Best Of
Re: The new and last Survey about Vengeance is out
ShutUpitsRed wrote: »I wonder what determines who gets sent a survey vs who doesn't.
We sent the survey to those who participated in Vengeance during the Vengeance timeframe. This also includes if you participated in Vengeance and Gray Host when they were both running. The goal of the survey is to gather feedback from those who either participated in Vengeance or both Vengeance and Gray Host during the test period.robwolf666 wrote: »Anyway, I'm sure if ZoS is able to track who took part to send out the survey, they also know who those links were sent to, and which replies are legitimate. (When I go there now it says I completed the survey, so it obviously tracks it somehow).
We do. We've noted in the past that sharing the links skews data from users who have played and we would like to avoid that where possible. Since sharing links has happened a few times now with past surveys, we have taken additional steps to make sure the data does not get skewed and matches with those who are participated in the in-game tests. So if you did not get the link for some reason but participated in the tests, using the shared link will still allow your data will get recorded and matched.
Excluding those who rejected Vengeance but continued to play in Grey Host also skews your data towards those who view Vengeance positively. Kinda invalidating for the survey ngl.
Re: Cyrodiil & Imperial City – Community Q&A
YandereGirlfriend wrote: »Q. Do you have understanding, knowledge and/or insight in what addons that are "private" / "non-official" and used in PVP? Do and do you have plans to monitor or limit the behavior of such addons? - Idinuse
A. We’d like to provide a bit on our philosophy before answering this question. From a technical sense, in our perspective, there is no such thing as a private or unofficial mod. All mods are technically unofficial, since we don’t have a verification process.
Now in the sense that a “private” add-on is something someone made and didn’t post it for wider distribution, we do have a way to keep track of those by user. However, that does not give a clear indicator of what every add-on does. We go by reports and behaviors. So, if we see behavior that shouldn't be possible, or a user reports that an addon-on someone built is suspected of allowing for actions or behaviors that violate our Terms of Service, we investigate, close any API loopholes we find, and ban anyone caught exploiting. To be fair, the add-on community is rather good at self-managing instances like this. They would rather settle it among themselves without our intervention, as that could lead to everyone unintentionally losing more tools in their creative toolkit.
If you see instances where someone is exploiting through add-on usage, please report it to us so we can properly investigate and reduce the harm done.
With respect, this policy is SO backward.
This overly permissive stance toward the use of API features allows bad actors to run roughshod over the restrictions that are commonly believed to be in place (but in reality ARE NOT) and the burden is somehow on the rest of the community to out them and bring them to justice? Do you realize how unreasonable of a standard that is to set? To successfully bust a private and nefarious add-on you are forcing players to literally infiltrate those groups of users in order to expose them. Or to rely upon the sloppiness of those users to out themselves. Both of which are tasks that should fall upon ZOS not players.
Further, for the cost of add-on performance, particularly on the game conditions of NON-USERS, it is IMO outrageous that there is seemingly no effort being made to track the impact of add-ons that are being run in the wild. Recall that it took a random forum poster to blow the whistle on Bandits and the Map Pins hack that was degrading the performance of users and non-users alike. It wasn't an add-on dev or ZOS technical staff. It was a motivated player with Wireshark. That should never happen!
Make these add-on devs prove that their mods are behaving as they claim. Make them prove that they are only impacting the user of the mod. This burden should always fall upon add-on devs and their users and never on non-users.
It was alluded to that ZOS was monitoring the server overhead of the new data-sharing library. May we have a report on those findings? Is it being actively monitored or is it simply able to be monitored but nobody is really checking-in on it? IMO, it feels bad to know that common server resources are being siphoned away to transact the essentially private requests of add-on users. No matter how large or small those requests may be. It is unofficial and unsupported software, after all.
Finally, may we finally have a TEST of removing add-ons (and at the very least data-sharing add-ons) from Cyrodiil? It boggles the mind that, among all of the other iteration in Vengeance, this was never even attempted. Even if you think you know the answer it's worth doing anyway to gather actual hard data under real-world conditions. Just as with Map Pins... the conventional wisdom was that the hack was harmless... until it was actually scrutinized and that turned out emphatically not to be the case. This could well turn out to be the case again.
