Topic: ฿0.00 Low Hash Lotto Number 29 by Dabs. Provably Fair. Weekly Draws. ฿ - page 18. (Read 31355 times)

So will future lotteries be hosted on that site or is it just to post the hashes?

Got a domain name. http://www.LowHash.com ... eh, for lack of a better idea. Nothing there yet. It just points to this thread.

Confirmed.

Total tickets: 22. (also the odds per ticket)

Bought five =).
caf86300a99318ba7fa9fc6d8626fa544dda8c2ca33e7b32f181707adf65b664

Another ticket from unknown player.

Using Visual Basic 6 (Classic) I made a slow, but working program. Using a simulation of 600 character strings to account for the combined concatenated length of the transaction hash, the ticket number, and the 7 secrets, I ran sha-256 in an iterative loop with results as follows.

' interpreted (in IDE)
' 10   seconds for 1,000
' 94   seconds for 10,000
' 230  seconds for 100,000

' compiled
' 1.17  seconds for 1,000 10 BTC
' 11.5  seconds for 10,000 100 BTC
' 142.5 seconds for 100,000 (less than 3 minutes) 1000 BTC
' 5675  seconds for 1,000,000 (1.5 hours) 10,000 BTC

These results are automatically sorted as the array (listbox with 1 million rows, actually) is added to.

The 1 million ticket simulation took that long. It would have been faster to just create the hashes and then sort them all afterwards. As it is, my code was sorting as each item was added into the list. Near the end of the simulation at around 800,000 to 900,000 the speed of computing the hashes and sorting was significantly slowing down.

This is all on my 2009 year model T4400 @ 2.2 Ghz Intel Pentium laptop.

I then redid the simulation using actual previous secrets (that don't necessarily match to any date, I just picked them up from whatever I saw.)

In theory, I can "upgrade" this capability to sort up to 2 billion tickets, and make it significantly faster, so it's not something I'm going to worry about for awhile. 2 billion tickets is about 12 million BTC.

12 million bitcoins
1.2 billion regular tickets
400 million early bird bonus tickets
400 million volume bonus tickets
Total: 2 billion tickets

I can also make this faster by optimizing the sha256 code I'm using, but 3 minutes for 100k tickets works for now.

I will make the final program / calculator available, but for now it's still super beta version, has no error checking, and will certainly crash if you feed it invalid input (such as strings where it expects a number), negative or zero values, etc. And it's in VB, so maybe most people think it's butt-ugly or something.

Actually, all I have is this simulator. For the few tickets I've been getting, I just do all the hashing manually on a spreadsheet. I've since upgraded from hashing each string one by one (back in Dabs Lotto Number 1) using online sha-256 javascript function.

I've also experimented a bit with blocknotify and walletnotify on my bitcoin-qt client.

Using 8 digits as the ticket number, the game allows up to 99 million tickets per transaction. If I stuck to 10 digits then it becomes 9 billion tickets.

Again, not something I am going to worry about now, no matter how sloppy or slow my code is. There are only 11 million bitcoins in existence at this point in time. I'd be happy if someone bought 100 BTC or even 10 BTC worth.

Time: That's about two and a half hours. Okay. I can still use them.

I just did a UTC Midnight test again. It is now 3 minutes after midnight of UTC.

Here are my results


PrimeDice was actually early by a few minutes when I checked at 23:50 UTC



SRoulette uses Host Gator. hehe.

SD is the last one that still hasn't revealed it's secret. But it's only been 10 minutes past midnight.

So I will wait a bit and see when they publish the last day of May or 2013-05-31.

All other previously mentioned sites with secrets take a few more hours or another full day before revealing their secrets.

Oh. But that's done across hundreds of millions of accounts. I don't need to do that since I'm clearly taking a percentage once the total pot exceeds 1 BTC. The thing with bitcoin, there is currently nothing smaller than 1 satoshi. So I can't hide that. I'd just openly say I'll take the 1 satoshi instead.

