If the servers are overtaxed and not functioning properly, everything else is a waste.

MincMincMinc

MorallyBipolar wrote: »

MincMincMinc wrote: »

JustLovely wrote: »

MincMincMinc wrote: »

JustLovely wrote: »

I will forever believe that ZOS knows exactly what is causing the performance issues in Cyrodiil. They've fixed them before in the past, most noticeably when they replaced the PC NA server hardware. But we also used to get better performance for MYM/Whitestrakes that would magically get better for the event, then drop back to normal after the first maintenance after the MYM event. ZOS knows what changes they would make to make this happen.

If ZOS doesn't know exactly what is causing the performance issues and how to fix them then they are not the AAA studio they'd have us believe they are. So enough of the gas lighting and misdirection. (vengeance) ZOS just needs to make the investments they know will bring performance back to historically established levels with the same population caps we used to enjoy.

This is just not true. You can go back to that time and read through the threads. Zos initially put out a statement saying they would not see any performance improvements and they were just maintenance avoidance upgrades. People initially said all was fixed for a few days because they trusted a placebo that people dreamed up thinking somehow it was like upgrading a graphics card or processor on their computer.

It is far more likely that it was just player dilution that caused any difference. Like with older MyM when the pve new players joined the servers would all dilute outwards and ball groups and zergs would split up and not run into each other as much. Look at the recent MyM events. The past few have been abysmal for PvE quester turnout. IC was dead, cyrodil was still the usual daily crowd causing the same levels of lag. There are no incentives for these players to participate, but every incentive to not participate.

I don't need to review what others have posted on the forum. I was playing hours/day in Cyrodiil when the change was made. Newer hardware is what made the performance better. All of us who were PvP mains and playing hours/day at that time saw it happen in real time. Same goes for the performance improvements then degradation associated with the MYM events.

Well countless others documented during that time period disagree. Even react had a thread about how great performance was.....and then in the same thread he came out and said that actually it still lagged and abilities still failed to fire. (Sorry react, but its a perfect example). I also played during the server swap and thought the performance improved for a day or two....but then noticed it seemed to go back to normal once the regular playerbase came back and the hype pugs left again.

Just like the aoe tests and group heal tests zos did half a decade ago, tons of players flocked back to the game just on the hope that the silver bullet hit. The only performance improvement was that the usual lag ball groups were diluted out or not participating because of the downtime scheduling issues.

Player dilution is a far more witnessable event than believing in some conspiracy that zos can secretly flick a switch and make the server ignore terrible code and design choices. We know it is a software issue, hardware can only do so much anyways even if the upgrades were performance based.

Read Matt's post on it

"This will not result in any appreciable performance gain in-game, but it will result in a more reliable service overall that needs fewer unplanned maintenances."

Your posts regarding PvP performance past and present are making it clear you don't PvP much now or then....at least not in Cyrodiil.

Nice random personal attack, how about to elaborate to any degree on why you think this is the case. Funny considering I posted actual evidence of my claims and then you go on to a baseless personal attack.

Explain to me how live cyrodil is all peaches and rainbows? I guess lag is an illusion, player caps are higher than ever, que times dont exist, group interactions don't affect server performance.

edward_frigidhands

MincMincMinc wrote: »

Sluggy wrote: »

SeaGtGruff wrote: »

I have a hot take on Cyrodiil performance.

If deer were removed from Cyrodiil in an effort to improve performance-- presumably to little or no effect; it was long before my time, so I don't know whether it had any effect-- then maybe it would actually be more effective to remove wolves and all other wildlife that can become aggroed by our characters? That way, the server wouldn't need to keep track of whether or not we'd aggroed a dozen wolves and tigers while riding to a battle, keep up with moving those critters after us as they chase us down, keep up with how far outside of their normal territorial areas they've chased us, and how long they've chased us, etc.

I mean, how much processing might be reduced by eliminating wolves and tigers and such from Cyrodiil? /rubschin

I don't remember-- how were wolves and tigers handled during the first and second Vengeance tests?

The thing is we all know that's not what is causing lag. It lags when large groups, especially ballgroups, are in a small area fighting one-another. It happens every time and there is absolutely no way they don't have recorded metrics that show what is causing it and why.

It doesnt even require groups to be fighting. It begins lagging the moment groups dismount and begin buffing up. It happens all the time. I would follow an ep group in while they pvdoor and sit on the walls since there are only zerg guilds running around and 1vX is few and far between. The server is running perfectly fine until a 12 man dc group starts buffing up outside to back cap.

