1000 Players!

[Sonarpulse]

http://aoswiki.rakiru.com/webpages/playercount.html#all
Just wanted to announce that AoS broke 1000 players online for since the BnS master server went up!

[Jdrew]

I wonder how jagex will react to this

[Sonarpulse]

Well their DLC gave them a boost too. However this weeks peak was smaller than last weeks, so I suspect/hope that like 1.0's sale and the introduction of classic mode, the DLC will only create a temporary boost in the player count, and in the long run we will close the gap even further.

[StackOverflow]

We technically hit 1k players a few days ago, but due to a hiccup with the master server, it didn't show up publicly. Regardless, it's nice to see that there is a steady inclination of players, unlike 1.0's sporadic player counts which are based entirely on the game being 50% off.

[Jdrew]

We technically hit 1k players a few days ago, but due to a hiccup with the master server, it didn't show up publicly. Regardless, it's nice to see that there is a steady inclination of players, unlike 1.0's sporadic player counts which are based entirely on the game being 50% off.

most of them probably don't know its free

[Handles]

IT being?

[rakiru]

If you change the number in the box in the lower-left corner to 10080 (a week, in minutes), you can see the data smoothed over a week, showing a steady increase in 0.x player count and a steady decrease of 1.0 player count, with each boost aligning to a major update (classic gamemode, DLC) and/or a sale.
Also interesting, and more clear if you set that number to 1440 (a day, in minutes), you can see how large the peaks are over weekends. It looks to me that a lot more kids play 1.0 than 0.x, who can only really play games on weekends.

Edit: Note - changing that number to anything high will take a few seconds/minutes (depending on your browser/computer) as it's having to do a fair amount of number crunching. If you do it, just be patient (or refresh to reset it).

[Sonarpulse]

10080 1440

Ah, good. I was having trouble finding the value of 24 hours to smooth over the daily bump.

Edit: Note - changing that number to anything high will take a few seconds/minutes (depending on your browser/computer) as it's having to do a fair amount of number crunching. If you do it, just be patient (or refresh to reset it).

Hmm, I kind of want to contribute upstream to fix this. Dynamic programming can make this constant time with respect to the length of the rolling average. Something like

Code: Select all

l = lenth of rolling average -1
values = map (\x -> x/(l+1)) values

avg[l] = sum(avg[0:l])
for k in 1 : (length(values) - l -1)
  avg[l+k] = avg[l+k-1] - values[k] + values[l+k]

[rakiru]

Right now, it has a record for every minute (unless the box goes down or one of the sites was down). I've been meaning to make it grab every 5 minutes when you request a week, or something like that, since it would reduce the bandwidth (and thus loading time) and also the time it takes to smooth it out, but if you can get that added to dygraph, that's one of those issues solved without any effort by me. :P

[Handles]

hey, rakiru, it's good enough as it is! It's bloody brillient!

[Sonarpulse]

Does it reload the data when you change the rounding and or time frame? I assumed most of the delay was rendering, especially since I tried it on a machine with a 250 mbit/s connection.

I might just poke my head in at the dygraph github then.

[rakiru]

It reloads it when you change the duration.

[Handles]

On my computer it takes a rediculously long time to load the higher the minutes.

[Space Inspader]

three days ago, we hit 1115 players.