So I wrote an EQ Server population parser... here's some data !

[xious]

Originally Posted by baalzy [You must be logged in to view images. Log in or Register.]

Does your database store a 'last date seen'? It would be awesome if we could filter to only include toons which you've seen in the last 30-60-90 days or something.

Also, if you really really really wanted to get fancy(re:tedious) you could park alts in each zone that way you can log into them and /guildstatus all the anon players.

Yeah, I store the date at which the character was last seen and I could do that but I don't want to do it until I am able to pull data on a regular basis. Otherwise there would be "holes" in the data.
Like, right now, I haven't pulled data since January 6th.
I have an idea on how to tackle that but I haven't had time to do it yet.

I introduced http://p1999pop.dmsimard.com/population yesterday and I have some other stuff I want to work on - maybe within the next few weeks.

The guildstatus thing would be too tedious, ugh [You must be logged in to view images. Log in or Register.]

[xious]

I just did a small tweak to the application that parses the character information.
Kind of hard to explain but if it gets fed a character that is anonymous (or roleplay) AND it already has a record for that character, it won't update this character's information.

Before this change, one day I could have picked up a guilded 60 druid and if he went anonymous the next day, he would show as an unguilded anonymous in the database. Now he will remain as a 60 guilded druid and ignore the anonymous record.

I did this in hope of reducing the amount of anonymous characters there is in the database. Close to 10% of recorded characters are anonymous. Out of 15k characters, that's a lot !

Edit: For the record, there are 1396 anonymous characters in the database right now. Curious to see if I can get that number to drop in the next few days [You must be logged in to view images. Log in or Register.]

[Messie]

this is awesome. thank you for posting this!

[xious]

Cool, we're able to pinpoint when there has been a downtime or a patch [You must be logged in to view images. Log in or Register.]

http://p1999.dmsimard.com/population

[xious]

Hey. It's me again.

I've added time ranges to the population graph - I've been mostly lazy in the implementation and I attribute that to my lack of free time recently. It does the job, however.

See for yourself: http://p1999.dmsimard.com/population

Enjoy.

[Tubben]

/who count 10 all
/who count druid all

If you want to see all online per level / class

[Valoril]

What would be interesting but then the database grows really huge and you need a script, would be a statistic about the "hours played".

Like :
Over the last month :
20 % players represent 0-10 hours played
35 % players represent 11 - 50 hours played
etc

This would give a good idea about the percentages of addicts (f.ex > 100 hours/month) vs casual but regular (f.ex 25-50 hours/month)

[xious]

Originally Posted by Valoril [You must be logged in to view images. Log in or Register.]

What would be interesting but then the database grows really huge and you need a script, would be a statistic about the "hours played".

Like :
Over the last month :
20 % players represent 0-10 hours played
35 % players represent 11 - 50 hours played
etc

This would give a good idea about the percentages of addicts (f.ex > 100 hours/month) vs casual but regular (f.ex 25-50 hours/month)

Originally Posted by Rhambuk [You must be logged in to view images. Log in or Register.]

I don't think its possible because you cant /who this information but

would love to see a male/female character perspective, Ogre warriors not that rare, FEM ogre warriors on the other hand.

Unfortunately those are things I can't retrieve client-side ! Sorry !

[xious]

Originally Posted by Langrisserx [You must be logged in to view images. Log in or Register.]

omg you rock that kicks my pop tracking project right in the face!!

Sorry ? [You must be logged in to view images. Log in or Register.]

What I want to do for the time being:
- A timeframe tab (Maybe something like 24h, 7 days, 31 days, 1 year, all time..)
- A moving average as per calculated by the data I have, not EQEmu's average which doesn't seem to be calculated very often

[xious]

Originally Posted by Langrisserx [You must be logged in to view images. Log in or Register.]

I was manually tracking population in Excel, when i see your beautiful graph it makes me shed a single tear.

THANK YOU for the GREAT STATISTICS!!

can we get independent readout for Red and Blue on the same graph axis

Yeah, I can. I'll add it eventually.