John, that new increased sound buffers option does that.

Well the game does that normally, to some extent, but by raising and lowering the sound buffer levels you are choosing how much space is set aside for holding common sounds in memory.

@Aprone:
After rplaying through the single most awesome defender I've ever been on, I wanted to ask if it'd be possible to add trading during them. There doesn't seem  to be any downside, and it'd really help us focus damage (I had 90 points available for my single-shot ACR whereas somebody else was blasting away with a vulcan). It'd also let people hand off items if they're about to die, and hopefully give the rest of the team a better shot.

Just do a gift wrapping  event to keep us happy LOL.

@aprone
people really don't care about defender, mainly because it doesn't reward you at all.
It's pretty fun though when  you're on skype or teamtalk with a bunch of friends.
I however have suggestions, I think xp and rep gain from killing zombies should be enabled on defender missions, (encourages people to kill zombies),  and the reputation pot also needs a boost, At around wave 10-15 the pot gets up to 500 at maximum. imagine  sharing that amount with 7 people. so you'l basically get nothing out of it.

To reward players for the million zombie challenge, the server has been updated to add in a team of 10 AI controlled bots.  These 10 bots (Stonedust, Magicdust, Coaldust, Golddust, Bonedust, Sawdust, Chalkdust, Cricketdust, Metaldust, and Claydust) will automatically jump into launched CTF missions, to balance the teams.  In the past if you got together 2 friends and wanted to play a CTF match, your only option would be a boring 2 verses 1 match.  With the introduction of these bots, you could choose to have all 3 of you on one team, and 3 bots would be added to fill in on the other team.  So the same 3 friends would now be able to have up to a 3 verses 3 game without needing to convince anyone else to join in.

I sincerely hope this will breathe some new life into the CTF matches.  I've always thought it would be cool if we had an actual league, with teams that competed in official tournaments.  We never had enough interest to make that happen, but maybe by some small chance this will help push us in that direction.

There is a client update in the works, but I still need to do quite a bit more to it before it is ready.

hello.
@Aprone, Is it possible to add an options in Swamp practice mode, to make the items respawn like zombies? now, if you start a map with 100 zombies and 100 items, the items will not respawn, but the zombies will automatically respawn. this idea will make swamp single player more fun, and some people who can't aford swamp because of their geographic location or lack of money, can still enjoy playing swamp

I love the new CTF LOL love killing the bots, Thanks prone.  I wish you could steal their weapots when they die.  Thanks again.

I'd originally made this post to mention bot reloading and fire rates, since they just seem to empty their hunting rifles without respect to clip sizes at an amazing rate. Instead, I'll turn it into some thoughts on the 5:5 CTF that just ended.
For the record:
Congratulations!  Your team won with a score of 7 to 2.  For the Red team: joão had 18 kills, 0 captures, and 1 returns.  Duckington had 106 kills, 6 captures, and 14 returns.  xorg had 48 kills, 1 captures, and 2 returns.  For the Blue team: Stonedust had 7 kills, 0 captures, and 0 returns.  Magicdust had 6 kills, 0 captures, and 0 returns.  Coaldust had 6 kills, 0 captures, and 0 returns.  Golddust had 2 kills, 1 captures, and 0 returns.  Bonedust had 2 kills, 1 captures, and 0 returns.

First, let me start by saying those kill and return stats look a bit weird, unless all the other humans just died a lot less than me. I wasn't counting, but would not be surprised to hear the vast majority of those bot kills were me alone.

Overall, this was insanely fun, and the most fast paced CTF I've played in a long time.
There was one point where I'd run out of medkits, and was standing just inside the door of our base (hidden by a half wall) with the hammer, smashing skulls as they came in. The bots were an awesome idea Aprone, and I can't wait to see where they go next! I'd also like to see some mixed human-bot teams, and suspect I'll have the chance to do so soon. The kill rate was unbelievably high.

On gunfighting:
The bots are absolutely lethal. They're incredibly fast, and they don't have to pause to shoot you while they run for your base like well... an army of robots after your children's blood. They do seem to die pretty easily; a couple bursts of an assault rifle does them in. That said, I also seemed to die a lot more easily than usual, so maybe its a balance change across the board. I had luck hitting them at long range, and there were points where they ignored me and just ran; my guess is they have randomly assigned strategy and a limit on range.
When they do shoot though, you die. At times I felt like I was just being utterly annihilated by a swarm of piranhas. I'd pause to return fire against a single bot, and another one would come running in and shoot me from  behind. I'd kill the first, turn around, and here comes a third.