Its no secret, we know it involves groups stacking every effect in the game with no limiting mechanics like the game used to have before the pve changes for things like stacking the same dots and hots. Power creep and design bloat has only gotten worse and is only going to continue to get worse. People love power creep though and will adamantly defend it with their dying breath.

CatoUnchained wrote: »

MincMincMinc wrote: »

They have already gone over and upgraded the hardware with really no difference. If the issue is terrible code, new hardware won't fix things. Well, sure maybe if we leap forward to quantum computing it could handle the shear volume of calculations. The game is just too bloated after 15-20 years of the code being worked on by different employees. It probably doesn't help that zos keeps adding in bloat and putting new layers ontop of new layers every other patch.

Having no control or not putting their foot down when it comes to power creep is only going to make things worse. We already have skills and set bonuses that are multiple paragraphs long. At what point will we have short novels for tooltips? We already have to look and compare multiple out of game Wiki's to understand status effects and offbalance.

The stat and combat system is just not clear and concise to any degree.

Patently false.

The only time in the history of Cyrodiil that we had near flawless performance for months, with a pop cap of 200-300 players/faction at the time, was the three months after they replaced the server hardware.

When they did the 12 man groups in like 2020 they cut pop caps down to like 100 players and then as many expect when they did the server hardware changes they reduced it further to around 64 players. Go on live today and try to get a player count. Typically GH will have one zerg guild on your faction of 20-24 players. Then it will have 2 ball groups of 12 players. Then just add up the remaining floaters and afk at spawn players and you end up around 64 players.

Zos had quoted early day numbers of around 200+ per faction which seems to match up with the vengeance test where they are saying we were able to 4x the population. People still like to think cyrodil is big, but in reality there are more people sitting in que than there are likely even in the server. Granted the graphs they showed on stream only showed a 2x player count so I am not sure what metric they were showing.....

React wrote: »

MincMincMinc wrote: »

They have already gone over and upgraded the hardware with really no difference.

For about 6-9 months post-hardware upgrade, the game ran at a level I'd never seen in it's history up till that point (and since!). I was streaming often during this time and would frequently be in keeps where there were 50+ players present and on screen at once with zero ability delay whatsoever.

I wish they would explain why the performance suddenly degraded back to pre-upgrade levels just a short 6 months later. It was like someone had flipped a switch - they released a patch and suddenly the ability delay was back during prime time. Since they never bothered to offer any explanation for this series of events, we're just left to speculate on what occurred. Whether the sudden drop was related to the return of ball groups en masse, the spaghetti code compounding the issues further each consecutive patch, or a straight up reduction in server capacity for a reason not given to us.

But to say there was no difference at all is just untrue - the upgrade was genuinely the one single thing they've done in the history of the game that actually made a difference, and a massive one at that.

I only did a lot of 1vX back then, but as far as I remember we had a good week or so where everything seemed good. Granted we all know it only takes a different dilution of the playerbase to have a performance impact so those few good days could have just been some event or influx in playerbase. This dilution effect is probably the main reason why previous more popular MyM events were so much better.....the ballgroups and guild groups actually split off to different servers and were diluted by light attacking pugs. (EDIT: honestly this is a big issue I see with the vengeance testing since zos alienated and did not support the pvp players. The vengeance tests were likely diluted with more pve quester players that only light attacked in combat vs hardcore groups stacking aoe hots as efficiently as possible.)

There are just as many people and forum posts in those 2-3 weeks after where people still experienced performance issues. Even you claimed you were still having ability delay and how there was still lag and ability delay. (not to fault you with a gotcha it was multiple years ago. I wish there was a silver bullet too, but It just doesn't seem to be the case.) Not to mention Matt had made a post about the hardware change likely would not affect performance at all and were only for stability and maintenance avoidance.

It is obvious from that link that the concerns React is raising those years ago are about anamolous performance instances while acknowledging that overall performance has experienced significant gains due to the hardware upgrade.

Matt's comments regarding the server hardware upgrade were verifiably inaccurate.

moderatelyfatman

noblecron wrote: »

I agree and would love if they improved server performance but the issue is it's also an mmo. New content is what keeps the players and brings in new players. Asking them to stop giving us new content and fix the game is just a recipe for folk saying it's going into maintenance mode and is going to do more harm then good. What they need is to hire on more folk. One team focusing on fixing bugs and whatever is cuasing issues and a second to continue so that we can still have running servers.

Tell that to the players who can't login after every major update.