Very well said.
I've always been tremendously suspicious a certain small group on PC NA is using some sort of AddOn or 3rd party interactions, as they are the only one of their kind (that I've ever run into) that generates the same sort of lag that the most impactful ball groups do.
I rarely if ever see lag in Cyrodiil, not even in the biggest zerg on zerg Faction Stack fights, unless multiple or specific individual Ball Groups are present. That alone has always made me wonder just why certain ones are so much more damaging to server performance (seriously, after so much gameplay, I can identify when specific ones arrive before I even see them just by the type of lag, because I don't see it otherwise.) To me, the possibility that AddOns and their interaction are playing a role seems VERY probable, but I admit I'm far from expert.
The fact a particular small group that's less than half ball group size manages to do the same (and I've seen Zone Chat wide complaints to know I'm not alone in noticing) kinda screams 3rd party something to me, because they certainly couldn't be generating the same amount of ability calculations as a much larger Ball Group, let alone multiple, in a similar area, right?
It's supposition on my part, certainly, but what's been stated here doesn't exactly fill me with certainty that it's being carefully monitored either.
Any chance we can get more on that topic, @ZOS_Kevin ? I know people love their AddOns, but I'd MUCH rather go without them in Cyrodiil entirely if they're not being watched with tremendous and constant scrutiny by you guys for their impact on performance.
Re: Cyrodiil & Imperial City – Community Q&A
Q. Have you considered that not the Skills, but other things lead to the poor performance? For example certain server side updates, which lead to big performance regressions. Even if you run around alone in Cyrodiil, running can be slowed down, or riding on your mount can be sluggish and get slow downs, without ANYONE being near. - RedJohn_COF
A. Yes, we've considered and tested what contributes to game performance impact and have looked at the various things including physics processes (the things that control destructible states on keeps and such). We’ve looked into collections, outfit stations, quests, inventory, etc. Everything pales in comparison to ability usage combined with passives (this includes item sets, CP, and skill line passives), which is why they were the focus of our Vengeance tests.
@ZOS_Kevin Vengeance loadouts can only be changed at base because they then become locked, constant for this particular player. But in all other parts of the game, each passive is used in calculations every time (every cast, tick, etc., depends on the nature of passive), right?
Take "Dark Veil" as an example. It adds 2 seconds to Shadow abilities, and if I understand correctly, it applies a passive (36552) that is used to calculate the duration each time a Shadow ability casted. Is that right? If so, it's good design for a flexible environment where everything can be changed on the fly and still calculates correctly, but it adds significant overhead when unnecessary. I wonder if this has been discussed internally, because I see big potential for optimization here.
Possibility one: It could be possible to "snapshot" these values when a player leaves a faction base - with their bars and gear locked from changes (similar to Vengeance). By snapshotting, I mean making them constant for that player, so these values would no longer be recalculated on the fly each cast (or tick) but taken pre-calculated. The same for gear: take stats from gear, add them to base stats, and store them server-side when the player leaves base, and restrict gear changes. Considering these elements won’t be changed during combat anyway, it’s safe to make them constant server-side and remove part of the calculations. This "snapshotting" could safely apply to 90% of passives, skills, and gear. The only limitation is that to change your build, you’d have to return to base. That’s an okay trade-off if it helps improve performance imo.
Possibility two: Create separate skills for PvP, as was done for Vengeance, but take default skills and recalculate base values including all constants. For example, Battle Spirit decreases damage by 50% - make a copy of every available skill with 50% less damage (do the same for healing and shield) - and you can remove Battle Spirit from calculations. Next, require players to fully level skill lines before using them in PvP. That way, you could bake all skill line passives into the skill itself. Using the "Dark Veil" example, every skill would simply be 2 seconds longer by default, not calculated. And again, this applies to almost all passives. Allow players to join Cyrodiil only after level 50 with all passives purchased - then add class passives to base stats (regeneration, stamina, magicka, etc.). This would change nothing for 95% of players (who have everything leveled anyway), while reducing calculation overhead. Only 5% of players might be affected (they would have to finish leveling skills up before entering PvP), but the performance gain could be substantial.
