Layering Adjustment: Gundabad and Elderslade

[Cordovan]

We just a bit ago made a change in both Gundabad and Elderslade to reduce the number of active players in an area that it takes to create a new layer for these areas. This causes the areas to layer once the population goes above fifty. Players can migrate to their Fellowship leader's instance using the layering alert that pops up. Let us know how thing go for you in terms of game performance, and please do make sure to state your time (and time zone) along with server. We'd be particularly interested in hearing from players on Arkenstone, Evernight, and Gwaihir. Thanks!

[Braer]

All day on Evernight the main topic has been the lag. Also failed gold spam reports, but ignore works. Reduced Awareness popping up in the least taxing places like Lothlorien Ally Hall, homesteads.

With many running instances I took a trip to Elderslade and "who" only reports 27 in the area, but then we don't know how granular the layering is; the whole area or pockets as in the first few days of Far Anorien going live where we couldn't move far before being thrown into a different space with another set of spawns and nodes in several areas.
The ride through pre-war Elderslade was as much as I tried, no issues really but on Evernight we aren't used to going to such places at peak times. The lag felt everywhere else kind of puts you off making it worse for yourselves. A brief visit to the War area just reminded me of the flakey tech you use to create these spaces and missions so I just quit doing anything more. Until you fix the way you build them no amount of poking is going to make things better.

Non-anons consistently 1000+ all day. Test ride 13:30 eastern, our peak.

[Braer]

All you need to do is have some staff log on to Evernight in the afternoon and follow /world chat for a few hours. You can alter the default chat window and font sizes, even colors to "listen" to other players.

[Gandolf_TheOld]

Server Arkenstone.
Area: Gundabad - Mattugard
System: AMD FX-9590 not OCd, 32 GB RAM, RX 480 video card, not OCd.
Time: about 730pm till about 8 - 830pm EST (Detroit, MI)
Class Lore-master with Bear Pet

Last night, Nov 13th, the issues were real bad, continues herky-jerky movement, mounted or on foot: take a few steps and stop for 5 seconds, take a few more and stop again rinse and repeat.

Casting delayed, stuttering, laggy: cast, delay, animation, delay, delay, delay, delay, recover from cast, delay, mob times out and herkey-jerky returns to their spot, delay, delay, delay, then the effects go off: Ancient Craft sound, then Lightning Strike hits, ground burns: herky-jerky flames die away, fight over damage taken 0, damage taken by mob 0, fight over, Herky-Jerkyly leave the area and do it all over again. Manage to get back to Noble Gate through milestone skill, long delay in the animation for execution. Called it quits for the night. In short, the game last night was unplayable.

For the past month or so, Arkenstone has suffered various times and amounts of stuttering, delay and lag. Since the modifications were made, it has been getting worse, with last night being the worst yet.

Areas Mattugard, Pit of Stonejaws, to Deepscrave entrance, steepset, up to the exit to Car Bronach, up into, a little bit past the map change to Car Bronach via road and near to the top of the mountain from Elderslade.

Was on earlier yesterday around 12 noon to about 3 pm, same thing but much less. Thus, probably population related.

Will see what happens Sunday, the 15th again around 12 noon EST.

Additional info for the 13th, network bad hop at Chicago Level3 to Newark Level3 (ae-6-350.ear2.Newark1.Level3.net) pkt lose up to 90% over 15 minutes. But this is nothing for me with Level 3, historically the hop from Chicago to Newark via Level3 is flakey.

14 Nov: played from 12:00 pm to 5:12pm a few small problems, mostly minor intermittent delays. Dinner...

[Loox]

11/13, evening, east coast, on Arkenstone was horrid.

Relogging was a quick fix that only provided short term improvements.

Latencies were in the 40s with no packet loss while extreme stuttered movement persisted.

Ping tests to Level 3 were consistently in the low 30s.

Speed test to ISP showed full bandwidth.

Framerates tied to monitor refresh were capped and consistent.

Hope that helps.

I'm happy to give more precise answers (if I can) to more precise questions.

My guess is server side received packets are being delayed or forced retransmitted with the new layering.

[OghranNasty]

I did notice some improvements on Evernight, EU player playing EU evening hours.

Reduced awareness not on all of the time, I'd estimate 30% of the time now.
Interactions with objects, npcs, inventory has delays of several seconds instead of dozens of seconds.
Skill delays when fighting typically several seconds, plus several seconds for the damage to register on the target. Amount of 'invalid target' messages - even mid-fight - about the same.
Rubberbanding and stutter seems to be reduced a little too.
The amount of wild random camera angle changes remains unchanged.

So yes, I noted some modest improvements. Just not enough to want to play the game.

[OghranNasty]

Hm. 21:13 GMT+1 Evernight:

Moving an item from bag to shared took 1:35. Ouch!

Edit: there might be a network issue in London, which might explain EU players seem to be suffering more. My trace suggest that ae20-xcr2.nyk.cw.net, a Vodaphone node, is having issues. Could be a total red herring, as routing is isp dependent.