Jump to content

What is the Highest Populated Server? We have the answer right here! (NA-Servers)


kanechart

Recommended Posts

So I know a lot of people ask what server is most popular. We also know the Tor Status sites, etc are 100% wrong in dictating server populations.

 

So today I went ahead and create a simple spreadsheet that is not 100% accurate as we have not taken dozens of samples at different times and the simple 100 /who cap. We could find ways around the who cap do something like 60 Powertech a then b then c. But this is suppose to be an idea not perfect.

 

Here is the chart:

https://docs.google.com/spreadsheets/d/1MeGHMvZKZ619M7Zoc6QycJwQmIkckTpfDWron7BIqgc/edit#gid=974265505

Created 1/7/2015.

 

I hope this helps whoever needs it or wants it. I really did it for myself and a group of friends but it sounds like a lot of people are interested in it but also might help find some new players a new home.

Link to comment
Share on other sites

appreciate this being done, a few people used to do this around here now and then. Was curious about Revan dropoff rates now that its been a month. Pretty much sums up the way the order has been known to be for awhile though, (harb-shadow-hawk) but seeing how so many can't even field 100 at various classes, some ANY classes, is pretty bad. I hate to ask what op and wz queues are like on some of those servers. Edited by Maxil
Link to comment
Share on other sites

Thanks for taking the time to create this! While the actual numbers are of course not indicative of actual server populations, it is interesting to see the difference between the servers. I would have expected Harbinger to have a much higher lead over the east coast servers, but I guess it's because Harbinger has a high population 24/7, while the other servers are only high at primetime. Also, interesting to see how Imperials are dominating certain servers, especially the PvP ones, while on the others the factions are mostly even, with no server having a clear Republic lead.

That's about all I'd interpret into the data since it is not precise enough, but once again thanks for sharing!

Edited by Jerba
Link to comment
Share on other sites

Last night (01/06/15) around 10:30 PM Eastern, The Harbinger had a 15+ minute queue time. I would think this a strong indicator of server population/over population.

 

The server isn't full, there is some error causing it to stop people from logging in.

Link to comment
Share on other sites

:rak_04:

So I know a lot of people ask what server is most popular. We also know the Tor Status sites, etc are 100% wrong in dictating server populations.

 

So today I went ahead and create a simple spreadsheet that is not 100% accurate as we have not taken dozens of samples at different times and the simple 100 /who cap. We could find ways around the who cap do something like 60 Powertech a then b then c. But this is suppose to be an idea not perfect.

 

Here is the chart:

https://docs.google.com/spreadsheets/d/1MeGHMvZKZ619M7Zoc6QycJwQmIkckTpfDWron7BIqgc/edit#gid=974265505

Created 1/7/2015.

 

I hope this helps whoever needs it or wants it. I really did it for myself and a group of friends but it sounds like a lot of people are interested in it but also might help find some new players a new home.

 

I am a believer in sampled data being reflective of a broader understanding of what is going on inside the servers. And while your data set is limited in nature at the moment, it does generally track with my observations on the PvE servers in NA. So while absolute numbers are not known... your data does give a good feel for the relative population levels between the different servers.

 

Various people have done this in the past and unfortunately some people will pick and choose what to believe... but this data makes for good player-2-player discussion. It certainly puts to rest the often used and very hyperbolic "servers are dead" nonsense we see every month in the forum.

 

Two suggestions to help make the data better:

 

1) take your main samples at ~ the prime time peak for the server in question. It does make a difference, though less so with The Harbinger then the rest of the servers (since Harbinger is fairly busy 7/24). torstatus still gives a pretty good view as to when peak prime time on a server takes place.

 

2) take specific sample data to get a feel for the population distributions for a server from 1-60 during peak prime time. This is a one time exercise (takes a bit of time and effort though) to help profile the distribution curve of character levels (rather then just focusing on level 60s). I know you did this on sheet one, but it would be best if done in sync with my suggestion number 1.

Link to comment
Share on other sites

The server isn't full, there is some error causing it to stop people from logging in.

 

It looks very much like they have a population cap set on The Harbinger at a lower level then normal at the moment. Probably to help mitigate the server load and it's negative impact on ability delay. OR, it could be the server is simply malfunctioning (as some sort of artifact of the current performance issues players are seeing).

