No, it just chokes apparently at random on a very small percentage of transactions. I don't know what causes it - I occasionally post a list of outstanding transactions, and fireduck sometimes fixes them. I posted a list yesterday, but they're still not fixed.
Yeah, I think I worked out the issue here. Basically transactions were being saved into the wallet db and not the tracking/bet db so I didn't see that they were not being confirmed. I've corrected some ordering there so it should always be left in a state such that if anything is wrong it will be clear in my reports. After a few hundred thousand transactions anything that can go wrong will. |
|
|
|
|
Those are processed now. Sorry for the delay.
|
|
|
|
The problem is, there is no other use for a mining ASIC besides mining,
There's a coming vanity key gen market brewing... I doubt ASICs can be used for that. EC Key generation is an entirely different beast. |
|
|
|
Seriously, what if the developers decided to change the block hashing algorithm from double SHA-256 to something else (WHIRLPOOL or SHA 3 maybe)? doesn't it renders ASIC useless?
My understanding is that an FPGA could probably be reprogrammed and flashed to whatever. However, an ASIC cannot and does one thing forever. |
|
|
|
Suggestion (from the competition, but whatever): I often use my phone to pay things and scan QR codes from my computer. To do that I end up having to take your address and pasting it into blockchain.info to get a QR code. Also, there is a URI standard for QR codes that can specify the amount as well: https://en.bitcoin.it/wiki/URI_SchemeExample: bitcoin:1NS17iag9jJgTHD1VXjvLCEnZuQ3rJED9L?amount=20.3X8Just tested and works with Blockchain android app. You can use this to specify the amount as well which makes it easier for users like me. I just scan and hit send. |
|
|
|
Can you clear something up for me? If I bet 2 BTC on Teh Beast, do I get (a) the currently advertised odds (b) the odds recalculated as if my 2 BTC was already in the pot or (c) different odds for my two spins?
It looks to me like it's (b), which means I'm better off sending two separate 1 BTC bets in 2 different blocks - because that way my first bet gets better odds. Is that right?
You get the odds recalculated for the block in which your bet is confirmed. So if you are the only playing, splitting your bets gets you slightly better odds. The risk is that someone else could come along and win. |
|
|
|
Your job seems to have cleaned up most of the old pending bets. What's going on with this one?
03a804d79a24ea92112c968652474911666ee57a5b67b10cd265c2d5c2393843
Its 3rd output (output 2, counting from zero) is unspent, whereas all the rest are spent. That's the oldest pending standard transaction now - and it's only 2 days old.
That got missed, maybe it wasn't old enough when I ran my program. I was basically deleting transactions that haven't confirmed in over two days. It is handled now. http://blockchain.info/tx-index/10905279/8140c939e8437d46d8a3536dd72e88be54d3a4c5a31f0542788186218141c46e |
|
|
|
It looks like you've changed it again since writing that message. I was just going to say that it's unlikely to grow very fast when you're offering a 1% chance to turn 1 BTC into 27 BTC. But now I see it's a 3.7% chance... Also, did the re-import finish? I see that the outstanding bets are mostly still outstanding. For example, look at: 010bc3ccaab3394915e6e1ecd026c126572f21501d6d89ca5efc7f3970df554d (2012-06-16 23:39:32) It makes 10 bets in one transaction, but only 9 of them were processed. And: 0e67d2804b78cd32fcac334b807b62bff7d4d0d3d7af293a42d80a29be031f0a (2012-06-17 08:39:55) only makes one bet, but it wasn't processed either. They're both pretty old. Here are more old unprocessed tx's. I think all but the last one pre-date the re-import you said you were going to run. 021f5d99ad2ddbaeacd2be982343e1994ebdf5a6e86fe1cc95c29a0533ba6b9f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, the different progressives now have some different tuning parameters. I have teh beast setup to not grow very much but to always give a good expected return. Anyways, I tracked down one of the lost transactions you posted. It looks like some things were marked in the database in such a way that my software lost track of the fact that the transaction to pay them never confirmed and wasn't being tracked. I have a job to clean that up that I'll run overnight which should fix the others. Thanks for bringing it to my attention. I'm not sure exactly how things failed in such a way where my view didn't show any problem. |
|
|
|
The 0.05 and 0.25 ones look kind of reasonable, but the 1.00 one is still nowhere near worth playing.
Do the odds of winning max out at 1%?
Yeah, there is a max at 1e-2 and a min at 1e-5. I'm trying to strike a balance between decent expected return and reasonable chance of growing to a good jackpot. |
|
|
|
I suspect part of the motivation for this is that the progressive games are entirely risk-free for the house. They take 1.5% of every bet, win or lose.
The site is doing very poorly on the regular bets - they lost 500 BTC just yesterday - so a risk-free alternative is surely attractive.
I've changed how the progressives work to make them more interesting for players. Now the win odds are dynamic based on the pot size. I have some further tuning to do but I am pretty happy with them right now. My simulations show them often paying out in the low hundreds and occasionally making runs into the thousands of BTC for some big wins. The advantage is that there is no upper bound to how high they can go and are not at all limited by our available funds. If no one is interested we could drop it but I think it could be engaging. And since almost all of the money bet goes back into the jackpot (even if it is the next jackpot) the overall payout is probably much better than Vegas progressives. However, I don't think they publish numbers so who knows. 98% of each bet will eventually be paid out to a player. |
|
|
|
The first 3 have weird input scripts (one paid to a pubkey, and two paid to anyone who could find an address with the correct 160 bit hash). The last 5 look pretty vanilla.
Yeah, those look fine. I'm reimporting all the blocks now, which will take some time. Maybe some transactions didn't make it into the database and thus didn't get processed. I have something to prevent that now but it might have happened. |
|
|
|
There are 8 bets which are more than a week old and which haven't yet been spent in a block. Every bet should be used to fund (part of) the transaction that's sent back to the bettor. What happened to these 8? Rightnow i think, satoshidice really f*cked something up. I haven't yet looked through all of these but on the ones I have checked the issue seems to be from this issue in BitcoinJ: http://code.google.com/p/bitcoinj/issues/detail?id=192I can probably process some of them through by hand while we try to work on a better solution. The bottom line is for whatever reason these transactions were created with inputs from unusual (but still valid) scripts. |
|
|
|
Maybe this has been up before, but what's the point having hundreds of recent bets on the front page? It often jams my opera making the site (and my laptop) pretty much unusable. For Satoshi's sake, I hope the page is not dynamic...  I have actually thought the same thing. Especially with CloudFlare they are losing a lot of the benifit. It is dynamic like a fox. Also, it is CloudFront, not CloudFlare. |
|
|
|
|
House edge is about to be temporarily raised to 3% from 1.5%.
|
|
|
|
I posted a this request on reddit as well and the winner is from there. http://z1x.dk/llama/ (z1xwitch) Thanks everyone. |
|
|
|
Here are my submissions, Dark:  Light:  White:  Closer Look (Light Version):  PM me back if you would like to discuss further details. Thanks, these are very nice. 43ac82e8341999ba458c2d39adea5ef03c99cbbf01a9ce47486f333705dce471 |
|
|
|
either way here is my submission  I like it. |
|
|
|
|
Show Posts
