Bitcoin Forum

how complicated is to build a wallet generator, which will generate you adresses till you hit the jackpot ? ???
is it even possible ?

Mathematically insignificant chance for this to happen.

http://bitcoin.stackexchange.com/questions/7724/what-happens-if-your-bitcoin-client-generates-an-address-identical-to-another-pe

This sums it up:
https://i.imgur.com/VjtG3.jpg

You would make more money putting all that computing power into mining.

Well thanks for da quick answer.. i think this closes the deal "There is also a chance for you computer to catch on fire, and some of the materials to melt together into a lotto coupon with winning numbers on (and a valid barcode), but it just won't happen because of the chance is so unbelievable small (it's the same with the "click and generate another persons bitcoin address")."

I hadn't seen that one... may be my new favorite  :D

To answer the first part of that... no, it's not complicated at all. All private keys are 256 bits. Just start at a reasonable number (or even at the lowest possible private key, which is think is 1,) and calculate it's public key and address. Check for prior funding to the address. If empty, add 1 to the private key and repeat. Hit Ctrl-C when thoroughly bored.

You won't get rich, but it might be a useful exercise in coding key generation. ;)

Incredibly easy and incredibly worthless.  If you managed to get an address that someone had already generated (which is so unlikely it is relatively impossible), it would likely be an address that you yourself had mass generated.

I love that picture. Simply amazing.

You can use vanitygen to crack addresses with giant sums but it will take you a billion years
Statistically it could happen in a second tho.

You have no idea of how massively "a billion years" underestimates the true length of time it would take vanitygen to crack an address with a large amount of coins.

Overestimate billion^billion

So basically, you're saying it's possible. Just need to get a bigger star.

brain wallet mining is apparently really profitable.  i guess the majority of people don't realize that the first round hash of "this is a super secure passphrase because it's long" has less entropy than they think it does.

It's so possible that someone has done it already. ;D

https://bitcointalk.org/index.php?topic=107172.0

Oh really...

The probability of finding a private key to a specific address is the same probability of finding a hash of all zeroes. I don't know what the lowest hash found to date is (maybe 17 leading zeroes of of 64?), but its not even close. You'll need to have computing power many trillion-trillion times more powerful than the current bitcoin network, and a few trillion-trillion+ years to run it.

Awesome explanation!!!

Win win either prove security or hit the jackpot

I'm surprised that no one has created a *coin to do this  ;)

Wouldn't using two factor authorization make that method less likely of success?

outstanding

There is no 2 factor

Properly packaged, these could be an effective promotional item for Bitcoin.  The ideal package should be preloaded with orphaned and dormant addresses, require that the "player" download a bitcoin wallet before using the generator, and ideally have made some Bitcoin transfer to it.

As with any lottery, near misses to the big payoff should be displayed, and can provide tantalizing reinforcement.

If we can't market unlimited free chances to win millions of abandoned dollars in a way that promotes more widespread Bitcoin awareness and usage, we really are a bunch of nerds.

Regardless of the odds, they are a noticeable improvement over zero.  And spinning the wheel is free, at many times per second.

The screen saver method is only one such implementation.  A simpler approach would be a batch file that ensures that the player has a Bitcoin wallet, and then runs vanitygen https://bitcointalk.org/index.php?topic=25804.0 coupled to a GUI that provides just the right amount of user feedback and "near misses", with a search file of the largest abandoned addresses, and the matches written to an output file.  By using just the first 8 characters of the target addresses, a small number of encouraging hits would be generated.

No they aren't.  It is like trying to get to another star and your progress is 0% so you decide to jump.  You get too feet off the ground and call that a noticeable improvement over zero.

Not really, it costs energy to do that.
Even if the computer is on anyway it would use more energy doing all the work.

We can sell it as a bitcoin robot

No but multi-sig would.  However the chance of success is ~0%.  You are far more likely to die before reading the end of this post then ever having a collision occur.  Still alive?  Good your coins are probably safe too.

No.  It is not like that.

You decide to jump, hundreds or thousands of times per second.  Most times you only move 0.01 inch, and occasionally you jump 46 feet, and occasionally someone posts documentation in a forum that they jumped 273 miles.

The number of encouraging results increases over time.  And "players" would get closer to a match.  And the value of the Bitcoins keeps increasing.

We would have a much harder time promoting something with a zero chance.  Unless we were a religion, offering "eternal life".

Yeah but promoting stealing will make them scared to keep bitcoins after

We should not promote stealing.  It should only target abandoned addresses.

