had no real impact. The only notable performance benefits since 2016 have been them reducing the cyrodil population over time. Which has gotten us from the 2014-2016 era 900 players down to the current 300 players. Ive argued with countless people on this topic and not only zos denied it at the time, but even those people who argued hardware changes had their own statements back then saying it helped, then a day later saying it was lagging the same. Even some streamers at the time did the same thing where they ate the placebo and a day or two later said there was the same issues again............What I assume happened was during the shutdown alot of normal coordinated groups didnt play for that following day or two. So much like MyM the player population was diluted with casual or solo players. Then a day or two later got diluted again with coordinated spam groups.

Its alarming how clear it is when the server can lag from ballgroups yet zos doesn't break it up. Zos did say they can look at the logs and see whats occurring during the spikes, so you'd think they would see mostly the ballgroup spam or a list of player names being hit all at once. I still wouldnt be surprised if it has to do with smartheals or similar AoE check systems that have always had issues......of which groups are big fans of spamming.

Otherwise, yeah it is rather disappointing how many PvP players had rather watch the ship sink instead of taking over vengeance threads and leading zos in a better direction. If people had just spam asked for basic item sets back and some basic nonproc passives we could have seen a whole different test path happen. Instead it was rejected and zos is replacing GH with this mini campaign, then veng will inevitably be the only cyrodil.

Except the new hardware made a huge and positive difference until they added a bunch of new content and changed how the servers were allocated within the game.

ZOS also said they've been working to improve Cyrodiil since 2015 and even banned people who complained too much when this effort was not recognized by the players. And that vengeance is a test to determine what they needed to do to restore performance to live cyrodiil. How well does ZOS' current statements align with their past statements?

MincMincMinc

ToddIngram wrote: »

MincMincMinc wrote: »

Sluggy wrote: »

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

No i mean that hardware change itself had no real impact. The only notable performance benefits since 2016 have been them reducing the cyrodil population over time. Which has gotten us from the 2014-2016 era 900 players down to the current 300 players. Ive argued with countless people on this topic and not only zos denied it at the time, but even those people who argued hardware changes had their own statements back then saying it helped, then a day later saying it was lagging the same. Even some streamers at the time did the same thing where they ate the placebo and a day or two later said there was the same issues again............What I assume happened was during the shutdown alot of normal coordinated groups didnt play for that following day or two. So much like MyM the player population was diluted with casual or solo players. Then a day or two later got diluted again with coordinated spam groups.

Its alarming how clear it is when the server can lag from ballgroups yet zos doesn't break it up. Zos did say they can look at the logs and see whats occurring during the spikes, so you'd think they would see mostly the ballgroup spam or a list of player names being hit all at once. I still wouldnt be surprised if it has to do with smartheals or similar AoE check systems that have always had issues......of which groups are big fans of spamming.

Otherwise, yeah it is rather disappointing how many PvP players had rather watch the ship sink instead of taking over vengeance threads and leading zos in a better direction. If people had just spam asked for basic item sets back and some basic nonproc passives we could have seen a whole different test path happen. Instead it was rejected and zos is replacing GH with this mini campaign, then veng will inevitably be the only cyrodil.

Except the new hardware made a huge and positive difference until they added a bunch of new content and changed how the servers were allocated within the game.

ZOS also said they've been working to improve Cyrodiil since 2015 and even banned people who complained too much when this effort was not recognized by the players. And that vengeance is a test to determine what they needed to do to restore performance to live cyrodiil. How well does ZOS' current statements align with their past statements?

I don't want to rehash again and continue off topic, but again the hardware change literally was lagging the exact same a day or two later when people started regularly playing again after servers were down. You can go back to those times on the forums and there are loads of people that have whiplash posts where "performance is so much better! Woah OMG!" .......then a day later "performance is back to lagging exactly what it was before".

Anyways, I really have to hope zos picks up on this mesh concept. Otherwise I kinda just expect these mini gamemodes to turn into one ballgroup bullying the server/match much like how BGs always turned out when all ques were group ques. Probably still lagging the server.

TheAwesomeChimpanzee

MincMincMinc wrote: »

Sluggy wrote: »

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

No i mean that hardware change itself had no real impact. The only notable performance benefits since 2016 have been them reducing the cyrodil population over time. Which has gotten us from the 2014-2016 era 900 players down to the current 300 players. Ive argued with countless people on this topic and not only zos denied it at the time, but even those people who argued hardware changes had their own statements back then saying it helped, then a day later saying it was lagging the same. Even some streamers at the time did the same thing where they ate the placebo and a day or two later said there was the same issues again............What I assume happened was during the shutdown alot of normal coordinated groups didnt play for that following day or two. So much like MyM the player population was diluted with casual or solo players. Then a day or two later got diluted again with coordinated spam groups.

Its alarming how clear it is when the server can lag from ballgroups yet zos doesn't break it up. Zos did say they can look at the logs and see whats occurring during the spikes, so you'd think they would see mostly the ballgroup spam or a list of player names being hit all at once. I still wouldnt be surprised if it has to do with smartheals or similar AoE check systems that have always had issues......of which groups are big fans of spamming.