I hope it can be seen by the correct person, who can say if these ideas were discussed previously, if they are viable, if they can be tested, etc. It can turn out to be undoable with current codebase, so while it would help with performance, it is simple can't be done in observable period of time, or it can turn out that these approaches are doable but simply was not considered before. Is it worth to write ideas like this at all?
A. Yes, we've considered and tested what contributes to game performance impact and have looked at the various things including physics processes (the things that control destructible states on keeps and such). We’ve looked into collections, outfit stations, quests, inventory, etc. Everything pales in comparison to ability usage combined with passives (this includes item sets, CP, and skill line passives), which is why they were the focus of our Vengeance tests.
@ZOS_Kevin Vengeance loadouts can only be changed at base because they then become locked, constant for this particular player. But in all other parts of the game, each passive is used in calculations every time (every cast, tick, etc., depends on the nature of passive), right?
Take "Dark Veil" as an example. It adds 2 seconds to Shadow abilities, and if I understand correctly, it applies a passive (36552) that is used to calculate the duration each time a Shadow ability casted. Is that right? If so, it's good design for a flexible environment where everything can be changed on the fly and still calculates correctly, but it adds significant overhead when unnecessary. I wonder if this has been discussed internally, because I see big potential for optimization here.
Possibility one: It could be possible to "snapshot" these values when a player leaves a faction base - with their bars and gear locked from changes (similar to Vengeance). By snapshotting, I mean making them constant for that player, so these values would no longer be recalculated on the fly each cast (or tick) but taken pre-calculated. The same for gear: take stats from gear, add them to base stats, and store them server-side when the player leaves base, and restrict gear changes. Considering these elements won’t be changed during combat anyway, it’s safe to make them constant server-side and remove part of the calculations. This "snapshotting" could safely apply to 90% of passives, skills, and gear. The only limitation is that to change your build, you’d have to return to base. That’s an okay trade-off if it helps improve performance imo.
Possibility two: Create separate skills for PvP, as was done for Vengeance, but take default skills and recalculate base values including all constants. For example, Battle Spirit decreases damage by 50% - make a copy of every available skill with 50% less damage (do the same for healing and shield) - and you can remove Battle Spirit from calculations. Next, require players to fully level skill lines before using them in PvP. That way, you could bake all skill line passives into the skill itself. Using the "Dark Veil" example, every skill would simply be 2 seconds longer by default, not calculated. And again, this applies to almost all passives. Allow players to join Cyrodiil only after level 50 with all passives purchased - then add class passives to base stats (regeneration, stamina, magicka, etc.). This would change nothing for 95% of players (who have everything leveled anyway), while reducing calculation overhead. Only 5% of players might be affected (they would have to finish leveling skills up before entering PvP), but the performance gain could be substantial.
I hope it can be seen by the correct person, who can say if these ideas were discussed previously, if they are viable, if they can be tested, etc. It can turn out to be undoable with current codebase, so while it would help with performance, it is simple can't be done in observable period of time, or it can turn out that these approaches are doable but simply was not considered before. Is it worth to write ideas like this at all?
imPDA
3
Re: What has caused you to quit the game or barely play any more? What would bring you back?
@Veinblood1965 Thank you for starting this thread and constructively expressing your frustration. And thanks to those who have shared their opinion. We are taking a look through this and sharing this with some folks internally for their visibility. Just wanted to note that this thread has been seen.
Thanks ZOS_Kevin, I'm glad there's been a lot of helpful replies instead of a lot of bashing which was what I was hoping for. I think the reply that resonated with me was someone mentioned releases like Summerset were when the game was at it highest peak. That's when I started and Wrothgar was also impressive.
Skill style for gim focus does not change the colour of thw red glow permanently affixed to our weap
Feels like a rip off to be honest. Why bother changing the skill at all if the main pain point of it doesn't change?
No Snow Bear mount?