Plus, it can't be stealing if they don't get an exact match.  And we have already stipulated that the chances are very, very, very small.

So, how do you know if an address is abandoned?

That is an excellent question.  Some people may be comfortable waiting 100 years before trying.  Just to be sure.

There goes someone's stash

Owners of addresses with a very large amount of Bitcoins in them, that have been dormant for years, should consider immediately moving them to multiple addresses with smaller balances in them.  This will demonstrate that the accounts are not dormant.

Unless they are confident that the chances of someone obtaining the private key are "zero".  You know, like jumping to another star.

Your computer would be burning electricity.  Do you CPU solomine Bitcoins?  That would be more cost effective.   You could make all kind of "fun" stats on how close to solving a block you got.  In the end it would just mean wasting an asinine amount of energy that has a ~0% chance of every providing a benefit.

I agree with you.

But for promotional and entertainment purposes ...

I can't think of a worse promotion then someone buying the hype running their jackpot generator for a month or two racking up $100 in electrical cost only to find out they have been duped and they didn't even have a quadrillionth of a percent chance of making a cent.


"Bitcoin is a scam you will never win anything".
"My son downloaded that Bitcoin thing and ran up $500 in electrical bills.  We can't afford this".
"Everyone is paying a lot so someone must be getting rich off this free lottery anyone have any ideas".
"Bitcoin destroyed my computer".

...are tons of ideas more suitable than "hey, you could empty a Bitcoin address" that sounds like Bitcoin addresses are insecure.

Which is a better deal mathematically?

• Buying lottery tickets for $1 each, with a $100 million prize
• Having unlimited free chances (maybe 1000 per second) to find a private key to a dormant Bitcoin account with millions of dollars in it

Did you even read our posts?
They are not free.

Yes.  But I think he grossly overstated the cost of electricity.  Don't you agree?

1. Gather a list of high-value bitcoin addresses. Or just use this one: http://bitcoinrichlist.com/top100

2. Dump them into a text file

3. Run vanitygen against the list of addresses: vanitygen -t "addresses.txt"

4. Wait billions of years

5. PROFIT!

You left out

Package it with a GUI

Have it require that a wallet be installed

Have it provide tantalizing feedback (like a slot machine or lottery ticket)

I appreciate your answer, but you have not provided any mathematical analysis.

I really tried find full Bitcoin address in vanitygen, but it showed no progress or even likelihood to find the Bitcoin address.

For purposes of this promotion, do not search for full addresses.  Search for the first 8 characters of each one on your list of dormant accounts.  As you have demonstrated, humans require positive feedback.

Waiting billions of years may be too discouraging.

*facepalm*

This is like trying to find a diamond with random digging with a shovel and then saying "For the purposes of this promotion, we will search for a normal rock instead".

I agree with you.  Using uppercase and lowercase letters, along with numbers is a remarkably compact way of representing "really really huge" numbers.

I don't think you managed to grasp how huge those numbers are.

It's more possible for one of us to Fuck Jennifer Anniston consentually

That could be our backup promotion.

I don't think you managed to grasp the point of my first post in this thread.

Use the regex option, otherwise it will complain that your prefix is too big.

vanitygen.exe -r ^ADDRESS_HERE$

Remember, the search is not cumulative, so you are always just seconds away from a huge jackpot. Don't lose hope! Waiting the first billion years is the hardest.

regex is too slow.  It would take zillions of years.

Truncating the address provides the needed positive feedback for the target audience.

I will let whoever reads this post now decide on that, I think we've answered your assumptions of it being a good advertisement for Bitcoin already.

It's more possible that Jennifer Aniston agrees to have every member of Bitcointalk simultaneously do that than it is brute force the key to an address.

Surely the OP is trolling at this point.

Fair enough, since the chances are greater than zero.

Brute-Force Concept #2

1. Create website that allows anonymous visitor to enter public key of target wallet (suggestions include suspected Satoshi address, FBI address, etc.) and public key of their "payout" wallet.

2. Visitor presses button to make 100,000 attempts to brute force the target wallet.

3. Visitor enjoys an advertisement while they wait for 10 seconds or so while the attempt is made.

4. If successful, coins are transferred to payout wallet, site shows garish WINNER!!! banner.

5. If not successful, site loads a new ad and invites the visitor to try again.

6. Site operator pays 10% of weekly ad revenue to wormbog for sharing this brilliant technique for extracting money from gamblers who suck at math.

Preferably a Jennifer Aniston ad.

http://www.adweek.com/files/adfreak/AdFreak%20new/Jennifer%20Aniston%20SmartWater%202.jpg

You should probably code in a couple consolation images with inspiring quotes or "Don't worry, your mom and Charlie Sheen still thinks you're winning" that pop up once every couple billion losing clicks, just to keep people interested.

