I havent read the code from qzjul bots, or rather I dont know how to read them. But from my point of view, the bots login to often and they are 'taught' to place prices between a certain percentage, so for example food on the market right now is $36 and their range to place is say +10/-10% , they will place all their food at $32-$39, and the cycle repeats every hour or so.

the graphs do not reflect 5 minute updates, as it says clearly, average price above the graph. thats why the only way to verify what KOH said was true is to watch the 6/12/24/72 hour updates next set if he happens to be able to sell $50-$100 food.

Aye that was my point.

I played all of previous to last set, I know what I sold for how much, I wouldn't of finished shy of $17m NW if your "observation" was correct.

From what I see in the bot history which we both have access to- is that they sign in every 140mins.

http://www.eestats.com/express/oldcountry/762/58
Click History.

And you're right about the market selling prices. That is one of the disadvantages that the bot has- it doesn't recognize trends, or how it should be. I'm certain if the bots spent their $ on bushels as stock rather than stock 1-2 bill cash on hand the bushel prices would go up.

I actually misread it- it's every 1hr and 40mins once turns are all used up. Destocking he has a different code.

Honestly, I don't like the idea of having bots in the game, especially since the game isn't inactive. I know the idea was to give attackers more targets so the newer players can get a chance, but if a player's willing to learn it doesn't take too long to get out of the complete noob stage. Just doesn't seem worth it for the downside of randomized market prices and randomized retaliation from a bot that can't understand diplomacy and emotion. I'm sure the randomized market could be improved with time, but you can't really give a bot emotion. Just my 2 cents :P

I also I don't like the idea of bots attacking targets based on an acre, nw, etc. formula (couldn't remember the formula) because if the point of the bots was to help new players, why implement something that would mostly target new players who don't know better? Also, having bots that attack in this way limits diversity of strategy, some people might like to take bigger risks for higher rewards and try to sneak past the other players, or any number of other strategies that require a low nw/acre ratio for a period of time. You can't possibly attempt these strategies when there are bots that can always calculate whether you're a good target or not and attack without fear of consequences (it's just a program). I personally think the bots shouldn't landgrab on their own at all and should just retal.

If the bots have to be in the game they should atleast not have the ability to landgrab, I feel like that would break the game. (I meant to stop with the first paragraph but I rambled on lol)

@assassin:
These are good points. I would like to present possible solutions.

Your question: I also I don't like the idea of bots attacking targets based on an acre, nw, etc. formula (couldn't remember the formula) because if the point of the bots was to help new players, why implement something that would mostly target new players who don't know better?

Possible solution: It would seem logical to write a formula for a bot to maximize its netgain regardless of who it attacks. So, your argument is certainly reasonable that newer players would be targeted more. However, there's a fix to that. The bot can still maximize its netgains but it could have an additional check. For an example:

Based on turns saved, production, stock, current networth, cash on hand, acres, all these could factors could give a player a score. The newer players would likely have a low score and the expert players more likely to have a higher score. The bots could then increase their probability to attack/retal the stronger players than the weaker ones, thus working almost like a handicap does in bowling. And if the word 'handicap' freaks you out, just think of it like this: it will make the game more challenging for veterans and provide opportunity for new players.

I think as BSTRONG mentioned in another thread- the bots should emulate human players. Some are LGers, some all-explore, some are newbie bots, some are smart and strong, some are killers. All this is easy to do.

I think I got retalled by a bot...is that possible?

Qzjul added the function for the bots to defend, but I don't think it went live last set. I took a quick glance at who attacked you. I didn't recognize any of them as bots.

Bots only "defend" as in buying military; no bots attack yet. This is an interesting discussion, I'm enjoying reading it; good points on both sides; I think I can address the concerns of assassin, tbh, though I do agree the bots are a bit dumb now; we *COULD* make them do market trend analysis relatively easily, in fact that might be a worthwhile exercise; and yea, we can make them stock bushels, especially if they are actually getting past $2B anyway (not sure if they are)

Well I don't really think the fact that they go past $2B is the problem. Sure, the bots should do better by holding close to $2B and stocking with the rest of their cash, but I wouldn't think the point of the bots should be to do the very best they can no matter the risks, I think the point should be to use the bots to improve the game. I don't think most normal players would hold 2B on hand unless they were on top and willing to risk it, most players would continue stocking below 2B so their money doesn't get stolen from them. If the bots played this way and stocked until under $100mil then I believe it would bring back the complexities of the bushel market instead of a flat rate throughout the entire set like last set. (Also, bots with 2B would be great bomb banks targets)

I like the idea of having different kinds of bots, definitely better than all LGers, all dumb, or all strong. I must have missed BSTRONG's post. Maybe diversify their stocking habits too? Some risk it with 2B, most stock completely, and some don't stock at all and buy units?

This needs to be adjusted to max of 4000.

function buy_tech(&$c, $tech = 't_bus', $spend = 0, $maxprice = 9999){