Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

How can we see overall sub numbers?

STAR WARS: The Old Republic > English > General Discussion
How can we see overall sub numbers?

Emeda's Avatar


Emeda
06.20.2012 , 05:54 PM | #91
Quote: Originally Posted by jarjarloves View Post
torstatus does it's recording during prime time. the transfers stopped after what 4pm? So no the skyrocket in subs is because people are logging back in and resusbcribing to play.

Sorry to tell you that
No its not. I said this would be used to act like there was a increase in subs once the transfers started,

You take all the players from the low pop servers and give them to other servers and then when you count the players you get an increase in the destination servers (because they got all the origin servers players) and you dont subtract anyone from the origin servers.

Also TORSTATUS tracs 24/7 I believe
If this game is so great then why do so many people that love the game spend all their time in the forums?

taliph's Avatar


taliph
06.20.2012 , 06:07 PM | #92
I wonder what the servers will do this winter when all us seasonal workers can actually log on regularly.... rather then maybe 6-12 times a month.

ZudetGambeous's Avatar


ZudetGambeous
06.20.2012 , 06:26 PM | #93
Quote: Originally Posted by Emeda View Post
One question with the skyrocket after the transfers. Could that be because you still credit light servers with 250 average? In all honestly I do not think that number is accurate anymore.

Its like they took all the people from the light servers and added them onto other servers but still were counted being on the light servers.

Thats the reason you see a skyrocket.

Have you noticed any change in the server caps? Or are the basically the same?
I'm also interested in knowing this. If you are still counting light servers as 250, then you are double counting a significant portion of the population. Using say 10 servers. Previously all 10 of those servers never went above light, so if we count those 10 as 250 each you get 2500 players. Now after the transfer with the same 10 servers you get 2 destination servers which now read very heavy (1000? players, just using random numbers to make the point) now you have 8 light servers at 250 and 2 heavy servers at 1000 gives you a total of 4000 players even though you haven't actually added a single player. Now since we went from 230+ servers down to 23(?) the effect will be slightly less noticeable, but it may completely negate the seemingly large gain you are calculating.

Of course all this is assuming population caps haven't changed, because if they have then the data is meaningless for this calculation.
I tell you this: though neither he nor they yet know it, he is the greatest of all the Jedi. Jacen Solo is the living Jedi dream. Even without the Force, he is more dangerous than you can possibly imagine."

Scorpienne's Avatar


Scorpienne
06.20.2012 , 06:27 PM | #94
Quote: Originally Posted by Emeda View Post
One question with the skyrocket after the transfers. Could that be because you still credit light servers with 250 average? In all honestly I do not think that number is accurate anymore.
Its like they took all the people from the light servers and added them onto other servers but still were counted being on the light servers. Thats the reason you see a skyrocket.
I just don't know. The disposition of the light servers is a continuing uncertainty. It's become my white whale. I've been chasing it around and I don't know that I'll ever catch it and solve what to do about low phi servers.

I think that what's going on with the low-phi servers in this data set is uncertain. Particularly because this data is a mix of pre and post-transfer data. So some of the data still does have population on the low phi servers from before transfer.

If you have a better way to account for them, please let me know. I'm desperate.

Quote: Originally Posted by Emeda View Post
Have you noticed any change in the server caps? Or are the basically the same?
The way that Mythbusting originally estimated the number of logins at each server status (LINK HERE) was to write a program to tell him when the server changed status (it looked at the server status page and alerted him) and then he went in and /whoed the characters logged in both factions. /Whoing the server like that, and doing it that many times, is a metric buttload of work. I have not replicated it because my life is waaay too short for that.

If any of you smart programmers reading this have some way to write a program to make a character login and then /who a server's population and dump it into a text file, pleaseohpleaseohplease let me know.

However, here are some baby steps... DATA HERE.

Note that these are *single faction* populations censuses. Either Rep or Imp, not both, so they're presumably about half of the total population.

