Chom Isis uses an ebay-like model for bidding at the auction. I don't think it's perfect but Pete has raised an issue I do not think there is a solution for.
Here's how it works, with aid of an example
A pawn is given a starting value:
Pawn A:
+ Current Bid: 110
Pawn B:
+ Current Bid: 105
The system calculates the next bid that is allowed:
Pawn A:
+ Current Bid: 110
+ Next minimum bid: 120
Pawn B:
+ Current Bid: 105
+ Next minimum bid: 110
We have three players. Player one comes in and makes a new bid of 200. It only shows up as 120 because if no-one else bids, then player one will only pay 120. No-one else knows player one has bid 200.
Pawn A:
+ Current Bid Shown: 120
+ Bids:
---+ Player 1: 200
+ Next minimum bid: 135
Player two makes a bid of 250 on Pawn B:
Pawn B:
+ Current Bid Shown: 110
+ Bids:
---+ Player 2: 250
+ Next minimum bid: 130
So, player B comes along and makes a bid. All player 3 can see is the current bids of 120 and 110. So, he aims a bid of 160 at Pawn B:
Pawn B:
+ Current Bid Shown: 180
+ Bids:
---+ Player 2: 250
---+ Player 3: 160
+ Next minimum bid: 200
Now, the bid player 2 bid is still winning, so the price is rammed up to be the next lowest bid (by player 3) plus an extra bit.
The added issue is that you can only bid up to the resources you have. 400 at the start.
The resources are only released when the auction item is complete.
If a player bids 160 on each of the pawns then they will end up with nothing and can't rebid the money. Player 3 doesn't know (and should not know) what other bids are being placed.
Pete recommends not locking the resources for those bids that aren't winning. The problem with this is that player three can drive up all of the prices without ever worrying about losing cash. He could do this:
Pawn A:
+ Current Bid Shown: 200
+ Bids:
---+ Player 1: 200
---+ Player 3: 199
+ Next minimum bid: 220
Pawn B:
+ Current Bid Shown: 250
+ Bids:
---+ Player 2: 250
---+ Player 3: 249
+ Next minimum bid: 270
Player 3 is not in any jeopardy because his bids aren't going to win. This is the extreme case, which is a little unfair but I think the principle of it would work.
I don't see a way around this problem. All bids must lock out cash, there needs to be a limited amount you can bid. You just have to be canny in the way in which you bid. The guide might have to be a bit explicit on this but I don't see another way around it.
Of course, by adding pawns to players at the start (which was always the plan) will reduce this problem but there is no way out of the cock-fight that occurs right at the start.
Unless your big brain can think of a way around it...?









Of course player 3 has
Of course player 3 has "risk". He doesn't know the maximum level of the bids, so if he is blindly increasing the amounts he may move into a winning position on one. That's how an auction works. If you bid on eBay and lose then eBay doesn't commit you to not spend anything for the next 3 days.
You're looking at this from a perfect knowledge situation, not the blind-bid scenario that it actually is. The entire reason for a blind-bid model is to inflate prices to what the richest person can afford.
Also, consider this a game-levelling mechanism which favours the poorer player over the richer ones...
So, are you saying that
So, are you saying that anyone should be able to bid anything they want and there is no resource locking?
Oh and I've not given the
Oh and I've not given the perfect information. You can work out what someone else has bid by increasing your bid by one each time, stopping by round numbers.
Perhaps the best thing to do
Perhaps the best thing to do is say that any given bid cannot be more than your resources. So, you have 400 resources, then you can bid 400 on six items but you can't bid 983475394875 on them.
How's that? Is that better?
I don't want people cancelling bids, either. If the winner can't buy the auction item, then it should go to the next person.
I fully agree with the "no
I fully agree with the "no higher bid than your resources", and also happy with the winning bid having it's resources locked up, which stops the richest player buying one, and locking all others (which will fail, but he doesn't care as his money will have been spent on the first pawn).
Also agree with the no cancelled bids thing (though as above, there could be an option to move/spend the money elsewhere prior to an auction completion...for example;
Player A bids 1000 on Pawn A (auction end time 12pm)
Player A Passes Money to Player B
Player B bids 1000 on Pawn B (auction end time 1pm)
12pm - auction on pawn A fails, player A has no money, so no punishment
12.01pm - player B passes money to Player A
1pm - auction fails on pawn B, player B has no money, so no punishment
This could also be done using an alliance as a holding account, and could effectively let one rich player block some pawn auctions).
To be honest, there are so many ways of fucking the system that I wouldn't worry about it...put together a simple, robust auction using the standard online model. Lock up the winning players resources, and that gives the lesser players a chance to get back into the game...
Cool! Result. I'll code that
Cool! Result. I'll code that up (not a lot of work, actually, it's simpler than what's there now) and have it in the next Beta. I think with things like this, the best thing to do is code it and then let you chaps have a play. As you rightly point out, there will be emergent smart-arse techniques but the best way of finding them is through play. I'm going to change the postit note graphic for the resources used to make it obvious that only winning bids count.
