Bitcoin Forum

is it possible for the cryptographic function to break?

SHA256 is also what secures your online banking.  Do you think online banking will die out?

short answer:  yes

long answer: just fyi "breaking" could mean a few different things in this context. i assume you mean that an attacker has fully mapped SHA-2 so that he can determine your private key and spend your bitcoins.

anyone that could "break" SHA-256 right now could win a nobel prize and would likely launch a new branch of cryptography.  it is just not feasible with our current knowledge of mathematics, quantum computing, and cryptography.  in 5 to 20 years it could be possible, but if an attacker developed such technology, it would likely to be much more profitable to use it to mine than to search for the private keys of publicly known wallets.

with respect to bitcoins:  satoshi anticipated the "breaking" of SHA-2 and outlined how bitcoin would adapt to a new cryptographic function before SHA-2 could be feasibly broken.  the core bitcoin protocol doesn't really care which cryptographic algorithms is utilized, and the algorithm can be changed without disrupting bitcoin's functionality.

that's interesting ive been wondering that for a while now, you seem like you know quite a bit about bitcoins so I have another question, when a block is fully mined [releasing 25 coins] does that mean that the other people have to start on a totally different block, that is if three pools are working on a block and pool #1 solves block does that set pool 2 and 3 back?

Generate a valid hash (large integer), get random data from top block, mix and combine random data until you find your valid block, done.  Good Luck!

Just try and let us know  :D

yes, that is correct.

There is no "set pool 2 and 3 back". Back from what? Also a block "fully mined"? Either it is mined or not. There is no concept of "working toward a block". Either you find one or you do not.

"Chance" based on hashrate. Pool 1 finds block first, then that is the block. It's a done deal. Next block "approximately" 10 minutes away. Existing pools with existing hashrate have the same chance to find block 2 as they did block 1, but pool 1 found block 1. No one was "set back", they're still hashing away and on average will break a certain amount of blocks per x time.

Breaking BTC?
- breaking SHA-2 (low possibility)
- 51% attack (it is still possible but after a year or two difficulty will be too high. Someone would have to invest billion of dollars to buy necesary equipment)
- goverment intervention - closing all the exchanges available. Imagine what happens next: price would drop to pennies, btc would be again a toy without assigned value to traditional currencies.

yes

The current cryptography including sha-256 will surely be broken someday, maybe in the next few years, but I think the cryptocurrency idea with all the advantages and features it has over our current payment systems means it will carry on in some form.
Bitcoin will have to evolve if it wants to stick around as the digital currency of choice.

Yes
But I think before SHA-256 function is broken most of governments countires will issue laws to virtually kill bitcoin currency.
If Bitcoin volume is enough can challenge monetary politics creating a parallel uncontroled value chain.
Thailand started some of this mesures.

No, and by the way, today Germany recognized Bitcoin as a private money (which goes in the opposite direction some user predicted)..
See this: http://www.telegraph.co.uk/finance/globalbusiness/10252383/Germany-plans-tax-on-bitcoin-after-virtual-currency-recognised-as-private-money.html

Good news ( for Bitcoin recognizance ) but think main reason is to try to collect additional taxes and not to help Bitcoin grow.

I'm not knowledgeable about the technical side, but I keep seeing more and more signs that Bitcoin is going mainstream, and therefore, here to stay.  I personally think it's a good idea to own Bitcoin, even if you can only accumulate it a little bit at a time (http://creativecurrencies.cu.cc/your-complete-crypto-crash-course/).

Bitcoin and its sha256 clones mostly use this (sha256(sha256(ripemd160)) Gavin has said that if this is broken they will patch it but he has also said that the security of the function is not as important as most people think and it could use a simpler function and still remain secure due to most of the weakness on a function like sha256 is to do things like extend length, reduced rounds etc. but in the bitcoin system(wallet/network) these types of attacks are not a main threat.

things like malware that may steal your private key or a weakness in the implementation of the ecdsa public/private key pair e.g the java/android latest exploit show some of the real threats, also note that the double sha256 is not more secure than a single sha256 in fact it has 1 more possible collision but at 2^255 it is safe as a function but not necessarily safe if implemented using what the programmer is told is safe but may turn out to be unsafe e.g "SecureRandom" or a bug in say one of the library's e.g openssl.

how a bug is solved in the network is quite interesting, bug is found, fix is made and new wallet released (bug in network not fixed yet) and once most of the network nodes have updated the network converges and goes with the majority e.g the new wallet and the bug is solved mostly.

