ESO latency /vpn

spekdah

When I first started playing 2 years ago, I had a 200ping from New Zealand. That popped to 280-300 up until late last year where they fixed something and it returned to 200. Now it's back to 280ish. I can run a VPN to drop it 40-80 more but it seems kind of daft when other games don't require this.

Thing is other MMO's both popular and Kickstarter, alpha, beta don't have this issue.
WoWs: 80-90 ping
EQ, DAoC, EQ2, 90-120 ping
Crowfall, other ealy access stuff, 100-150 ping
Tempted to find some free weekend to test some more.

Why is ESO latency like double other games with US based servers?
I have a newish machine, i7-7700K, Overclocked 1080ti. It's not that. Disable addon's nothing seems to improve it.
PvE is mostly OK except vet lazer room. Cyrodil is kinda horrible with the delays as its more pronounced.

Do other people have this issue? Can people please list their location with comparisons with other games? Is it similar?

Zealand

I have the same issue, from Auckland, NZ. I also had 200ish ping and now at 300 since this latest patch.

I'm back to using a vpn which gets it down to 230ish but also gets some bad lag spikes at times. I think last time it was high was something to do with the ddos protection, not sure why it went bad again?

Saelent

Out of curiosity, how many of those other games use megaservers?

spekdah

Saelent wrote: »

Out of curiosity, how many of those other games use megaservers?

EVE online is the most famous megaserver, but I only played that while living in England many years ago. Moved since and not played it. Could check ping when I get home though. Or ask on some forums.
Google Bloodbath of B-R5RB, 20+ hour fight with 7K+ players.

Others are old school choose a server with servers added or merged as population moves. Early access typically don't have production infrastructure in place.

I don't think you should use Megaserver as an excuse for high latency though. It's both an allocation and instancing technique on a server farm. No one shares server specifications or how many boxes they are running, or how much they invest in upkeep so it's hard to draw anything from it. There shouldn't be anything stopping them from adding more boxes to spread the load, unless it's architectural and say, each campaign is stuck in one instance.

ghastley

Ping has nothing to do with either client or server performance. It's purely a measure of the connection between them. The megaserver concept puts all the servers at a single location, so you can't find another server closer to you, because there isn't one. And by closer, I don't necessarily mean distance, so much as faster connection. A slow connection can be the fault of ISP's prioritizing other traffic because it makes them more money.

And megaservers were chosen over smaller, more numerous ones, so that populations in instances would always be sufficient for group content like trials. You could theoretically pass a character between VM's on geographically separate machines, but then your ping would be drastically inconsistent, which is harder to live with than consistently bad. You can argue about how many, and where, but it's always going to be a balance between the popular instances being overloaded, and the niche stuff being deserted, so you'll never keep everyone happy. Fewer also means they're farther apart, so the outlying players get worse lag. There's a case for an Asian server, but also a chicken-and-egg problem of not enough customers because of current lag to the others.

DaveMoeDee

spekdah wrote: »

I don't think you should use Megaserver as an excuse for high latency though. It's both an allocation and instancing technique on a server farm. No one shares server specifications or how many boxes they are running, or how much they invest in upkeep so it's hard to draw anything from it. There shouldn't be anything stopping them from adding more boxes to spread the load, unless it's architectural and say, each campaign is stuck in one instance.

You can use it as an excuse because they have consolidated their infrastructure in less locations. Instead of megaservers, they could have had servers in more diverse locations to better serve different regions.