I think the idea is to take fractions of less than 1 cent, because no one but the bank sees it. People aren't going to miss what they can't see. You can also make it larger, such as a few cents (or whole dollars) of the interest. People aren't going to notice.

Here in the Philippines, one of our highways (the SLEX and NLEX) has an e-pass system (RFID tag for prepaid toll). One time, I noticed a PHP 8 transaction in my monthly statement. Upon calling the customer service hotline, they said they will refund the charge. He slipped and also mentioned that everyone got that same charge, but they were only refunding those who called to complain.

100,000 cars pass that highway system with these e-pass tags. How many are complaining? Few. They can easily get a cool million.

I've seen it taken to a larger extreme in another movie, Firewall (Harrison Ford) but the baddie took $10,000 chunks from 10,000 accounts.



patrickim, I don't see your 2 tickets. Did you send to the correct address?

Quick go check the CoinWin lotto.
It accepts all kind of wallets !

Bought 2 tickets.

Good luck 

ie. he was rounding the satoshis. 

Amended Rules to 83.33% / 16.67%.

Short Link to Block Chain:
http://blockchain.info/fb/1lotto3

Less than 85 blocks to go for the Early Bird Bonus!

P.S. I might be out for the weekend. The links and graphic should update in almost real time. (And, in case of a bus, ... but let's not think about that.)

P.P.S. I just noticed, that Stunna sent 0.05 so he got 1 bonus. But if he had sent 0.06, he'd get 2 bonus tickets for a total of 8. But rules are rules. Bonuses are only applied on a per transaction basis. I can't combine two different transactions into one.

Bought 3+1 tickets.

Lower % to next sounds good to me.

Good luck

Bought 5, good luck all   

f575ea840cc9e23d39bcd07b3b2aeadd10f0c761668227c3c7d7d67a0d7883e0

I think 16.67% retain would be much better! Not sure why you need my 'vote' through. It should be that if no existing ticket buyer (proper to the official announcement of the rules change) objects, then rules can be changed. But enough yaddering about this, I sent 0.03 BTC so I should get 4 tickets

Ok. I need one more "vote" to concur, specifically Mr. TradeFortress, since he was the one that complained.

I don't quite get the Superman 3 reference, although I know about the giant computer. Lots of giant computers tend to kill humanity when humans attempt to unplug them. Skynet (Terminator), Seth (Universal Soldier). Maybe even Eddy (Stealth, the movie about the plane with AI).

Bitcoin currently only allows 0.00000001 as the smallest unit, which most people call a satoshi. The weird math can make all sorts of values come up.

Aye.


Are you attempting to pull a Superman 3 on us?  Spoiler alert: the giant computer you build ends up trying to kill you when you attempt to unplug it.

Mr. TradeFortress, kindly check my math. I think I would achieve the same goal by changing the percentages to 83.33% / 16.67%.

83.33% to winner
16.67% to next game.

While keeping the rest of the rules the same. The sole ticket buyer I don't think will complain if this rule is implemented, as if he wins, he gets a higher percentage.

*edit* Dooglus posted. Testing out 83% / 17%

Ok. That one works nicer. I didn't use any formula. I just made a simulation of the next 20 games with those parameters. The retained amount about becomes close to 20% as intended. But it changes when bets are not constant, like if there are more this game, or less next game.

All in favor of amending rules for Dabs Lotto Number 3 to 83% / 17%? I will update and also post here.

Due to the weird math, all results will be rounded to the closest satoshi.

Here's how to work out what the retained percentage ends up as if the number of bets is the same each week:


ie. a weekly retain of 20% results grows week by week until it gets arbitrarily close to 25%.

If you wanted to end up retaining 20% each week...


... you'd have to retain 16 and two thirds percent each week.  Which is ugly. 

Here are some alternatives:


I say go with 17% per week, which ends up around 20.5% in the long run.