Before I  go watch porn, let's agree the target shall be fbi address

A guy who used sha256 of poem written in exotic language as private key lost his brainwallet recently (http://www.reddit.com/r/Bitcoin/comments/1ptuf3/brain_wallet_disaster/), so yeah you could make your own dictionary, maybe someone thinks that nobody will think about using password directly as key or MD5 of passwd and you will be the one who will teach him a lesson about security practices. Be creative in choosing which keys to check and you might hit big. Of course properly generated private keys are safe (well until you use them with bad RNG (https://bitcointalk.org/index.php?topic=274050.0)).


That could actually work quite well. Or at step 2 propose something to be added to dictionary.

Highly doubt it, I bet it was a simple key logger or spouse

Great idea. Visitor can paste in a target address, OR enter a phrase to generate into a brainwallet to sweep. The system tries a set of variations on the phrase - different word order, some words missing, letters replaced by numbers, etc.

Once people realize that any bored kid with a computer can spend their spare time attempting to hack brainwallets, perhaps people will stop relying on them so much.

Say you know a particular vps was used to generate many wallets , and you know all the settings of that server (ram, HARDDRIVE, MAC address,etc), and rolled back your system clock to around the time that you knew it was generating its wallets. Would that increase your chances at all?

No.  Not unless they is a flaw in the PRNG.  The entropy pool is filled by chaotic events like pagefaults, mouse click intervals, cpu temp, etc.  The same user likely couldn't produce the same entropy pool if they tried.  Now if the distribution of the PRNG is flawed then one could increase their chances by focusing on the area with higher occurance but unless it is horribly flawed that would just give you an academic speed up at best (i.e. a planetary sized supercomputer needing "only" 1B years instead of 100B years).

Still on a more general level the security of a wallet any wallet is only as good as the PRNG so that should be an area of scrutiny.  If you are really concerned there are hardware RNG.

hmmm, I think I will start by running the entire script of the princess bride against the FBI wallet

That thread made me think of a little game:
https://bitcointalk.org/index.php?topic=326545.new#new

Come and beat me (while checking for a positive balance for each attempt) ;)

Let's check the math:

The "theoretical minimum possible amount of energy required to change one bit" is 1.38×10^23 J/K
That's according to wiki (http://en.wikipedia.org/wiki/Landauer%27s_principle) - Theoretically, room‑temperature computer memory operating at the Landauer limit could be changed at a rate of one billion bits per second with only 2.85 trillionths of a watt of power being expended in the memory media. From this, the energy required to count to 2x10^256 is 0.7x10^235 Joules and would take 2x10^247 seconds.

Total energy of the sun - estimated to be 1.3 x 10^44 Joules (as calculated here (http://curious.astro.cornell.edu/question.php?number=389))

Since the difference in power is 10^203, cooling that theoretical computer would do nothing (but a nice Sci-Fi bit there).

In fact, another wiki (http://en.wikipedia.org/wiki/Orders_of_magnitude_%28energy%29) says that the total energy of all observable Universe is 4x10^69 J.
For those who is interested the total number of atoms: 10^82.
That's just to show how HUGE 10^256 number is...

subcoin of course temperature matters.  It is right there in your wiki link.  Hint K is Kelvin a measure of temp.  The higher the temp that a perfect computer operates at the lower the efficiency.  It can't be more efficient than at absolute zero.

Ok, let's compare:

0.7x10^235 vs 1.3x10^44 -> at room temperature -> Difference of 10^191
or 0.2x10^232 vs 1.3x10^44 -> at 0 Kelvins -> Difference of 10^188

For this example, chilling to 0K does precisely nothing.

But funny thing, I just realized - in this example the time is the problem, not energy.
2x10^247 seconds is inconceivably long.

It isn't time or energy, it is time AND energy.  x J for y seconds is the same as 1000x J for y/1000 seconds.  

If you have a perfect computer and the ability to scale it to an arbitrary size your computation time is the inverse of the energy output. 10x the energy, 1/10th the time. 1/10th the energy, 10x the time.

For it to "only" take 2E247 seconds would require a sustained energy output of ~1 E 235 J however our sun doesn't have that much energy (nor that much time).
Using less (say the output of our star) would mean your throughput would have to be lower and the amount counted would be less in the next 5 billion years.
On the other hand if you had significantly more energy (just the output of a billion stars) then you could do it in less (still asininely long) time.

