In multiple posts Kabam themselfs said that there are multitude of game performance issues that they cannot fix on a quick note, thats why they are sending you compensations on a weekly basis as a way to say sorry about current state of the game, and thank you for staying with us through these dark times.
My packet loss has miraculously disappeared. Strange…
No complaints, but I’m also done for the evening and will see what it looks like/how it plays tomorrow.
Dr. Zola
I’m curious to see if packet loss has any impact, or if it was just a weird anomaly that has no effect. Coincidences do happen. For a few weeks a while back I saw both deteriorating YouTube streaming performance and slower loading times in MCOC, and it turned out someone moved in next door and deployed a bunch of Wi-Fi routers that were using adjacent channels. Switched it up a bit and problem went away.
Another possibility: for about a month or two I’ve been seeing fewer crashes, but more odd “pauses” that are happening during combat that are unusual for me, having rarely seen similar issues in the past. Not stutters that affect inputs, but literal pauses. I suspect this is part of a separate set of changes to address memory leaks in some of the resource allocation heavy parts of the game (it is something a few of us reported to the devs during testing). I’m not seeing what you’re seeing, but maybe you’re seeing a different version of what I’m seeing.
I’ve been messing around with different channels and shutting off a second router I had tied to backup my drives. No new neighbors, but I am trying to examine closely how the connection is distributed across channels and devices.
@DNA3000 packet loss could explain some of it…switching channels, I ran a speed test immediately after experiencing the same type of frame rate issues I’ve been seeing.
Packet loss was 34%, which seems inconceivable. Nothing else is affected, regular web activities on laptop unaffected.
So there’s that.
Dr. Zola
P.S. Disagree guy up above—just trolling all of my posts today or do you have something constructive to add?
I found those lags or sudden dash forward issues getting worse this month vs last month. I finished exploitation of all MEQ and SQ, the impact is particular obvious in arena. I stopped grinding since last week.
My internet environment support 4K streaming without any lags. I think the internet setup of players is not the key factor in this particular issue.
In general, I would tend to agree. But I've been discussing performance issues with Zola for months, and troubleshooting a single instance of a problem is not the same thing as attempting to determine the general cause of a widespread issue. The widespread issues are probably related to the devs efforts to refactor the input system and resolve some of the memory leak/resource contention issues known to exist in the game. Across the majority of players, those two are likely to represent something between 90 and 99% of all the observed issues.
But we also had issues before the Unity platform upgrade, and people were experiencing crashes before the current round of known memory leak issues, and we have Android players complaining of input issues that cannot possibly be due to the iOS Unity engine timestep changes. So for any one player, when they mention something wildly out of the ordinary, it is worth exploring.
There's a difference here between systemic troubleshooting and specific troubleshooting. Kabam has had complaints about input issues since forever. The iOS issue was detected in part due to a dramatic statistical rise in such complaints combined with a statistically significant set of tests. Kabam wasn't seeing an input bug report and tracing it to Unity. They saw a rise in reports and traced that eventually to something that changed coincident with the rise in reporting - the Unity platform upgrade. We're statistically certain that most of those increased reports are due to this bug.
But this is happening within the background of a "static" of constant bug reporting, and as Kabam reported in their tests, there's evidence to suggest that some fraction of bug reports are false positives - players reporting bugs that don't actually exist - and some bug reports are due to anomalies not connected to either the iOS issue or the memory leak issues.
So while when he hear hooves we're supposed to think horse not zebra, a good troubleshooter will always ask the affected person to look for stripes.
No one should be reading this thread and saying "well DNA and Zola are saying that Kabam's reports of the problems are false." Quite the contrary: I was saying basically what Kabam eventually reported *before* they reported it, with less detail,, and I've worked with Kabam to help refine their knowledge of the issues in both cases. I believe their findings are sound. But even they admit that the things they are doing might not solve every problem for everyone, because while they know there is *a* problem they don't know if they are solving *the* problem. There could always be other problems, something which they have admitted directly (and which all good troubleshooters should keep in mind: it is one of my "Gibbs rules" I tell our engineers: there's always another problem).
I don't have check-in privileges on MCOC so I can't help resolve either an input system refactor or memory leaks in the client engine. But I can work with a player to see if there is some other potential cause for at least some of their problems, and if it turns out to be a waste of time (which it might ultimately be), well it is my time to waste. If there's only a 1% chance I find something that would help another player, I think it is time worth spending.
Comments
Dr. Zola
Packet loss was 34%, which seems inconceivable. Nothing else is affected, regular web activities on laptop unaffected.
So there’s that.
Dr. Zola
P.S. Disagree guy up above—just trolling all of my posts today or do you have something constructive to add?
But we also had issues before the Unity platform upgrade, and people were experiencing crashes before the current round of known memory leak issues, and we have Android players complaining of input issues that cannot possibly be due to the iOS Unity engine timestep changes. So for any one player, when they mention something wildly out of the ordinary, it is worth exploring.
There's a difference here between systemic troubleshooting and specific troubleshooting. Kabam has had complaints about input issues since forever. The iOS issue was detected in part due to a dramatic statistical rise in such complaints combined with a statistically significant set of tests. Kabam wasn't seeing an input bug report and tracing it to Unity. They saw a rise in reports and traced that eventually to something that changed coincident with the rise in reporting - the Unity platform upgrade. We're statistically certain that most of those increased reports are due to this bug.
But this is happening within the background of a "static" of constant bug reporting, and as Kabam reported in their tests, there's evidence to suggest that some fraction of bug reports are false positives - players reporting bugs that don't actually exist - and some bug reports are due to anomalies not connected to either the iOS issue or the memory leak issues.
So while when he hear hooves we're supposed to think horse not zebra, a good troubleshooter will always ask the affected person to look for stripes.
No one should be reading this thread and saying "well DNA and Zola are saying that Kabam's reports of the problems are false." Quite the contrary: I was saying basically what Kabam eventually reported *before* they reported it, with less detail,, and I've worked with Kabam to help refine their knowledge of the issues in both cases. I believe their findings are sound. But even they admit that the things they are doing might not solve every problem for everyone, because while they know there is *a* problem they don't know if they are solving *the* problem. There could always be other problems, something which they have admitted directly (and which all good troubleshooters should keep in mind: it is one of my "Gibbs rules" I tell our engineers: there's always another problem).
I don't have check-in privileges on MCOC so I can't help resolve either an input system refactor or memory leaks in the client engine. But I can work with a player to see if there is some other potential cause for at least some of their problems, and if it turns out to be a waste of time (which it might ultimately be), well it is my time to waste. If there's only a 1% chance I find something that would help another player, I think it is time worth spending.