Different teams, so that's why I voted no - But I think the team responsible for game performance needs to be enlarged and working 24/7 (three shifts), until they get it nailed down.

This should have been done months ago, when things started to go off the rails. I mean, what good is everything else all these different teams are creating for us if the game itself is unresponsive?

Game performance is an emergency situation in my opinion, which should justify immediate increases in staffing and probably bringing in outside engineering expertise as well. Things can't go on like this.

Lady_Karleya wrote: »

I understand if some folks have issues stemming from hardware that potentially could negatively impact their experience coupled with any performance issue server side.

Right.

MOBO: MSI Z390-A PRO.
CPU: Intel Core i7 9700K.
GPU: RX 5700 XT.
Internet: 6.7 Mbps. Ping: 30 ms.

Computer says no..

I wouldn't mind this at all. I have plenty to do in the game as it is.

No, a Chapter release is paid content and I sure as hell won’t pay ZOS for something that should be released base game. Besides, I think everyone expects some kind of new zone content in a chapter. Now if they wanted to pre-empt a dungeon DLC to substitute a major performance improvment package I would probably be OK with that.

Prefer black and white. If it doesn't matter one way or another, no need to vote. This is cut and dry. Visible. Clear. Voices are being heard, and numbers speak for themselves. Really the bottom line here. This thread will remain visible for a long time to come. I telework until January. I got nothing but time .

hafgood wrote: »

I can see lots of software developers in this here thread, not.

I'm actually studying to become one, but nevermind me!

hafgood wrote: »

Do you honestly think stopping future content would lead to bugs being fixed? Sorry but that's not how it works. New content is one team, bug fixes a different team. As developing new content involves a lot more than just programming you are putting a lot of people out of work.

I don't think anyone here is telling Zenimax Online to stop developing future content. Just not to rush the content and maybe spend a bit longer or more on optimization.

hafgood wrote: »

In addition you do realise that if there is a bug in a particular area of the code that is being looked at for new content they will look to fix the bug at the same time? They don't delight in the software having bugs in it.

I think we get that.

hafgood wrote: »

All pieces of large software has bugs, windows 10 has bugs, google chrome has bugs, the banks software that you use has bugs. Look up Interactive Brokers and the bug in their software that has cost them $113 million.

Sure, but a solid 200-999+ latency and FPS drops to well below 30? The FPS drops have been an ongoing issue. It seems that issue just got worse.

hafgood wrote: »

At the end of the days the bugs and performance issues cause a bit of frustration but what do they actually cost? You may lose a little playtime, be thankful that you have that playtime.

I'm not frustrated. I would like ZOS to know that I, and many other people, are absolutely fine with it if they decide to focus on optimization for a while. That's the aim of this thread. To give full support.

Hello. Neither. I want that the new content will be paused until a stability is given. I realy cant uderstand this 2 options.

Sorry laggs, strange forum bugs and missing letters after submitting

BXR, nailed it dude.

I think it'd be a good-faith gesture. We'd need to see releases that are actually effective though. They're saying that U25 and U26 fixed things - what things specifically? That's my problem, this disconnect. Based on the last interview with Rich Lambert, sounds like they're putting a lot of people on these performance issues and still don't know why they're happening.

Not that I won't buy content, but personally I have enough to play on this game for a long while. If they paused new content to make the existing content more enjoyable, I'd be more than happy to support them.

eKsDee wrote: »

Destai wrote: »

I think it'd be a good-faith gesture. We'd need to see releases that are actually effective though. They're saying that U25 and U26 fixed things - what things specifically? That's my problem, this disconnect. Based on the last interview with Rich Lambert, sounds like they're putting a lot of people on these performance issues and still don't know why they're happening.

Not that I won't buy content, but personally I have enough to play on this game for a long while. If they paused new content to make the existing content more enjoyable, I'd be more than happy to support them.

Based on the fact that they were looking at non-combat pets and character metadata loading as potential culprits for performance issues this patch, I absolutely believe they're just shooting blind, hoping to hit somewhere.

That's what worries me. It's this shotgun approach. I don't know if it's a skill vacuum or what, but something needs to change internally. I don't trust them to release new content on top of this without introducing new issues. Sure, some'll get fixed, but the foundation is weak.

Destai wrote: »

eKsDee wrote: »

Destai wrote: »

I think it'd be a good-faith gesture. We'd need to see releases that are actually effective though. They're saying that U25 and U26 fixed things - what things specifically? That's my problem, this disconnect. Based on the last interview with Rich Lambert, sounds like they're putting a lot of people on these performance issues and still don't know why they're happening.

Not that I won't buy content, but personally I have enough to play on this game for a long while. If they paused new content to make the existing content more enjoyable, I'd be more than happy to support them.

Based on the fact that they were looking at non-combat pets and character metadata loading as potential culprits for performance issues this patch, I absolutely believe they're just shooting blind, hoping to hit somewhere.

That's what worries me. It's this shotgun approach. I don't know if it's a skill vacuum or what, but something needs to change internally. I don't trust them to release new content on top of this without introducing new issues. Sure, some'll get fixed, but the foundation is weak.

What worries me more is what that says about the intuition of the developers, or lack thereof.

As a developer myself, if you know roughly the timeframe of when an issue first started (say, desyncs), the first thing that should immediately come to mind is to go back through your commits during that timeframe (in this case, around Harrowstorm's release), and figure out which ones could potentially have caused the issue.