I think the biggest threat to bitcoin is from goverments (is bitcoin a Napster or an iTunes?) or due to the the load on the network e.g block size 1Mb (can only fit so many transactions in it) and the slowness of the sha256 function itself, Oh what about litecoin e.g scrypt based alt coins well they are different but very slow and still use sha256 and the 1Mb block (sha256(sha256(scrypt)) and overall the tps (transactions per second) are lower than the main bitcoin network.

I think bitcoin will be around for some time what will be a real game changer is if someone (maybe me) release a real new coin that uses a different algorithm for the tree hashing method(means writing a new protocol) and a more secure function for the public/private key pair.

some of the new hash functions that have been proposed as candidates for the sha3 or their derivatives could be used to replace sha256, I would use a faster algorithm as it would increase the tps, and newer methods for key pairs like MQQ(Multivariate Quadratic Quasigroups) but that would be quite a bit of work ::)

most people that are developing new coin are not even taking this into account they just find and replace a few names, create a new root hash and release the next CrapCoin which is fine for learning but in the long term most clone alt coins will fail mainly due to unstable low network hash rate.

This is just my view on the current situation but I do hope that it is helpful to someone  ;D    
  

It is possible for mathematicians to find weaknesses in the SHA-2, RIPEMD-160, and ECDSA algorithms that bitcoin uses.  It is extremely unlikely (for all intents and purposes you'd be safe to consider it impossible) that significant enough weaknesses will be secretly and suddenly discovered in all 3 algorithms simultaneously by someone who intends to use the information for fraud and theft.

As such, bitcoin won't "die out" just because someone finds some weaknesses in some of the algorithms.


Your private key for any address that has only received bitcoins and never sent them is protected by all three algorithms (ECDSA, SHA-2, and RIPEMD-160).  A SHA-2 algorithm that is broken will allow miners to mine much faster, but it won't break bitcoin and it won't allow anyone to determine your private key.  Once you send bitcoins from an address, and future bitcoins that are sent to that same address are only protected by ECDSA.  As such a broken SHA-2 will have no effect at all on those addresses since the public key is already known to the world.


You need to do some reading on how the mining process works before you start making guesses at the effects of a solved block.  You have some very common assumptions about the mining process that are incorrect.  It is much like rolling dice.

Hand 10 six sided dice to 100 people and ask them to start rolling those dice all together as fast as they can.  Any time someone rolls 8 sixes in a single roll they are rewarded with a "block", and then the rolling continues.  When one person rolls 8 sixes, have they set back the other people at all?

In the case of mining, the miners are each essentially rolling a single 2²⁵⁶ sided die.  Whenever a miner rolls a number that is lower than the current target difficulty they are allowed to publish the block that they are attempting to create, and the rolling continues.


No, you are giving out bad information.

The cryptographic functions can be replaced with better ones if SHA is ever broken.

Does anyone know where I can find a calculation of the likelihood of a collision of two randomly generated private keys?

if it was truly random generated then it is simply the size of the bit space of the hash function 2^256 for a single sha256 over ecdsa then - any short cuts the devs have taken to produce those keys in a realistic time frame e.g limiting the curve to reduce the time taken to generate the said key pair, and you would produce a pair not just a private key.

but I think the issues is that no computer can really do true random its always pseudo random like with the "SecureRandom" function in java/android and this has shown that a functions weakness is not in the bit space its in the way it has been implemented

read some of this to get a better idea as this has been asked quite a few times on this forums in more than 1 post/section Gavin has also talked about collisions https://bitcointalk.org/?topic=62.0

but like I have said its not about the function and bit space as 2^256/2^255 is quite secure and if brute forced would take longer than the dinosaurs walked the earth to create a collision

Quote: molecular
"take an average of 1,618,542,460,620,902,128,345,579,373 years to generate a collision"

it is more about how the devs have used said function and if that function has any exploits or weakness that would allow an attacker to shortcut this time

my apologies.  i was just trying to point out that the other miners previous work does not give them advantage over the publisher of the block when a new block is generated.  but to get more technical, isnt the hash function SHA256(SHA256(Block_Header)).  so doesn't that mean that everyone's "dice" effectively changes every block and thus everyone starts over?

This is true.  There is no progress ever made toward solving a block, so the amount of time spent working on a block in the past gives no advantage or penalty towards the work that will be done now or in the future.


Yes.


Everyone starts over with every roll they make regardless of whether there is a new block or not.  The dice don't remember the past, and neither does the SHA256 function.