Given enough time, lost coins will be reclaimed by technology that is able to claim them.  Of course when the time comes non-lost coins will not be vulnerable to the same technology that claims lost coins.

Throwing in the towel is a product of how many bitcoin cycles you have been a part of.

Propensity to throw in the towel:

0 cycles: High propensity, desperate to "catch up" and no understanding of hodling
1 cycle: Medium propensity, you remember getting a 3x return, somewhat fearful that history won't repeat itself
2 cycle: Low propensity, you have seen an order of magnitude expansion of your wealth, good things can happen
3+ cycle: No propensity, you have seen it all, bitcoin don't care how you feel, life is good

Why not just make a couple hundred of them yourself?

You are correct, and there are many other ways to exploit the server memory.  For example, it has been shown (and I tested it on my own servers) that you can dump the HTTP headers (no trace left!) and extract session IDs.  You can then trivially use that session id to masquerade as a logged in user.  It's REALLY easy.

I don't know why you would have a space before bitcoind but that is irrelevant because you are getting the error.

My bet is that the privkey is bad, otherwise it would work.

All bets * (1-.981).  That is roughly how much the house is expected to rake.

Think of SatoshiDice as a laundrer, it returns you 98% (roughly) of your money over time.  Now some people play wierdly trying to "beat the numbers" and they might actually lose 100% of their money.  That sweetens Satoshi's profits.  But as Dooglus has shown, pretty much it's just laundering and taking a 1.9% fee.

I hope he won that $38,400 before he had already bet that much in losses 

If you are developing several applications for bitcoin I highly recommend you learn the bitcoind API.  It does a lot and works really well.  Why would you want to avoid having a local server?  If you post here you will get help very fast.

I will even get you started with this script that sends money in Python:


How easy is that?  Super easy!

I already offer a 0.01 initial bid.  You are saying people want to bid lower than that?

The game has run it's course, thank you to everyone who played, I had a great time writing the scripts to work with bitcoind and the blockchain. 

Exactly correct, I do not send the initial bid until the reveal address is seen by a player.  So imagine the database is configured to start a game at 12:00:00 PM.  12:00:00 PM may come but if the address has not been revealed to anyone then the initial bid is not sent and the game therefore can not prematurely end.  I put this in place so that games wouldn't be started by the initial bid and then ended without anyone ever seeing the address.

Correct, you can win ANYTIME you have the highest bid in the first block.  Blocks do take a long time, it's not that hard to get 30 minutes or an hour (although it is less common) between blocks.  In your case the initial bid was sent immediately and your bid 30 minutes later both made it into block # 211953.  That's just exactly the game.  The initial bid might be sent and ten seconds later it is in a block or it might drag on and on and on.  You never know.

If you send immediately then you have the maximum chance of getting included with the initial bid but you also have the maximum chance of being outbid.  So, again, that's the game.

Hmm, I think you are right...  Maybe I need to put more time between the pots so that each feel scarcer...  I don't know, I think you have a good point.  If someone has already bid, what is my incentive to bid, I know that someone with initiative is already betting.  If I bet I'm probably just going to get outbid.  Going to have to think about that...

What do you mean "initial block didn't get in"?  You mean initial bid?

Argh, I just updated my bitcoind and I lost the paytxfee value, so indeed I was not attaching a fee to the initial bids.  This is fixed.  I have sent you the missing .10 that you should have had for the bid and I sent you an additional .25 for reporting the bug.  Thanks.

I am not an advocate of ensuring the initial bid is included for three reasons: 1) It is rare.  Excluding this issue where I wasn't sending a fee, the initial bid almost always get the first block.  2) The player is only losing a potential win, they get their initial bid back.  So yes, you don't get the initial bid but "no harm done".  3) This game is not about winning the initial bid, it's about winning the pot.  If there were more than 1 player on the pot then 99% of the payout will come from the bidding.

You are lost my friend.  You might see other operators as competitors but that's really not how we see ourselves when speaking to you about issues.  We all want to see Bitcoin succeed and crap websites don't help the community image.  So we just try to help, no one told you to go away.  If we wanted you to go away we wouldn't have said anything and your site would have taken the inevitable plunge to obscurity because your site really is garbage design and won't appeal to people.  Remember most people owning bitcoins are fairly technical and can smell crap from a country mile away.  We don't engage you to have a buddy to talk ciphers to.

This is different in that you can win many multiples of your bet (not just 1x like black/red on roulette) if you get in late enough to get the whole pot.  You can only martingale if nobody else is playing against you.

For sure, I like the idea of bigger stakes.  So far I am keeping larger stakes for events... like I am going to have a big stakes for New Years Day.  That reminds me, I should add the New Years stake.

You placed your bid 12 minutes after the address was revealed and you are not happy?  That's pretty risky, people who win just place their bid immediately and then pay close attention to whether anyone is trying to outbid them.

In form of an update I added a small amount 0.001 for those who don't want to commit a lot of BTC and I staggered the start time.  So 0.001 starts 1 minute between, 0.01 is 2 minutes, 0.05 is 5 minutes and 0.1 is ten minutes.

Enjoy!

How many nodes are you connected to with your sending client?  Blockchain.info is very easy to get broadcast to because they have 2500+ connections with their custom implementation.  You need more connections to increase your chances to get to miners maybe?