Does anyone else get hit by these without actually touching them? I use a birds eye view, making sure I don't touch any but I still take damage from sonic rings. It's honestly infuriating. I do the same thing every single attenuation phase, and most of the time I take no damage but sometimes I take like 3 rings in a single phase. On my screen, it looks like other people are running straight through rings but not taking damage.
Is it lag? Am I just bad and should go delete my character?
Originally Posted by auBerg
You are a certified crackpot that is subservient to the manipulators of science who are dreaming to control knowledge.
Since I started hugging the inside rings the times where I'd take dmg without being close was drastically reduced.
I still sometimes take dmg that I can't explain and honestly I think having such a precise mechanic in a game like WoW is a pretty bad idea.
The thing about the rest of your raid is justlag , doesn't look like that for themselves.
It's very different how you strafe. If you strafe using right click + A/D or Q/E (depending on what you use) you will not get hit by them if you run properly. If you strafe with just Q/E or use W in some way then it's much easier to get hit when you're not 'even close to them' as the game updates your position slower than when you right click-strafe.
Yeah, that seemed to be a problem for some of our raiders (it was mostly getting hit by the Sonic Rings, the pulses are really easy to avoid anyways), but it was solved once everyone started to strafe with their mouse.
We had a problem few people last night with these rings.
Some people are under the impression their player model is bigger (e.g. bear form and should go cat). Other people claim they are running ahead into the rings in front of them and they take no damage because of lag. I think the point about mouse strafing is the most valid one, and it's to do with the way the game executes path estimation on client as well as the server.
Pressing any directional button without holding down mouse-direction sends a single instruction to the server per tick, and that instruction is "this way" meaning your player model and estimated stopping place is checked each tick for collisions. So if you consider how this works with the sonic rings, players are tested for intersecting rings for their target path as well as rings that would be in the place they stop at when taking into account the roundtrip instruction time between client and server. I'm not sure if this is also the case, but I'm pretty sure the roundtrip time is considered behind the player too, which means they need to be moving ahead much sooner or they are penalised for where they were.
Holding right click and moving works very differently from a single directional command. The player is given much more grace since the player direction cannot be estimated so easily, and if it was estimated, it could be so far off it looks like the player is darting around. Ever seen a rogue come out of stealth and appear to bounce around in spikes while they kick your ass? That's path estimation failing for right-click movement because the target was in stealth, which prevents the game from correctly estimating the original direction of the player. After a few seconds the path estimation smooths out and you can see the player move more coherently from then on. When using right-click strafing with Sonic Rings, the server also estimates the player direction and speed, and possible stopping place after roundtrip time, however the collision detection uses a smaller model* in this case, because it cannot easily determine if a player has minutely or completely changed direction since the last command it received.
There's probably plenty of game developers who could comment on WoW's path estimation system and server collision detection in much more detail than I can, but bottom line is you should always use right-click and move if you want to avoid colliding with things in the game, even if it looks like that would be impossible - the server may estimate your position in your favour (and usually does).
* it's not actually a smaller model, but tightens the estimate of the path moved by a certain grace amount