Link to comment
Share on other sites

It looks very much like they have a population cap set on The Harbinger at a lower level then normal at the moment. Probably to help mitigate the server load and it's negative impact on ability delay. OR, it could be the server is simply malfunctioning (as some sort of artifact of the current performance issues players are seeing).

 

If the cap was reset it should say "full" when there's a queue, but it happens at "normal" and "heavy" instead. Others have claimed it has experienced similar problems in the past, so perhaps it's a hardware issue, something is old or malfunctioning.

Link to comment
Share on other sites

If the cap was reset it should say "full" when there's a queue, but it happens at "normal" and "heavy" instead. Others have claimed it has experienced similar problems in the past, so perhaps it's a hardware issue, something is old or malfunctioning.

 

There is auto-code on The Harbinger (probably all servers) that set log-in queues based on server loading (not population load), but actual server load. The most common time you see it is after a server is restarted, but it can also be seen when a server is suffering in one or more other areas.

 

So NO.. it's not a forgone conclusion that if the login queue limit has been reached = server FULL.

 

It would be more accurate to state that a server has maxed out it's available resources when it flips over to a queue for login. It is quite possible that they have changed some variables around on our largest server to limit the total population (or more likely resource) load (as a mitigation for some of the ability delay issues until they actually fix them).

Edited by Andryah
Link to comment
Share on other sites

I'm sorry, are you all seriously accepting this as a legitimate estimate of server populations?

 

No it is but a sampling and I think people do understand that. When you see multiple numbers at "100" and nothing over that, it leaves me to believe there are more people in that class than what his sampling can account for.

 

But regardless it is still a useful tool. It gives a broad understanding of just how unbalanced class make up is on the servers. The low population numbers for the non-force use classes on the PVP servers is not surprising at all. And his data supports what I see everyday in most PVP matches on Po5.

Link to comment
Share on other sites

And his data supports what I see everyday in most PVP matches on Po5.

 

Sorry, you can't legitimize bad statistical data by adding anecdotal data on top of it. His chart is literally a one time sampling with numbers completely unknown past 100 and not including anything other than level 60 characters.

 

So today I went ahead and create a simple spreadsheet that is not 100% accurate as we have not taken dozens of samples at different times and the simple 100 /who cap. We could find ways around the who cap do something like 60 Powertech a then b then c. But this is suppose to be an idea not perfect.

 

For gods sake, he even states the fallacy right there.

Link to comment
Share on other sites

I'm sorry, are you all seriously accepting this as a legitimate estimate of server populations?

 

Dude.. he is not estimating populations... he is using the same sampling methodology on each server to build a table that shows "relative populations" between servers. And for that, his data is fine, and in fact looks historically accurate to earlier "relative population" data gathered in the past by other players.

 

It comes up often in the forums ---> "hey, which server is the most populated? ...."

 

NOBODY, except Bioware, knows the exact numbers for population data on the servers. Other then water to use on stumps.. we as players do not need to know exact population numbers either. Sure.. we would like to.. but we don't need to.

Edited by Andryah
Link to comment
Share on other sites

Hard to go by this...of course the west coast server was the highest populated, it was taken during their primetime...meanwhile for the east coast servers it's midnight on a Tuesday when the numbers are taken...clearly not primetime for anyone with a regular job or school to get up for the next day. If numbers were taken during prime time regionally specific, the higher pop eastern servers like Ebon Hawk would be much closer the Harbinger's number I'm sure.
Link to comment
Share on other sites

Another issue this doesn't address somewhat brought up above is how long a server stays near its upper numbers throughout the day, as noted somewhat above time zones play an issue. Harbinger being the unofficial Asian-pacific server though allows its numbers to stay pretty high throughout the dead of the evening most other servers go pretty empty during.

 

While the shading between the second highest and the least may require some fine tuning and testing, I don't think anybody would ever doubt Harbinger maintains the highest population most of the time, and holds a a much higher population grasp over other servers during high pop times. Such is a reason the Harbinger tends to have such frequent server issues

Link to comment
Share on other sites

I thought it was widely known already before this that The Harbinger was most populated?

 

You would think. But a lot of people end up finding tor status, and then spread here, twitter, reddit sub, etc. what it states the populations are. Then people join x server, then at some random dead point of the night ask why there is nobody playing. People have corrected TOR server for a long time now, and the updates for it go back all the way to 2012 at least as listed. It needs to be either taken down or explained straight up on its page how it gathers the data and is not representative of actual population.

Link to comment
Share on other sites

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.