If it was a public system in the sense that it was available on every server, you would be correct; however, there are two major populated servers, and I think we can coordinate a level of trust between them.
Sigh. Again, you're missing the point. Your back end would end up being a SQL server. Fine. What stops me from getting into that and bumping my skill up - or down - however I deem best to manipulate the system to allow me to stack. Or give me a bigger e-peen.
a password? anyway i think your skill (howerver you calculate it, i think total points per total playtime is good enough for a balancer) should be stored per server anyway. servers might run individual scripts or your ping simply sucks. also what does epeen matter if noone sees the tracked stats in raw numbers?
There's hacks for SQL servers exposed to the internet. You need to limit by IP, IMO. Hide the SQL server such that ONLY localhost can access it. Moreover, expose the SQL server and you can DOS it. Storing points in the server means servers will balance differently, and it won't be based on someone's global skill. You're going to do it, do it right.
Probably the same thing that stops you from editing the admin or ban list: security. Android, just wondering... do you think that we setup permissions with statements like GRANT ... 'user1'@'%'? It's easier to DoS the game server, and the SQL server can be easily concealed. Exactly.
No offense meant, but I think that most people in this world writing code are morons, until they prove otherwise. I'm in no way saying I think most devs are idiots - there's tons of very competant devs out there doing incredibly cool stuff. It's just too simple to do stupid stuff though. Far too many times do I look at people's code and want to beat my head against a wall. As I look back on some of my older code, I think the same thing of myself. Having dealt with....overseas....developers as well, I've seen a lot of really....really...stupid shit. And people don't get what's wrong. They don't understand that coding up a hack isn't acceptable for shipping code. Or that writing inefficient code cannot be allowed. Or that you need to think out designs thoroughly to avoid leaving some security hole. Writing up an application is simple. Writing it well is pretty damn hard and not many people/companies do that.
Thats the intention ... ... i dont see how a global stat system is superior, it just wouldnt account for lag and stuff like server settings (or even worse exploiting by joining an empty server and idle to get a lower rating) anyway, any system is exploitable, anyone will try to exploit and if it be only to prove the point.
I think you guys are thinking to complicated. We want no ranking table. Also hours played is only for internal use. No need to show it somewhere as webpage. If one player reached 100 hours there is no need to adjust his hours any more. (Maybe adjust it anyway if the veteran limit is raised somewhen). And why to join an empty server? Just to become a veteran faster? Then you will have only less people on your side at real games? Nobody would do this. You can only exploit the system if you and some other veterans are getting a second steam account for skill stacking. Too much effort just to play some boring rounds.
You think others are thinking? How very dare you! Anyways, the suggestion just to disable scoreboard and mute all players till round start would probably fix the initial stack, making it much more difficult to stack afterwards.
Exactly. Hours played, keep it internal, it's that simple. It's so simple it's nowhere nearly as exploitable as skill-based systems. Second accounts (which, if we find, we can manually add) and then we can just say that a minimum of 8v8 is required for it to track, etc. Sure, a few people will find a way to exploit it, but why? It just makes it more difficult for them to join the team they want. That will never, EVER happen. Makes organisation far too difficult.
Again, hours != skill. I can't say that enough. Tons of people play emp for ages and still aren't that good. If you're going to do something that's a bad metric, then I should just introduce you to rand(). Shuffle teams randomly.
Hours don't equal skill, but they're close enough, and pretty unabusable. On top of that, there are people ahead of their hours, better than they should be for the time they've played, thus balancing it out.
I've read bad code, and I've written some myself. Generally, it's something I avoid. True. Often times people don't care about writing it well. It's not a rankings table. The columns would be Steam ID, hours played, and the last update timestamp
Since apparently noone read my post I. This thread, and half of my posts are apparently invisible, I will attempt to ask his question again. P.S. Cellphone typing kinda limits my editing opportunities, so sorry this quote is a little blah.
Randomly mixed teams are bad because many people want to play with their friends in the same team. And a scramble voting makes only sense if it will happen before round start, because on many maps the first minutes are very important. Killing all players while a round running because of a team scramble will cause a lot of raging. Just think of a commander who has just stepped out to build an abandon refinery. Also by scrambling the team it will happen that most good players are on side. Then scramble again until everybody is happy? Could take a long time.