I was hoping we'd get it around the christmas holidays. I missed it the last time it was around. I don't want to wait a whole *** year again.
pls ZOS. I'll even pay double the crowns
pls ZOS. I'll even pay double the crowns
Re: Community Update – Vengeance Testing & Cyrodiil
tomofhyrule wrote: »spartaxoxo wrote: »spartaxoxo wrote: »spartaxoxo wrote: »MorallyBipolar wrote: »Anomander72 wrote: »I think it is safe to say based on prime time last night on NA server that GH blew Vengeance out in population by a wide margin. Locked, Locked, Locked to 1, 1,1 in Vengeance. Case seems closed to me. Vengeance will be another Ravenwatch
Last three nights, including Friday night, prime time PC NA has seen all three factions pop locked for Grey Host, and only one bar pop for vengeance.
If ZOS ignores this undeniable side by side popularity test it will be the same as telling us they want us to leave the game.
They've already explained that the pop bars are different because the population sizes are different.Hi all, thanks for the continued discussion here. We want to share a point of consideration as we are seeing some comments around population when talking about the in-game graphs. The in-game population bar is representative of the current participants in a campaign, relative to the max cap of that campaign. So for example, if Gray Host is at 360/360, but Vengeance is 450/900, the graph will show Gray Host as 100% capacity while Vengeance is at 50%, even though Vengeance has more players. We wanted to provide that as you continue your conversations about population overall.
spartaxoxo There is no "zero bar" population. And poplocked Grayhost (4 bars) is still more than one bar of Vengeance, assuming that the one bar is legit and not due to the lack of zero bars. Grayhost is clearly more popular, especially when you consider that people waiting to get into poplocked Grayhost could only play Vengeance in the meantime, which would make the difference appear in favor of Vengeance by up to a factor of 2, and even then Vengeance ended up with a lower population. If you then also consider that Vengeance, as far as I'm aware, still has double AP and people on average are still choosing to play Grayhost over Vengeance, it is really telling which mode is more popular among PvPers - which is ultimately what really matters, since we also aren't designing dungeon hardmodes and trials around what PvPers think either.
When I went into Vengeance it was pretty active. I'm sure ZOS has the numbers.
Just take the Vengeance “L” and move on from this stale argument please
I mean if Vengeance caps at 900 people then 33% is 297. So 1 bar in Vengeance is a bigger amount of people than 1 bar in Grey Host. 297 people is equivalent to round about 80% of Grey Host, if I'm understanding what they said about pop caps correctly.
But the thing people are pointing out is that 1 bar Vengeance could be around 80% of GH. But we don't know if it is.
Vengeance will show 1 bar if it's got 80% of a full GH campaign, or 10% of a full GH campaign. Either way, that first in the "0-297" population.
This was in response to someone seeming to claim that Vengeance was not active during the time period when I personally went inside and tried it for myself. The whole 0% thing cannot be used to discount people's personal experience with the mode just because when someone else checked there was often 1 bar. The bars don't mean the same number of people. Part of determining if it was viable is if people participated. I don't have the ability to know the full numbers. I can only relay my personal experience. And my personal experience was that at the time I joined in, it was active.
Re: Crown Refreshing Drink... Why?
I've never seen much use for them as they are mostly identical to existing Distillate recipes. Their major advantages are that they scale to level so there is a very small benefit for characters who are between Distallate levels, plus no crafting skill required; Distillates need maxed Quality Improvement. I'd have to go look, but I think I have one stack of them in chest somewhere for those days when there is a "consume x beverages" endeavor and I don't feel like visiting a cooking fire, but I usually handle those endeavors by using leftovers from daily Provisioning writs.
Re: The new and last Survey about Vengeance is out
The best you can answer is to call to keep GH and ask that they keep adding to vengeance to bring back sets, enchants, mundus, morphs, etc. The perk and loadout system should have just been a fake gear system.....UI can be made to look like anything.
For GH to get fixed we need vengeance in a more comparable state with all the necessary build systems cleaned up for less effect calculations.
