Bitcoin Forum

Ok simple question, It’s my understanding that SHA256/MD5 and any other hash cannot be reversed hashed..?
So… What are all these ASCI/video cards hashing/hacking.. ?
Main reason I am asking was talking to someone today and was telling him about Bitcoin and what was going on but when it came to the hashing part I was stumped. Didn’t understand enough to explain it to him.

All that I learned from bitcon and how I got started was from this podcast  http://twit.tv/sn287.

Lastly one other thing if anyone say 15 years from now wants to start mining solo/other and they need the bitcoin block chain. Can you imagine how much data is going to be downloaded? Let along say 50+ years from now the storage/bandwidth needed just to get started to mine.

Thanks for the info/help.

I have always been a little fuzzy on this myself. I would love it if someone could explain this.

This has been spoken about many time and can be found in the forums.

You are mining blocks of transactions.

Would you like to know more?  ;)

https://en.bitcoin.it/wiki/Block
https://en.bitcoin.it/wiki/Mining

Bitcoins.

What a miner is doing is generating hashes and trying to find one that is lower than the current network difficulty.

A miner gets all unconfirmed transactions (that have paid correct tx fees!) and packages them up into a "block", this block now needs to be inserted into the "blockchain" with all the other blocks of bitcoin tx's.

To get the block into the blockchain the miner has to solve a mathematical puzzle. It gets the hash of the data in the block, plus a 0 (nonce) and checks if its lower than the difficulty, if it is, then it has solved the puzzle. If it isn't lower than the difficulty, then it increments the nonce, so this time it gets the hash of the data in the block, plus a 1, which gives a completely different hash.

It repeats this and keeps increasing the nonce for a while until it solves the puzzle, then it broadcasts the solution + the block and gets 25BTC + all tx fee's in the block as a reward, if the miner is a pool then this reward will be split between all the pool members.

It's pointless busywork designed to scale according to how many other people are doing the same pointless busywork.

This. Unfortunatey, proof of work concepts for bitcoin are hard to come up with for useful work and we are currently using a modified version of Adam Back's Hashcash to delay people from getting too many bitcoins (his is used to delay people from spamming emails and comments primarily).

Luckily, Bitcoin is not yet a consolidated effort of the brightest minds in the world so there is plenty of room for vital improvements (including removing the proof-of-work requirements of hashing blocks). Unless I'm mistaken, it's also possible to find a solution through Open Transactions.

It is not pointless, it serves to process the transactions in the bitcoin network so that everybody can say with certainty which order transactions were posted so anybody can verify their own transactions.

If you do not see why we need mining, it might be helpful to read the original bitcoin whitepaper by Satoshi.

Just 8 little pages:

Bitcoin.org/bitcoin.pdf

You might want to actually read it yourself; yes the round winner will process the aggregated transactions but the actual hashing is solving nothing in particular, though a scaling nothing in particular.

The whole network could be run on a NES if it was only about the transactions.

But its not about the transactions, its about the trust.

Bitcoin is a human project, not a machine. Psychology is what matters.

Modern currency is fiat. It exists only through the trust of its users. But Wall Street and the Square Mile are not cheap to run. So bitcoin can afford its overheads. But as for the future, sci-fi writers could come up with some interesting dystopias based on feeding the unsatiable maw of the great deity BTICNOI.

The network throws out a transaction every 3 seconds on a good day. I could almost run that by hand with enough coffee in me.

As for the rest of what you're saying, what? Psychology doesn't change the fact that the network is doing absurd amounts of pointless busywork to decide you gets this round's shiny coins.

Mining gold is pointless busywork; we already have all we'll ever likely need for industrial production, but people continue to do it because it has value.

I consider most of what goes on in Washington, D.C. to be pointless busywork.

The difference of course is that we don't purposely pour rocks and dirt over top the gold, put on a blindfold while digging and spin around 3 times in a circle to make it more difficult to retrieve and thus claim it's more valuable, we simply dig when we believe there is gold to be found. We understand that if you agree with bitcoin in it's entirety and have drank the coolaid, that "mining" coins is necessary and thus acceptable. Bitcoin's code however is not a law of nature but a constructed set of rules that can be changed with consensus, leaving absolutely no excuse not to find more intelligent and less wasteful replacements for proof-of-work concepts later on.

I admit it; Cultist defenders of energy waste are a sort of pet peve of mine.

Then search we shall; meanwhile, consensus still maintains that there isn't currently a better way, all things considered.

You're right, we do it with diamonds, not gold.

Uh, no. There's is a tangible relation between the actions going in and the results coming out.

You're not doing a bunch of sodoku puzzles while hoping that this time you'll be the lucky one who gets to process the transactions of the block.

It's pointless busywork because all that computing power doesn't do shit except try to win the round, not process the transaction, and that would happen whether it's only a single NES running the network or the sum total of every electronic device made by man.