Timestamp Server Logins Status
6/13/2012 18:14 Ebon Hawk 870 3-Heavy
6/14/2012 18:03 Jedi Covenant 983 2-Standard
6/14/2012 18:55 Ebon Hawk 1044 2-Standard
6/14/2012 19:16 Ebon Hawk 1121 3-Heavy
6/16/2012 21:19 The Ebon Hawk 1997 4-Very Heavy
6/16/2012 21:41 Ebon Hawk 1422 4-Very Heavy
6/16/2012 22:20 Ebon Hawk 1583 5-Full
6/17/2012 15:02 Ebon Hawk 1121 3-Heavy
6/17/2012 15:09 Ebon Hawk 1169 3-Heavy
6/17/2012 17:15 The Ebon Hawk 1258 3-Heavy
6/19/2012 21:01 Ebon Hawk 1658 4-Very Heavy
6/19/2012 21:13 Ebon Hawk 1555 4-Very Heavy

When I take the averages (God bless pivot tables!) and double them to approximate the population from both sides, I get this:

3-Heavy 2215.6
2-Standard 2027
4-Very Heavy 3316
5-Full 3166

Yep. Very heavy is greater than full. I have no idea why that is. /shrug Hopefully eventually I'll have more data and so that all the variability will be less important than the mean. At this rate, I'll have enough data in a couple of months...


Paige

Scorpienne's Avatar


Scorpienne
06.20.2012 , 06:30 PM | #95
Quote: Originally Posted by jarjarloves View Post
torstatus does it's recording during prime time. the transfers stopped after what 4pm? So no the skyrocket in subs is because people are logging back in and resusbcribing to play.
Sorry to tell you that
Quote: Originally Posted by Emeda View Post
Also TORSTATUS tracs 24/7 I believe
Jarjarloves, Emeda is 100% correct. The guy from Torstatus has his program running to track the data 24/7/365. The program gets the status updates every five minutes and the program updates the webpage hourly. I don't know where you heard otherwise, but if you'd please cite the source?

In fact, the owner has been kind enough to send me June data from Ebon Hawk. I'm still trying to figure out what to do with it. It's an immense data set because there are measurements every five minutes...


Paige

Emeda's Avatar


Emeda
06.20.2012 , 06:48 PM | #96
Quote: Originally Posted by Scorpienne View Post
I just don't know. The disposition of the light servers is a continuing uncertainty. It's become my white whale. I've been chasing it around and I don't know that I'll ever catch it and solve what to do about low phi servers.

I think that what's going on with the low-phi servers in this data set is uncertain. Particularly because this data is a mix of pre and post-transfer data. So some of the data still does have population on the low phi servers from before transfer.

If you have a better way to account for them, please let me know. I'm desperate.



The way that Mythbusting originally estimated the number of logins at each server status (LINK HERE) was to write a program to tell him when the server changed status (it looked at the server status page and alerted him) and then he went in and /whoed the characters logged in both factions. /Whoing the server like that, and doing it that many times, is a metric buttload of work. I have not replicated it because my life is waaay too short for that.

If any of you smart programmers reading this have some way to write a program to make a character login and then /who a server's population and dump it into a text file, pleaseohpleaseohplease let me know.

However, here are some baby steps... DATA HERE.

Note that these are *single faction* populations censuses. Either Rep or Imp, not both, so they're presumably about half of the total population.

Timestamp Server Logins Status
6/13/2012 18:14 Ebon Hawk 870 3-Heavy
6/14/2012 18:03 Jedi Covenant 983 2-Standard
6/14/2012 18:55 Ebon Hawk 1044 2-Standard
6/14/2012 19:16 Ebon Hawk 1121 3-Heavy
6/16/2012 21:19 The Ebon Hawk 1997 4-Very Heavy
6/16/2012 21:41 Ebon Hawk 1422 4-Very Heavy
6/16/2012 22:20 Ebon Hawk 1583 5-Full
6/17/2012 15:02 Ebon Hawk 1121 3-Heavy
6/17/2012 15:09 Ebon Hawk 1169 3-Heavy
6/17/2012 17:15 The Ebon Hawk 1258 3-Heavy
6/19/2012 21:01 Ebon Hawk 1658 4-Very Heavy
6/19/2012 21:13 Ebon Hawk 1555 4-Very Heavy

When I take the averages (God bless pivot tables!) and double them to approximate the population from both sides, I get this:

3-Heavy 2215.6
2-Standard 2027
4-Very Heavy 3316
5-Full 3166

Yep. Very heavy is greater than full. I have no idea why that is. /shrug Hopefully eventually I'll have more data and so that all the variability will be less important than the mean. At this rate, I'll have enough data in a couple of months...