On strategy:
The bots have some amazing coordination, or maybe it just seems that way. No matter what I did, if one bot was around, another would follow in only a second or two. They made for a very different ctf than what I was used to, though the sheer number of players did that as well (this is the first time I've seen 10 people since I started playing again at the end of April).
I'm honestly not sure how we scored as well as we did; my only theory is that the bots struggle a bit with flag capturing itself and try to make up with that by slaughtering their enemies utterly. If they were as good at running flags as they were killing me, we wouldn't have stood a chance.

John, as far as I can tell from the code the bots obey the correct clip sizes, firing rates, and reload times for the guns they use.  Since they don't suffer from any latency in 1 direction (they Are the server), it may make them seem a little faster than other players.  The other possibility is that, as you noted yourself, the bots like to move in teams for support and you could be hearing shots from multiple bots who are running side by side.

I don't think the server is correctly crediting the bots with flag returns.  I'm glad you posted that mission score because it made me notice that.  It's on my list to fix now.

One of the bot's biggest advantages is it's ability to seamlessly turn and fire at a player without abandoning it's travel plans, haha.  There are of course players who can time and balance their side step keys and backwards to do the same, but it's hard to do while concentrating on aiming and firing.  Most players wouldn't notice, but because they are constantly mixing side stepping (and backwards) into their movements, their overall speed is a bit slower than the human players.  It's nice knowing that their fancy spinning while moving trick comes with a bit of a down side in terms of overall speed.

I've raised their starting health from 175 to 200 today, since they do drop pretty fast once you can get some rounds headed in their direction.  That health seems unfairly high compared to human players, but the bots aren't able to benefit from health boosts or the armor that players pick up as they go.  In early tests I gave the bots only 125 health (what players get) and they were dying so fast it wasn't any fun at all.  If I end up coding them to benefit from armor, I'll likely go back and remove this starting health boost they get.

Speaking of their limitations, you are correct that they've been limited on range and what weapons they can use.  I didn't want them using anything too laggy (I'm looking at you vulcan!) since the server has to still do all of it's usual workload, plus the work of simulating 10 players now.  Giving them the same range as players was brutally unfair, as I saw in early tests.  I may tweak them over time, but for now I enjoy not being sniped from 50 tiles away, haha.

Keeping processing power low is also why they sometimes ignore you and just run to where they are going.  At times where processing would be over their "limit", they dial their brains down to simpler tasks.  I'd love to squeeze a bit more performance out of them so they do that less, but even them just racing in to grab the flag and run helps their team considerably.

Strategy is something the bots try to use to their advantage.  Moving in groups is something I'm glad you've noticed, and I didn't know if it would come across in audio.  I thought their mass of footsteps might make them sound like a big blob.  Like moving pieces in chess where you defend each with the others, the bots do what they can to always have a teammate in the area that can jump over to help.

This morning I made some changes to the bots.  They should use voice taunts after getting kills (sorta hard to be in the right place to hear it, but it happens), their offensive strategies were tweaked a bit, and they can use burst mode on 4 of the guns.  When it comes to bots in any game, code intentionally holds them back from operating at their true potential.  If in the course of the match the bots notice that you are better players than them, they will gradually dial up their skill to provide you with a better match.  That part hasn't been tested much, but I hope it works as intended.

@Aprone:
I really, really enjoyed reading that post. Great details .
Rather than try and really process it right now, I'll put your last bit to the test and throw a nonstandard situation at the bots. I'll go on a ctf, and apply a couple criteria:
(1) it'll just be a one on one game, and (b) I'll play for a long time to see what happens. Earlier the bot ended up grabbing my flag and running back to its base, but since it was just the two of us that gave me lots of time to go loot and then barbicue the poor thing. I'm interested to see if that behavior changes after a while.

As an aside, a while ago you'd mentioned that one simulation you'd created taught the computer to learn from past instances of itself. Have you implimented this here?

That CTF addition is nice can't wait to check it out. I didn't play many back when I played more, but I need to make some time to jump on the game. I'm going on a trip next week, but I hope to be able to do that when I get back