25 Feb, 2009, Vassi wrote in the 1st comment:
Votes: 0
So I was curious, how do people here test their codebases to see how they scale? Or is it more of a cross your fingers approach?

I suppose the correct way to do it is to build some kind of bot harness and spawn off a bunch of instances; so does anyone know of any code that exists to make that process simpler? Wouldn't want to reinvent the wheel, says the programmer creating a codebase from scratch.
25 Feb, 2009, Zeno wrote in the 2nd comment:
Votes: 0
I swear I saw a topic on this on Gammon. Let me look.

[EDIT] This: http://www.gammon.com.au/forum/bbshowpos...
25 Feb, 2009, Scandum wrote in the 3rd comment:
Votes: 0
It's fairly easy to do in tintin++.

For example:
#gag {#Trying to connect to %1 port %2.}                                                                                                                                
#gag {#SESSION '%1' CONNECTED TO '%1' PORT '%2'}

#loop {1 100}
#delay {0.25 * &0}
#session ses&0 {<address>} {<port>};
#format name {\x%d\x%d\x%d\x%d\x%d\x%d\x%d\x%d} {{60+1d19}{60+1d19}{60+1d19}{60+1d19}{60+1d19}{60+1d19}{60+1d19}{60+1d19}};
#ses&0 #send {NEW\r\n$name\r\npassword1\r\npassword1\r\n};
#echo {<ffa>Connecting session %03d, username: $name} {{&0}}

Save that script to a file, and use #read <filename> to execute it in tintin. Should be faster than writing a custom client.
25 Feb, 2009, The_Fury wrote in the 4th comment:
Votes: 0
Loading a bunch of players to test commands over and over is rather trivial. Generally i load between 10 and 100 players that play the game in a similar fashion to a real player. They sit stand sleep, use shops, kill mobs, recall to the city and the like. There are many ways to achieve the same thing, in the end i went with a custom client and a couple of scripts to handle them. Load testing is something that should be done IMO, it will help you find bugs and design issues that you would not other ways find by play testing with minimal people.
25 Feb, 2009, Vassi wrote in the 5th comment:
Votes: 0
Thanks for the references guys. It helps.