The resource bug isn't?
Scanning through the forums I noticed someone listing a few server variables.
One in particular stood out :
ns_playerresourcescalar = 0.233
If this does what it sounds like, namely scaling the resource income by the number of players, then it explains a lot of the resource problems.
I can see why it is done - not adjusting for players means that small games are relativly resource rich, while large games would be resource poor. Basically what resources there are are shared among more players.
However, changing the resource income in order to get the same 'kit'/upgrade density is just not possible.
Lets take an example:
Assuming 4 players per side, the effect of the variable isn't visible (num_players * playerresourcescalar approx = 1). At this rate, the marine commander starts out with an income of 12 per minute. Given his 100 starting points, he can build an armory and an infantry portal, then cap a resource. About 40 pts left. Now he needs to fortify his location, which will cost two turret factories and 6-8 turrets, say about 150 pts. At 24 resources per minute, that comes out at about 6-7 minutes before the base is secure and the marines can try to move out for the next resource or hive.
Now, imagine an 8vs8 game. Double the resorce income. 48 resources per minute, and suddenly its just a bit more than three minutes to fortify that spawn. Just 4 minutes into the game the marines are moving out.
12 players? Now its 72!! resoures per minute ... 2 minutes until the base and first resource is fortified. Then another minute gives enough resources to upgrade to HMG, another minute till you have a prototype lab, and three minutes later the HMG's are starting to spam out.
That's five minutes from gamestart and with only two resources... get a third resource and the commander will be forced to start spamming out stuff to keep down the resource pile.
Now, you could argue that it's the same thing for the aliens ... well, I don't know. I think the aliens are much more limited by time needed to build things, and also, if you have three gorges in a 12 player game, they will split the gorge allocation of resources three ways, so you have to wait EXACTLY as long time as in a 4 player game before the first resource tower is built.... which, bw, takes about 3 minutes. If a 12 player game used a single Gorge, it would be able to put down the first resource tower in about 40 seconds after evolving to a gorge, and after that point would be unable to put down resource towers fast enough to keep up.
So, what is the solution? Well, first of all ... get rid of ns_playerresource. I don't know exactly how it works - hopefully, setting it to zero disables it. Of course, if the formula is resources * num_players * ns_playerresource, then that doesn't work.
Next best thing would be to set it to 2/maxplayers. That would give you a correct, resource poor game if you have a full server.
For players, I'd suggest finding those servers with 8-10 max players. They are the only ones you can play and hope for any balance.
Secondly, if you want to be able to kit out more of your teams with MG, or allow the Khara to evolve, then you don't change the resource income - you change the costs of those things you want more of. Want more HMG's? Cut their cost then .. want more Onos? Cut their cost as well.
Personally, I'd advocate keeping the current costs. More players means less kit per player ... so be it. That HMG will be rare, and battles will be fought to retrive it. Turret factories must be a activly guarded instead of spammed with half a dozen turrets. Alients have to think before spamming the world fulll of offensive chambers, because RESOURCES WOULD MATTER!
Sure as hell beats the current situation.
Also, note that n_players means that the more players you have on your team, the more resources you get. How nice. Now we know why 7vs5 games is always won by the team with 7 players.
Oh well, that was a lot of writing triggered by just the name and value of an unexplained variable. I could be completly wrong of course, but it just fits _so_ well with what I've experienced the last week.
Anyone care to tell me if I'm wrong?
Did performance optimizations for UWE summer/autumn 2012. Not currently employed by UWE.