Thanks Pete, twas a tricky one that I think we've got nailed.
I was having a think about
I was having a think about this. Initially the simple solution is best, however if there is some kind of abuse/dirty filthy trick there may be some simple ways of limiting it.
The most obvious way is via a "bid fee", where to place a bid costs a certain value. The problem here is determining the value
Flat Rate - any flat rate tax screws over the poorer player, whereas really you want to use something like this as a levelling tool
Percentage - what of...current pawn cost, or players total resources? Pawn Cost again favours the ricker player, while player resources would lead to "tax dodging" techniques, so probably not effective.
Volume - each player gets x bids per day, and additional ones are charged on a sliding scale (new bid = previous bid + y). Again this really favours richer players
Scale on Game position - you already have some "logic" for determining who wins. You could put a fee on bids based on the players game position.
Money Ringfencing - The money juggling thing that I covered in the previous post could be countered by doing full ringfencing on winning bid money (in that it cannot be spent or moved in any way, shape of form while the winning bid is in place). This is effectively what credit cards do (and hence where I came up with it). You would end up with each player having 2 pots of money;
And only pot A can be used for bidding, tranfers etc etc. Ringfenced resources are released if a winning bid is outdone, or when spent when an auction wins. This would, of course, require more development work, as you would need to point all current transactions to the correct pot of cash, as well as making sure that fund release works correctly.
If we find these sorts of
If we find these sorts of exploits are common then we can add more complexity. I worry that the system is silly-complex already, without more caveats and so on. There is a lot of dev work for ringfencing and also problems where the second-place bid doesn't have the resources when the prime bid fails. They will either have all their cash disappear suddenly, go negative or lose the auction. because you don't have enough resources when the person in first failed. That is a world apart from seeing that you have 4 hours to get 500 resources together.
I think there might have to be a tax on Resource moving, as that might kill a lot of jiggery pokery off.
If you implement full
If you implement full ring-fencing, then your issue with the 2nd place bidder vanishes, as 1st place will never fail (as all the funds for the bid are ensured due to the said ringfencing, or as I refer to it the thumbprint), so you don't even have to track the 2nd place.
Generally, trying to do a second place in an auction is more hassle that it is worth, and most systems that have tried it do a "buy option" after 1st place fails. You can't really enforce a 2nd place buy, as;
1) the person may well have spent the resources elsewhere
2) they probably don't even know if they are in second place (with hidden bids)
If you don't do full ringfencing, I would suggest instead of a 2nd place bidder reset the auction, and block the failed winning bidder from joining in the next time round...
I will always have to track
I will always have to track second place as it gives us how much first place will have to pay. When the bid falls through, the auction end time is knocked forward and the new person should be (not yet, but I will put it in) that they are winning a bid. That will give them some time to raise cash / save finances.
Spending resources elsewhere is one of the features of the system and is available to anyone. It's daring but should be allowed.
Interesting discussion -
Interesting discussion - sorry I've not been about for it.
If you put in ringfencing there is another issue - the player could still lose money due to a suicide (if that still happens?) and not be able to pay or you could make the money safe from that, but then people would overbid to keep their money safe as soon as a pawn gets critical...
I can see a lot of issues
I can see a lot of issues with the transferral to 2nd place if 1st place falls through. Foremost amongst them is that the players funds will probably be tied up in other auctions...You could then end up with a situation where a player has his funds taken away for failing to complete a second place auction position, and as a result of that also losing his place on a winning bid later on.
Righty. I have coded half of
Righty. I have coded half of it - there isn't a lot to change, really. Here's the logic, should be what you've recommended but please cast an eye:
Pawn value is shown as 100.
Next minimum bid is 110.
Player 1 makes a bid of 200.
It's the first bid, 200 leaves their account. (ringfence)
Value is shown as 110.
Next minimum is 120.
Player 2 makes a bid of 120.
They are not the winner. No resources leave their account.
Value is shown as 130.
Next minimum is 140.
Player 2 updates their bid to 250.
They are now winning. 250 is removed from their account (ringfenced)
Player 1 gets 200 resources back (unringfenced).
Value is shown as 210.
Next minimum as 220.
Auction item completes. Player 2 gets 40 resources back: 250 were ringfenced but only 210 were needed. Player 2 gets the item. All other bids are removed.
Auction items never fail.
That it? Seems sensible to me. Many thanks for the help/thoughts/wisdom/business process analysis everyone!
That looks good to me, and
That looks good to me, and should be fairly self-explanatory for each player as to what is going on...
Thanks! I agree it's much
Thanks! I agree it's much more transparent than before.
I'll also be adjusting the screen layout at the top where the values are held (the postits) to make absolutely sure you know:
1. How many resources you have.
2. How many resources you have in ringfence (I'll call winning bid resources)
3. If no-one else bids, what your rebate will be (informational, not guaranteed).