Just one thing, it seems you are doing it in base 10 and the image is counting in binary.
2^256

Ouch, you are right!
Opps... how do I convert one to another?

It seems, most of the things is measured in 10^x something..

Screwing around with the windows calculator, I obtained this result 2^256 ~ 1,16x10^77
Hope this helps.

'the jackpot' would be other peoples money.

theres 1 no owner bitcoin jackpot worth M

Yes, it does. Your calculations are correct (according to this wiki page (http://en.wikipedia.org/wiki/Power_of_two))
2^256 = 115,792,089,237,316,195,423,570,985,008,687,907,853,269,984,665,640,564,039,457,584,007,913,129,639,936

Back to original calculation:
1,16x10^77 / 1billion / 1 trillion * 2.3 = 2.67x10^56 Joules
That is, indeed, more than 12 orders of 10 of the total sun's estimated energy (1.3x10^44)

Conclusion:
The very pretty Sci-fy poster is correct - even with perfect thermodynamic computer it is impossible to just count 2^256 using all of suns energy.
Which answers the original OP's question: No it's not possible to build a wallet generator.

TA DA  ::)

why do people keep asking this?

ignorance

also see "bitcoin is tulips"

see also:
should make block times 10 seconds so tx are fast
should steal/recover unused coins
should blacklist "bad" coins
should demurage coins so people will spend them
deflation will destroy the world
21M BTC is not enough to go around
all coins will be mined by 2040
how can bitcoin work once mining stops
etc

Good idea, but don't you think that DPR would also have a salt? 

very nice summary D+T

thank you

This is not possible. Sorry to disappoint you.

The effect you're referencing is called 'normalcy bias', the general belief (however false, intellectually) that bad things aren't going to happen to us because they never really have.  The problem with this statement, is that statistics and mathmatics don't have a normalcy bias.  Whatever can happen, eventually will; but whatever cannot happen, will not.  While it's not actually impossible for a random address keypair collision to occur, the odds are astronomically unlikely.  The odds that a potted petunia and an adult whale were to pop into existence 20 miles above the surface of the Earth is about as likely as any single individual, or even all of us, have of forcing a single keypair collision within our own natural lifespans.

Short answer not really

Long answer isn't there something more fruitful you could be doing with your time?

Yes it's 100% possible to build a generator.
Yes it's possible to hit a jackpot with it , but chances are 0.00000... my fingers hurt... 1%

Second , people like to cling on that 2^256 , which is not quite right.
Let's assume for instance there are only 10000 combinations. And the most advanced supercomputer is doing 1/second :D.
John has password 1823 , Jerome 5674 , and Maya 7777.
Well , it didn't take 10000 seconds for their passwords to get cracked right?

Same with passkeys , some will be safe till the dinosaurs are hit again by an asteroid , thousands of them are cracked right this second.
Also , do you know the amount of passkeys is not equal to the amount of addresses? :D
Just a though about how secure your password might be.

That's incorrect. The private keys have so many possible combinations that it would take pretty much forever with the current processing power. That's as far as I'm acquainted with the subject.

It will take forever to ... what? Get them all? yeah!  Get a few of them ? it's a nanosecond job.

If it was that easy, someone would do it.

Still, there are only 2^160 possible addresses from the 2^256 pool due to RIPEMD-160.

Got to https://www.bitaddress.org (https://www.bitaddress.org) , bulk wallet and there you go... you can generate 10000 adresses.
Chance on getting one with some bitcoins hiding there , close to 0 , but it's still a possibility.
It is not impossible.

Sorry , I think I read above in this thread 2^256 and took that for granted without even thinking back.

i have been running a random number generator for over a week, over 100 hours and hit 1 wallet with $4.13. it might take a million years to target one address but there's not just one out there.
the only thing i have not done is log the private key that had cash in it the program just keeps running. but i will be changing that soon.

If you really hit an address, please make a big post about this and let people analyse it, unless there is some bad random number stuff / low entropy stuff like brainwallet involved this should not have happened.

It was just luck. I would be surprised if he finds another one within the next 1 billion years.

If someone wins lotto 4 times in a row that may happen, but you still want to analyse if there is any flaw in the system.
Especially the random number generator used in this case.

Care to share the generator for analysis?

Sorry for hijacking, but I have another newbie question.

https://en.bitcoin.it/wiki/Technical_background_of_Bitcoin_addresses

If 2 different hashes at step 2 produce the same hash at step 3 (which is theoritically possible, although very unlikely, I know)
Then the final base58 address would be the same, though the ECDSA keypair would be different.

