If you mean like P10S, it's actually not that. I worked out that the snapshot in Pandaemonium is just a lot earlier so if you move way earlier you can consistently do the mechanics. Ways you can identify the start of the snapshot is: start of the animation (since animations are typically after the cast), counting 3 seconds from the previous effect (there was typically 3 seconds between each mechanic snapshot in Pandaemonium), testing moving at certain debuff timers (you see this mentioned in guides a lot because it's consistent) or sometimes another way is probably to enable certain things in the battle log to show when an attack is being prepared since it shows this even if it's not actually got a cast bar. So just simply by figuring out that I can consistently move earlier I was able to do Pandaemonium fights consistently.
Also, this is with over 150 ping and I don't have trouble as long as I understand how the mechanic is snapshotting and the order in which the snapshots occur.
What I will say about Pandaemonium is that it made it absolutely essential to understand what snapshotting is, in case people had just kinda gone through the game not understanding it and dying occasionally in High-End content.
Although I think most people that have trouble just don't understand snapshotting or the way in which a mechanic snapshots, I do agree that SE could do with improving the system, given that snapshotting is... not intuitive to new raiders at all, and many never figure it out despite all the explanations they have been given about it.
Don't personally experience this, with over 150 ping. But I just have a sixth sense for snapshotting from playing 10 years tbh. So I know when I have been hit before it displays that I was hit on screen. And if it says I was hit, then I completely agree it's because I was slow.The server’s movement prediction is so flawed that I can walk toward a ground AoE without stepping in, yet the game assumes I did and applies the damage and debuff anyway.
In any case, I do still support tick rate increases, multi-threading, helping people who suffer packet loss by selectively not requiring certain repeated packets to be in sequence (by using UDP), etc. In general, I think the code is really poor for 2025. It just has to be.



Reply With Quote