Atleast get vengeance closer to a noproc situation. No proc sets, no proc passives, no weapon enchants, no poisons, no status effects. Avoid any possible calcs that can trickle off and bloat the game. Then we need game rule tests like no cross healing, or dot/hot stacking, or aoe cap rules.
For GH to get fixed we need vengeance in a more comparable state with all the necessary build systems cleaned up for less effect calculations.
Atleast get vengeance closer to a noproc situation. No proc sets, no proc passives, no weapon enchants, no poisons, no status effects. Avoid any possible calcs that can trickle off and bloat the game. Then we need game rule tests like no cross healing, or dot/hot stacking, or aoe cap rules.
Re: Honestly - Is Vengeance Viable?
Definitely also want to chime-in that "One does not simply sit in siege..." as a ballgroup.
The debuff from oils is outlandish and it stacks with the number of oils active on a player. As such, it is not uncommon to see a powerful burst heal like Blessing of Restoration CRIT for like 800HP on a target that is sitting on a ram. Vigors, Radiating, etc. are ticking for like 100 per tick, which is basically nonexistent. You add a skilled enemy solo-bomber or a Negate on top of that and it is very easy to die. Anyone saying otherwise is peddling an uninformed take.
It IS very much true though that some factions siege better and some do it worse. This will directly affect your perception as to how "immortal" an enemy ballgroup feels. On PC-NA, for example, DC are absolutely tragic siegers and basically ignore it completely even when an enemy ballgroup shows up. DC, for the most part, also have never heard of this skill named "Negate". That makes them super easy to farm, as they are essentially uncoordinated solo players walking haplessly into obvious bombs.
Contrast that with AD who have quite effective PUG-wranglers that teach them the extremely high value of siege and of fielding Negates. Anyone who ballgroups against AD knows that it is the most difficult faction to fight because you face a huge number of Negates and that siege is the first thing to be deployed as soon as a call-out is made in their zone chat. It is very annoying but it is also very smart because it is BY FAR the best-in-slot tactic to use against ballgroups.
I do think that there is an odd sort of mentality where folk think that they are going to 1vX or bow-gank an enemy ballgroup into submission and then they tilted when that turns out not to be possible. You are simply using the wrong tool for the job. Instead go buy some oils, meatbags, and scattershots and use that at every opportunity. Then subclass Dark Magic and slot Negate and cast it when the ballgroup is under pressure from other siegers. Then tell your friends in zone to do the same.
The debuff from oils is outlandish and it stacks with the number of oils active on a player. As such, it is not uncommon to see a powerful burst heal like Blessing of Restoration CRIT for like 800HP on a target that is sitting on a ram. Vigors, Radiating, etc. are ticking for like 100 per tick, which is basically nonexistent. You add a skilled enemy solo-bomber or a Negate on top of that and it is very easy to die. Anyone saying otherwise is peddling an uninformed take.
It IS very much true though that some factions siege better and some do it worse. This will directly affect your perception as to how "immortal" an enemy ballgroup feels. On PC-NA, for example, DC are absolutely tragic siegers and basically ignore it completely even when an enemy ballgroup shows up. DC, for the most part, also have never heard of this skill named "Negate". That makes them super easy to farm, as they are essentially uncoordinated solo players walking haplessly into obvious bombs.
Contrast that with AD who have quite effective PUG-wranglers that teach them the extremely high value of siege and of fielding Negates. Anyone who ballgroups against AD knows that it is the most difficult faction to fight because you face a huge number of Negates and that siege is the first thing to be deployed as soon as a call-out is made in their zone chat. It is very annoying but it is also very smart because it is BY FAR the best-in-slot tactic to use against ballgroups.
I do think that there is an odd sort of mentality where folk think that they are going to 1vX or bow-gank an enemy ballgroup into submission and then they tilted when that turns out not to be possible. You are simply using the wrong tool for the job. Instead go buy some oils, meatbags, and scattershots and use that at every opportunity. Then subclass Dark Magic and slot Negate and cast it when the ballgroup is under pressure from other siegers. Then tell your friends in zone to do the same.