Otherwise, yeah it is rather disappointing how many PvP players had rather watch the ship sink instead of taking over vengeance threads and leading zos in a better direction. If people had just spam asked for basic item sets back and some basic nonproc passives we could have seen a whole different test path happen. Instead it was rejected and zos is replacing GH with this mini campaign, then veng will inevitably be the only cyrodil.

You’re just wrong on the hardware point. That server upgrade very clearly improved Cyrodiil performance for a long while, and it is still the only thing ZOS has ever done that produced a genuinely noticeable improvement to lag. Even when the server lags now, it still usually does not hit the same level of complete dysfunction it used to before that change, and that alone shows it had significant impact.

That said, over time the lag has absolutely been creeping back. Year by year it has gotten steadily worse, to the point where we are starting to see conditions that feel a lot more like the old days again. A lot of that is very likely due to ability complexity, and possibly some degree of server degradation over time as well. Vengeance, for all the problems I have with it, has made one thing pretty obvious: simplifying skills does help performance.

But saying the hardware upgrade itself had no meaningful impact is beyond wrong. It did. The bigger issue is that ZOS does not seem willing to invest in the only thing that has ever actually produced a clear improvement to Cyrodiil lag.

TheAwesomeChimpanzee

MincMincMinc wrote: »

ToddIngram wrote: »

MincMincMinc wrote: »

Sluggy wrote: »

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

No i mean that hardware change itself had no real impact. The only notable performance benefits since 2016 have been them reducing the cyrodil population over time. Which has gotten us from the 2014-2016 era 900 players down to the current 300 players. Ive argued with countless people on this topic and not only zos denied it at the time, but even those people who argued hardware changes had their own statements back then saying it helped, then a day later saying it was lagging the same. Even some streamers at the time did the same thing where they ate the placebo and a day or two later said there was the same issues again............What I assume happened was during the shutdown alot of normal coordinated groups didnt play for that following day or two. So much like MyM the player population was diluted with casual or solo players. Then a day or two later got diluted again with coordinated spam groups.

Its alarming how clear it is when the server can lag from ballgroups yet zos doesn't break it up. Zos did say they can look at the logs and see whats occurring during the spikes, so you'd think they would see mostly the ballgroup spam or a list of player names being hit all at once. I still wouldnt be surprised if it has to do with smartheals or similar AoE check systems that have always had issues......of which groups are big fans of spamming.

Otherwise, yeah it is rather disappointing how many PvP players had rather watch the ship sink instead of taking over vengeance threads and leading zos in a better direction. If people had just spam asked for basic item sets back and some basic nonproc passives we could have seen a whole different test path happen. Instead it was rejected and zos is replacing GH with this mini campaign, then veng will inevitably be the only cyrodil.

Except the new hardware made a huge and positive difference until they added a bunch of new content and changed how the servers were allocated within the game.

ZOS also said they've been working to improve Cyrodiil since 2015 and even banned people who complained too much when this effort was not recognized by the players. And that vengeance is a test to determine what they needed to do to restore performance to live cyrodiil. How well does ZOS' current statements align with their past statements?

I don't want to rehash again and continue off topic, but again the hardware change literally was lagging the exact same a day or two later when people started regularly playing again after servers were down. You can go back to those times on the forums and there are loads of people that have whiplash posts where "performance is so much better! Woah OMG!" .......then a day later "performance is back to lagging exactly what it was before".

Anyways, I really have to hope zos picks up on this mesh concept. Otherwise I kinda just expect these mini gamemodes to turn into one ballgroup bullying the server/match much like how BGs always turned out when all ques were group ques. Probably still lagging the server.

I will say, though, that even if we disagree on the server, your love for Vengeance, and a lot of other things, the mesh concept actually sounds like a great solution and one that could potentially save the version of PvP most veteran players want to play.

MincMincMinc

TheAwesomeChimpanzee wrote: »

MincMincMinc wrote: »

ToddIngram wrote: »

MincMincMinc wrote: »

Sluggy wrote: »

MincMincMinc wrote: »

There is also a post a while back where Matt I think said that they were doing the hardware upgrade for maintenance and they do not believe it would have any significant increase in performance. Yet people ate the placebo and were swearing that there was any difference........... When it comes down to it. The game works during vengeance, yet it doesnt when you re-enable all of the bloated skills, sets, and effects.

These days everything seems so hyperbolic so I actually can't tell if that's sarcasm or not. But...

It wasn't a placebo. Despite the glaring issues that still occur when ballgroups show up in mass, the server still runs decidedly better than I'd seen it run since, like, 2016. Gravity itself used to stop working, people would freeze in place, and going five or so minutes without skills firing was the norm even in some medium-sized fights. Now at least in these exceptional cases it's only like, five or so seconds - which, don't get me wrong, is still unreasonably bad. But it I orders of magnitude better than before.