What happens, then, if someone sends coins to this address?
I'm guessing both ECDSA key pairs would be able to spend those coins?

why would i do that.  

i started out thinking no way that this was going to work but then i had a hit. it might be just luck but its still running so i will wait and see.

edit its been running for 183 hours.

If there is some random number generator insecure, then this is seomthing that could costs people millions.
(and even worse for the unlikely case that Bitcoin addresses aren't secure enough)

Why? It's a potential security exploit in bitcoin. If people abuse it before it gets patched, it's over.

He just found a weak brain-wallet  ;)

Not by using a random number generator.

no

What's the public hash of the address you found?

Even with everything that everyone says about hashes, we did get a duplicate GUID one time where I used to work.  It caused major issues but from everything we could investigate, we really did get a duplicate.  It's the only place where I have used GUIDs where it has happened, but it did happen once.

Nobody believes me and I have been called a thousand names on forums for mentioning it, but I can assure you it happened to us once.

A SHA256 collision is billions of times less likely than duplicate GUIDs.

Yes.

Correct. 

GUID is 128 bit.  If you generated so many numbers that you had a GUID collision every second you would on average only have one 256 bit collision every 5,395,141,535,403,007,094,485,264,577,495  (for the record that is a trillion times longer than the universe has existed).

In fairness, SHA-256 isn't used to create bitcoin addresses, it's used to secure the blockchain.  The current address algo is 168 bit, IIRC; but if it ever looks to be at risk, it can be upgraded.  The first character of the address is what denotes the address version.  The primary reason that there is no 2xxx... address types yet is because there isn't really a better algo to migrate towards.

Well in fair fairness it is 160 bits; the checksum doesn't add uniqueness.   Still 160 bits is 4 billion times less likely than a 128 bit one.  The probability of a collision with a specific key (first preimage attack) is even less likely.

Thanks!


If you're talking about the question I asked, I think we should quit using "attack".
This is clearly not a valid nor sane way to "attack" bitcoin, as ROI would be ridiculously low compared to mining with the same processing power.

Can we just replace "first preimage attack" by something like "first preimage hazard" or something sounding even less likely?

And back on topic, There are currently just a little more than 11 millions addresses with unspent outputs.

Finding one by running a random generator looks less likely than winning national lottery 3 times in a row.

And currently there are only just over 11 million BTC in circulation, so the average address has about 1 BTC.  If the odds of such a thing occuring to you, by random bad luck or otherwise, simply divide your massive bitcoin wealth up into a few addresses.

Not correct. If you take out of the equation top ~10 addresses, the per address amount for the remaining ones is much less.

But I did not take the top anything addresses out of the equation, so the average win should an address collision ever succeed would be 1 BTC right now.  If someone is trying to collide with one of those top 10 addresses in particular, then the odds are nominally zero before the heat death of the known universe.

Okay then, mister heat death.

People have tried, and are still trying to bruteforce an address with funds in it. To my knowledge no one has done it yet

Nor is that likely to occur unless and until either the breakers aquire a 150+ qbit quantum computer or a serious flaw is discovered in the address algo.  Even if both were to occur, the odds of succes aren't particularly high.

Yes, there is a chance.

I know. Its a reference to 'dumb & dumber'.

Lloyd: What do you think the chances are of a guy like you and a girl like me... ending up together?
Mary: Well, Lloyd, that's difficult to say. I mean, we don't really...
Lloyd: Hit me with it! Just give it to me straight! I came a long way just to see you, Mary. The least you can do is level with me. What are my chances?
Mary: Not good.
Lloyd: You mean, not good like one out of a hundred?
Mary: I'd say more like one out of a million.
[pause]
Lloyd: So you're telling me there's a chance... *YEAH!*

There is a chance that you're actually a super intelligent alien with IQ of 1,000+. There's a chance for pretty much anything, but it doesn't mean that it will ever happen.

"Inventions have long since reached their limit, and I see no hope for future improvements."
   -- Julius Frontenus, 10 A.D.

True enough, but bitcoin also has methods for upgrading both the address algo and the blockchain algos while maintaining backward compatibility and without a need to even pause the running network.  If there is even a credible threat, bitcoin can change to address that threat, whether or not most people desire to remain with their old address style/algo.

L.O.V.E. IT!

-Minty

The Android RNG was flawed - it was announced across the top of this site that it was and that anyone who had a wallet generated from an android with the affected software should immediately make a new wallet and move their coins there. If people didnt' pay attention to that, they could be in for some pain. I'm a little surprised that more attacks haven't occured, honestly