Paige
Sorry I have no clue onto how to deal with light servers now.

Think the only way would to do a /who again for each server to get an average but I dont think anyone really wants to do that for a light server.

I did log onto my origin server just now 840PM EST (firkrann crystal) and there were a total of 2 people on imp side, thats not on fleet but total imps during prime time.

If I had to suggest a new value for light I would wait 2 weeks after the tansfers and consider it the same population as it was before the transfers with no people leaving or joining.
Those full servers had to get people from somewhere and I would say 99% came from the light servers. It would take some math but I think it would be close to accurate.

Just guessing I would give the light servers a 25 instead of a 250 but if it was 50 or 100 and the math works out I would be fine with that.
If this game is so great then why do so many people that love the game spend all their time in the forums?

Scorpienne's Avatar


Scorpienne
06.20.2012 , 07:04 PM | #97
Quote: Originally Posted by Emeda View Post
Think the only way would to do a /who again for each server to get an average but I dont think anyone really wants to do that for a light server.
Heh! Can you imagine sitting through opening cinematics on 188 origin servers to create level 1 characters to /who the population? At 3 minutes per, 2 factions per server, that's over 18 hours! There's not enough tequilka in the whole wide world to make me sit thorough that! :-D I guess 186 since I have characters on 2 dead servers, but still. You're absolutely 100% right - it's not worth it.

Quote: Originally Posted by Emeda View Post
I did log onto my origin server just now 840PM EST (firkrann crystal) and there were a total of 2 people on imp side, thats not on fleet but total imps during prime time.
Sanctum of the Exalted had 44 on last night when I went in to bug people about transferring. Def a case of YMMV.

Quote: Originally Posted by Emeda View Post
If I had to suggest a new value for light I would wait 2 weeks after the tansfers
Hmmm... yeah. I think I agree. We have to do *something* with them - but waiting until we're dealing with transfer-only data is probably best. Maybe I'll start reporting numbers a X + Y to Z where X is the population of the servers with a phi value > 1 and Y is the low end of the estimated logins on the low phi servers and Z is the upper end of the estimated logins on the low phi servers. That way people can call that number whatever the heck they want and how they interpret the variability is on their heads, not mine.

Of course, then we're all going to have to scramble to answer threads with ding-dongs who refuse to read the analysis running around like chickens going "bawk bawk bawk ONLY 20,000 PEOPLE SUBBED TO TOR bawk bawk bawk" and try to tell these chuckleheads what the numbers really mean...

I know you've rposted intelligently in my threads, Emeda. You have seen these people. Tell me I'm wrong about the chicken thing...

:-D

Paige

Drom's Avatar


Drom
06.20.2012 , 07:09 PM | #98
It's all messed up. That's exactly what BW wants. They transfered 200 servers, but not merged them, moved population caps on servers and F2P till lvl 15 is comming, that will boost it a lot more. We won't be able to say how populated the servers are anymore. So only official sub numbers metter now.
Dromfel, Imperial Agent (Operative lvl 50) | Imperator Mos | Sith Empire | Tomb of Freedon Nadd (EU PvP server)

DiabloDoom's Avatar


DiabloDoom
06.20.2012 , 07:11 PM | #99
I would guess around 800k-1m after transfers, and they people will hang around to see what 1.3 is like.

Scorpienne's Avatar


Scorpienne
06.20.2012 , 07:18 PM | #100
Quote: Originally Posted by Drom View Post
It's all messed up. That's exactly what BW wants. They transfered 200 servers, but not merged them, moved population caps on servers and F2P till lvl 15 is comming, that will boost it a lot more. We won't be able to say how populated the servers are anymore. So only official sub numbers metter now.
You know what buddy, you may be right. This has been a fun experiment, but maybe it's meaningless between the mergers, changed caps, and F2P trials. That's why my real project is the pop on my own server. I actually don't give a rat's patootie about subs. I mean, sure, I hope BW is making a reasonable and appropriate profit and that they'll continue to produce content, but that's in the hazy big picture. The only number that affects my quality of life is how many people are logged in on my server.

Paige