No i mean that hardware change itself had no real impact. The only notable performance benefits since 2016 have been them reducing the cyrodil population over time. Which has gotten us from the 2014-2016 era 900 players down to the current 300 players. Ive argued with countless people on this topic and not only zos denied it at the time, but even those people who argued hardware changes had their own statements back then saying it helped, then a day later saying it was lagging the same. Even some streamers at the time did the same thing where they ate the placebo and a day or two later said there was the same issues again............What I assume happened was during the shutdown alot of normal coordinated groups didnt play for that following day or two. So much like MyM the player population was diluted with casual or solo players. Then a day or two later got diluted again with coordinated spam groups.

Its alarming how clear it is when the server can lag from ballgroups yet zos doesn't break it up. Zos did say they can look at the logs and see whats occurring during the spikes, so you'd think they would see mostly the ballgroup spam or a list of player names being hit all at once. I still wouldnt be surprised if it has to do with smartheals or similar AoE check systems that have always had issues......of which groups are big fans of spamming.

Otherwise, yeah it is rather disappointing how many PvP players had rather watch the ship sink instead of taking over vengeance threads and leading zos in a better direction. If people had just spam asked for basic item sets back and some basic nonproc passives we could have seen a whole different test path happen. Instead it was rejected and zos is replacing GH with this mini campaign, then veng will inevitably be the only cyrodil.

Except the new hardware made a huge and positive difference until they added a bunch of new content and changed how the servers were allocated within the game.

ZOS also said they've been working to improve Cyrodiil since 2015 and even banned people who complained too much when this effort was not recognized by the players. And that vengeance is a test to determine what they needed to do to restore performance to live cyrodiil. How well does ZOS' current statements align with their past statements?

I don't want to rehash again and continue off topic, but again the hardware change literally was lagging the exact same a day or two later when people started regularly playing again after servers were down. You can go back to those times on the forums and there are loads of people that have whiplash posts where "performance is so much better! Woah OMG!" .......then a day later "performance is back to lagging exactly what it was before".

Anyways, I really have to hope zos picks up on this mesh concept. Otherwise I kinda just expect these mini gamemodes to turn into one ballgroup bullying the server/match much like how BGs always turned out when all ques were group ques. Probably still lagging the server.

I will say, though, that even if we disagree on the server, your love for Vengeance, and a lot of other things, the mesh concept actually sounds like a great solution and one that could potentially save the version of PvP most veteran players want to play.

Thanks

Well i dont necessarily like vengeance or think it should stay in its current state. I like testing and think its dumb people boycotted it instead of participating to have a voice in future pvp matters. Leaving most of the veng feedback up to pve players who probably wont play longterm.

I am more for trimming the fat off of effects and cleaning up the code. You know, over the years nobody pumped the brakes on how out of control the design team could make proc sets and new effects. I dont think its healthy to have 5 paragraph long proc sets coming out every dlc release. Nor should we have skills like deep fissure which are so absurdly complicated it has a 2 paragraph tooltip and interacts with several passives. Same goes for things like streak technically doing a ton of unnecessary aoe effects. No reason not to just reduce it to the blink+an aoe stun at the final location. You could probably remove the damage entirely since its negligible for kills and just exists to cause more server load. (I'm a 12 year stamsorc pvp player)

Sadly we are seeing zos go through doing the class skill reworks before even finishing collecting possible vengeance data to know if pvp was fixable to any degree. Which means they will cut the PvP pop again to account for the lag >> thus we are seeing the mini keep seige map. We know veng1+2 was successful at 900 cap.....We know GHpvp at 300 is unbearable alot of the time. What happens halfway between? Say you brought back nonproc sets and flat passives can we do 600 players? 700? 750? If you said we cut the amount of proc condition effects down and get 600 players in a campaign I would sign right up. Keep in mind they developed the ability to have pve and pvp split skills so if flushed out we could have brought back alot of normal elements to eso building while raising the server pop.

Lagzee

MincMincMinc wrote: »

As we saw in the livestreams zos is designing a one keep campaign. Where we have three spawn zones surrounding a single keep with a river going around it. There are bridges for bridge fights, and outside the river are resource nodes to contest. We know to solve the pvp lag issues they are simply reducing the player population again to account for the ballooning game code.

The real question is whether zos is going to take this mini keep campaign and go further with it. Theoretically they could distribute the playerbase across multiple mini zones each controlled by their own server to distribute the lag. Imagine if we took cyrodil and made a hex grid across the map. Then every hex was its own server where you could run between and swap to. Thus they could cut down the lag of a server in half or more depending how many keep fights are going on. If they were really smart they would probably disband the 3 faction concept and find ways to prevent massive singularity events where everyone in the campaign winds up at a single keep.

Ive been wondering about this for a while. But not on such a large scale. Like they said they cant give us large scale pvp with minimal lag and no changes to what we have access to. But i dont expect large scale combat with minimal lag. I just want to be able to have a 5 player fight at allessia and not be impacted by the massive zerg fight at chalmen. And i dont see why they cant give us that at least. Some form of lag during massive fights is expected, at least if you have any experience in large scale fights in any mmo. Maybe not always esos notorious, and abysmal, ability delay. But extreme FPS drops can be bad as well.

But i was thinking maybe they could split cyro into parts, with like a load screen in separating both halves. But thats just a thought, regardless i feel like they should be able to, at least, give us better performance away from massive fights in cyrodiil.