You completely miss the point. Mining is what miners do to prove(albeit usually without their awareness) to the Bitcoin investors/users their commitment to maintain and secure the payment network,  the bitcoins you mined is the proof itself, every payment network requires substantial infrastructure investment, you can't build a Paypal/WU competitor with no solid hardware.

http://upload.wikimedia.org/wikipedia/commons/thumb/b/b2/NES-Console-Set.png/250px-NES-Console-Set.png

http://buttcoin.org/wp-content/uploads/2013/03/nes-bitcoin-miner.png

Bitcoin miner capable of handling the entire Bitcoin network workload

No, you're missing the point; yes it's kinda required to establish who will be processing the transactions for the round, but the actual computation going on is nothing more than hashing against a random difficult but easy to prove puzzle, hence pointless busywork.

"maintain and secure"=protection against sabotage=beat any malicious third-party to it.

The hashchain itself is what tells you that the ledger is the result of network consensus, if you can't do it anyway else more efficiently,  then it's not pointless busywork, because it serves a useful purpose in the most efficient way, or you would be essentially saying all cryptographic works(at least those inlvoving hash functions) are pointless busyworks.

The dreaded 51% attack? That's not viable anyway. Fork off, kick out the responsible peers, resume mining on the 49% fork, 51% fork devalues, done. Whooppdeedoo.

Now if you want to kill bitcoins, there are way more viable ways to do it. Like cutting off the arms ears eyes and mouths. Which is coincidentally exactly what's been starting to happen these last few weeks.

http://codinginmysleep.com/bitcoin-mining-in-plain-english/ is a good explanation.

Had Bitcoin not being designed this way, it would have been killed by a different kind of 51% attack in its early days, Bitcoin doesn't only need to face threats of governments, but also that from botnet operators, random script kiddles etc. In fact, that we are dealing with government pressure itself is evidence enough that the system worked. Also, 51% attackers can follow you wherever you go, however many times have hard-fork.

So you have 2, 3, maybe 4 hard-forks which brought the attackers exactly nothing. How many more times would they then try after that? "Dude, stop trying to hit that ghost with your sword. It went clean through doing nothing the last 5 times!" - "Maybe if I swing extra hard next time."

With the current design, the only entities that are powerful enough to do a 51% attack are determined groups with vested interest in the old system(govs/banks, etc), so it's possible they will keep trying despite the cost. But again, the more important point is-what would have happened had it not being designed this way?  Some script kiddles in control of several thousand zombie machines could have subdued the whole network.

I give up. It's like talking to a wall of ASIC miners. "The 51% attack is not as problematic as it's made out to be, even if it occurs." - "But skript kiddies could do a 51% attack and what then?"

No, I was talking about the network in its early days, not right now, and if we give up on proof of work, it will shrink to what it was like in its early days.

If you don't use proof-of-work, a possibility is to use one-IP-per-vote, a script kiddle in control of zombie machines could then very easily pull off a double spend attack under such a system.

originally the POW system was created to account for increases in Hardware advances and improvements in hashing algorithms to maintain the 10 minute average to prove the time the transaction took place. The further the average gets away from that 10 minute mark, the bigger the number to be hashed; So when someone gets the hash all the transactions that they have received becomes the next block.

A double spend can be done during those ten minutes if someone accepts the transaction, but will disappear the moment the first transaction is spotted in the chain with insufficient funds, so a double spend can be done with 0 hashing power on a modified client that accepts 0 confirm time. you would have to have a lot of bitcoins to make it worth your while to pull off a double spend attack with a average of 6 blocks deep cause it's going to cost you 20 million in ASIC hardware and several million more to build the VPN infrastructure to do a successful attack at the moment, which will be an easy subscription soon to cause the necessary global blockchain penetration necessary.

to do a successful double spend attack would be extremely costly, a commited attitude to construct the infrastructure, proper distribution of ASICs to points of attack worse of all it would require multiple splits in the blockchain that would be destroyed the moment the network shutdown, that is a long time to keep it going waiting for delivery of goods and most multi-million dollar transactions rquire some form of ID and lots of surveillance, you'd be a wanted criminal ASAP.

So here's another set of questions:

1 - What is the average daily total volume of transactions in BTC?

2 - What is the total fees generated for the network from this and as a percentage?

3 - What is the cost of electricity for doing this?

I don't know the answers to this but would be interested, as this is the key to the future of BTC processing, once the incentive (and ability) to mine new coins falls away over time.

These numbers are way too high.

ASICMiner raised 163k external shares at 0.1BTC = 16,300 BTC at a then price of $15 per BTC = $200k total funds

Even allowing for the fact that they undoubtedly sold some of these BTC later at higher prices, the total cost to produce an ASICMiner equivalent currently would be well under $1m. Obviously the talent required to do so and the time of delivery is not clear, but the point is that if you simply set out with a plan to be able to take 51% of the network, then it would take a lot less than $20m......!

3. Last estimate I saw: 150000 $ US per day

\

according to blockchain.info

1: 28 million dollars every 24 hours are transferred worldwide

2: 5200 US dollars in transaction fees that is .96% of the miners revenue of 550,000 US dollars
   that's 5200USD for 56 thousand transactions valued at 28 million so that is .0002 of the total value.

3:324,000usd per 24 hour period,


but you have to remember the fee system has not been eased in yet, so we will get a taste of the future once we see the transaction limit being met, after that it's anyone guess,
no transaction limit? a market concensus will be created with ASIC farms charging more for their services than anyone with a lesser system, basically verification centers.