Whenever I have around 13-14 people in my server, gametracker still will say that there were a maximum of 10 people in my server.. this is quiet annoying, since I work hard to get people.

Will this be fixed?

Graphs show averages, not maximums.

burn wrote:
Graphs show averages, not maximums.

I have noticed the same thing, even when the player count has stayed steady for 2-3 maps (50-75 minuutes).
However, if it's the same for everyone, it seems relatively unimportant.

Well, it says: 'Players'

So i think that it should show the amount of players.

It even happend that, in the middle of the day, when the server had a lot of players, gametracker suddenly said my server was dead and set the player amount on 0.



I work really hard to get people in my server and I would love to see that back in the graphs :/

st0rmblood wrote:

It even happend that, in the middle of the day, when the server had a lot of players, gametracker suddenly said my server was dead and set the player amount on 0.


That's usually a network communication issue I would guess.
It's happened to us a few time too.

st0rmblood wrote:

I work really hard to get people in my server and I would love to see that back in the graphs :/

I get that. It isn't just the competitive aspect of how you rank against others.
It's also (probably more of) a personal pride thing.

So imagine GT scanned your server over an entire hour:

#1 - 0 players
#2 - 1 player
#3 - 10 players
#4 - 1 player
#5 - 0 players
#6 - 0 players
#7 - 0 players
#8 - 0 players
#9 - 0 players
#10 - 0 players

Let's think about how GT *could* show that in a chart:
1) The average number of players, how it does currently - this would show about 1.2.
2) The maximum number of players - This would show 10, this would be very misleading.
3) The unique number of players - This might show 10-12, this would be very misleading.

You need to keep in mind that player counts actually fluctuate on servers, especially at map changes. People come, people leave, etc.
Yes, your server going down can impact the history too. That's something you'd want to try to work to prevent.

The way to get your historical data graphs to be higher is to keep players on your server longer.

It's our opinion that these two servers should be ranked differently:

Server 1: 10 players played for 10 minutes, server was empty for 50 minutes
Server 2: 10 players played for an entire hour

The average number of players on the server over that hour would rank those two differently, whereas the other two methods would misleadingly say these two servers are just as popular. They are not even remotely close.

Absolutely understand.

GT appears to work on an approximately once / hour update of the graph that we see, having done an averaging of the number of players detected at several points in time during that hour (presumably during the approximately 4 minute scan interval), then draws a pretty much straight line between the two points.

To do anything more complex and detailed would increase resource usage considerably, and with the vast number of servers being tracked, a sensible middle ground has to be maintained.

I have witnessed naive developers design and deploy event logging systems that are so resource intensive in their desire to 'trap everything', that the event logging draws almost as much resource as the events themselves.

Personally; I was just dreaming

Explanation appreciated.
You might, given the vast amount of spare time you have during the day , want to consider this as a sticky somewhere. Though as we know, most people will post without having read anything first.

okay, thank you

I have revised post that was moved

Sry, have to agree with some of the above comments. Lately on both COD4 and UT2K4 servers, players can be on for an hour and the fact that there were say 7 players on server for that amount of time, does not show. I can post admin console picts and time stamps but it is what it is.