The second server event has been concluded for about 12 hours now (as of the time of this post), so here’s the results:
- Tangier – 10046
- Fallingferret – 3624
- KuromoriYu – 2302
- Crossi – 1416
- R422 – 673
- FaithY – 286
- bluecrab – 191
- treamcaster – 162
- aleronives – 138
- hawkzer0 – 129
I’m glad to see that even on shorter events we get a bunch of participants. Thanks to everyone who participated!
Oleeee—-
> aleronives
Well done Tangier, another mammoth effort.
Thank you so very much..*Roxx_Gang*………..
Off-topic: I just want to say this – Sylverant is on fire! Just look at the player numbers on a daily basis. Woot 😀
Is there A population log somewhere? The “Other Place” had one I recall.
I have been keeping some sort of statistics in the database for a while now, but I’ve never gotten around to making them publicly available. It’s on the list of things to do, though.
I remember doing something a few years ago where i would send lines of text to create an html file, it autoupdated by saving a portion of the html, we used divs i believe to save the lines (not sure if there was a better way) and then we would create a new file with:
– the head portion
– the saved divs
– the new entry (in a div so it could be saved)
– end portion
We would repeat this as many times as we wanted, i’m not sure if it’s the best way because it added content without having a concern about the time scale, so for sylverant i’m not sure if you have another method in mind that makes data easy to read, having logged data every hour of every day for 6 months might be overwelming i’m sure 😛
Actually, the way I do it is to save a new row to the database any time the count of players on the server changes. Each row contains only two columns: the timestamp (in UNIX timestamp terms, so with one-second precision), and the number of players currently on the server. Any time more than one update comes in during a given second, only the most recent is kept (so if 5 people all connect to an empty server at the exact same second, the two entries in the database will be ([some date/time in the past], 0) -> ([that exact second], 5).
In this way, I get up to single-second precision with the data without having to store the number at every single second the server is up. Of course, this has a few problems if the server is down for an extended period of time, but hopefully we won’t have that problem too often. 😉