You shouldn't need to use a shotgun approach to solve specific issues like this, because your intuition should naturally guide you towards potential culprits. This should be the first thing that pops into mind, and yet it doesn't seem to be the case here.

In a more general case, such as just trying to improve overall performance, the first thing that should immediately come to mind is to start profiling your code, take measurements to actually see whether your code is performing as expected.

In this case, Zenimax could drop in just a bunch of general profiling probes, that maybe measures the difference in server time before and after the code being profiled is ran (and so how long that code took to run), as well as maybe just some general stats of the server state at that time (load, memory usage, player count, etc), and then log the results coming from the probes in an easily readable data structure.

Push the patch, let it sit on the live servers for a few weeks, grab the logs, and start analyzing, figuring out what the data says. Look at where your performance is actually going, and plot it against player count or overall load to see how it scales.

Again, shouldn't need to use a shotgun approach, because you should be profiling and gathering data to help come to a decision, before doing anything. They might be refactoring non-combat pets for literally no reason, they might not even be that stressful on the servers.

The fact that they're using such an approach for this problem worries me far more than honestly anything else I've seen, because it goes to show that they really don't know what they're doing, they don't know how to manage such a large game. You don't try to blindly optimize without even knowing what you should be optimizing first, it's such a basic rule of development.

Eks,

I would be all for what they would need to do to analyze like that. If they communicated, came out and said this is going to cause' disruptions, we will need to pull the service offline throughout upcoming weeks however much, I would be ok with that. If they are actually committing themselves to looking deep into the issues behind it, how can you be mad at that? Not every game is perfect, not every game is going to 100% go without server flaws, capacity overload. But, games should not almost be neigh unplayable for hours on end for a large portion of the population because we let bugs long live and allow expansions/new content to bring in the droves and crash servers. I get we have more players, I get it's part hardware, part code.

I get that there are a billion moving pieces. So you're obviously spread thin if you can't do what you say here between releases and spend the time working as a team to gather and analyze this data. Are they just being lazy? Do they not want to go that in depth? I'm asking honest questions here. Not trying to bash or insinuate anything. But, if there are roads to be taken to get us to a continually playable game, why aren't these avenues being EXHAUSTED? Outside of money already in pocket mentalities. Surely they can't believe this will sustain...

eKsDee wrote: »

Destai wrote: »

eKsDee wrote: »

Destai wrote: »

I think it'd be a good-faith gesture. We'd need to see releases that are actually effective though. They're saying that U25 and U26 fixed things - what things specifically? That's my problem, this disconnect. Based on the last interview with Rich Lambert, sounds like they're putting a lot of people on these performance issues and still don't know why they're happening.

Not that I won't buy content, but personally I have enough to play on this game for a long while. If they paused new content to make the existing content more enjoyable, I'd be more than happy to support them.

Based on the fact that they were looking at non-combat pets and character metadata loading as potential culprits for performance issues this patch, I absolutely believe they're just shooting blind, hoping to hit somewhere.

That's what worries me. It's this shotgun approach. I don't know if it's a skill vacuum or what, but something needs to change internally. I don't trust them to release new content on top of this without introducing new issues. Sure, some'll get fixed, but the foundation is weak.

What worries me more is what that says about the intuition of the developers, or lack thereof.

As a developer myself, if you know roughly the timeframe of when an issue first started (say, desyncs), the first thing that should immediately come to mind is to go back through your commits during that timeframe (in this case, around Harrowstorm's release), and figure out which ones could potentially have caused the issue.

You shouldn't need to use a shotgun approach to solve specific issues like this, because your intuition should naturally guide you towards potential culprits. This should be the first thing that pops into mind, and yet it doesn't seem to be the case here.

In a more general case, such as just trying to improve overall performance, the first thing that should immediately come to mind is to start profiling your code, take measurements to actually see whether your code is performing as expected.

In this case, Zenimax could drop in just a bunch of general profiling probes, that maybe measures the difference in server time before and after the code being profiled is ran (and so how long that code took to run), as well as maybe just some general stats of the server state at that time (load, memory usage, player count, etc), and then log the results coming from the probes in an easily readable data structure.

Push the patch, let it sit on the live servers for a few weeks, grab the logs, and start analyzing, figuring out what the data says. Look at where your performance is actually going, and plot it against player count or overall load to see how it scales.

Again, shouldn't need to use a shotgun approach, because you should be profiling and gathering data to help come to a decision, before doing anything. They might be refactoring non-combat pets for literally no reason, they might not even be that stressful on the servers.

The fact that they're using such an approach for this problem worries me far more than honestly anything else I've seen, because it goes to show that they really don't know what they're doing, they don't know how to manage such a large game. You don't try to blindly optimize without even knowing what you should be optimizing first, it's such a basic rule of development.

Absolutely agree! When I encounter a nebulous issue with a product, I sit down with the user and watch them use it. Like you're saying, push the patch and watch. Do a live AMA and get feedback real time. Instead we get the combative tone from ZOS. I can't imagine saying to a client, "you don't have to be here".

You're absolutely right, they should be better at framing things. The fact they're looking at non-combat pets over the block bugs and desyncs is likely a management decision. We'll never know but working in software for the last 10 years tells me that low hanging fruit is pursued when you're being given goals to hit. I worry about the team size they're using on this, something tells me it's quite large, and that in-and-of-itself can lead to more problems.