Topic: When you start working on a block, why others won't do same thing? - page 2. (Read 3743 times)
May 26, 2011, 09:34:54 PM
wait, so If I have 2 cards in my computer, working as 2 separate workers, are they basically competing against each other? That is, are they doing the exact calculations at the exact same time, or will they be doing different/random calculations in an attempt to find the block?
May 26, 2011, 09:29:14 PM
So when one person in China finds block number 700000, how is the p2p lag between china and america dealt with such that someone in Texas stops working on that 700000'th block and moves on to the next one in the series?
Or does it even matter?
If it doesn't matter, and only based on local difficulty in the p2p network, then are there situations where there are pockets of lower difficulty in the p2p network where some people can still mine for coins easier than someone on the other side of the world?
Or does it even matter?
If it doesn't matter, and only based on local difficulty in the p2p network, then are there situations where there are pockets of lower difficulty in the p2p network where some people can still mine for coins easier than someone on the other side of the world?
If they don't find a block by the time they hear about China's block, they eventually stop working on it because they know it's not the longest chain.
If they do find a block before then, they publish theirs, but only the people who haven't already received the other block start working on a block on top of that one (the others have already started working on China's block).
Whoever finds a block first publishes it, and whichever block was "underneath" that one wins the previous contest. If the same thing happens again, then it continues to a third "round", until finally one of the chains is definitively the longest. This will happen even in high-latency networks because once one "side" gets much more than half of the network behind it, the split will even itself out.
In general, honest nodes will have included all transactions in those blocks anyway - just a different address gets the generation fee.
May 26, 2011, 09:16:29 PM
So when one person in China finds block number 700000, how is the p2p lag between china and america dealt with such that someone in Texas stops working on that 700000'th block and moves on to the next one in the series?
Or does it even matter?
If it doesn't matter, and only based on local difficulty in the p2p network, then are there situations where there are pockets of lower difficulty in the p2p network where some people can still mine for coins easier than someone on the other side of the world?
Or does it even matter?
If it doesn't matter, and only based on local difficulty in the p2p network, then are there situations where there are pockets of lower difficulty in the p2p network where some people can still mine for coins easier than someone on the other side of the world?
The first person the network successfully confirms as completing the block, gets the 50 coins.
So if china finishes a block 2 minutes before the USA, but there is 5 minutes lag time between the network, the USA will get the 50 coins (or the person who finished the block).
May 26, 2011, 09:14:41 PM
So when one person in China finds block number 700000, how is the p2p lag between china and america dealt with such that someone in Texas stops working on that 700000'th block and moves on to the next one in the series?
Or does it even matter?
If it doesn't matter, and only based on local difficulty in the p2p network, then are there situations where there are pockets of lower difficulty in the p2p network where some people can still mine for coins easier than someone on the other side of the world?
Or does it even matter?
If it doesn't matter, and only based on local difficulty in the p2p network, then are there situations where there are pockets of lower difficulty in the p2p network where some people can still mine for coins easier than someone on the other side of the world?
May 26, 2011, 09:02:09 PM
due to the way a hash function works (a property called the avalanche effect. google it if interested), every time the nonce gets changed, the hash output changes dramatically to effectively a random number.
Not so different to MD5 if anyone else is wondering.
exactly like MD5. MD5 is a hash function and a popular one. it's not suitable for security use anymore (collisions can be found), though it's just fine for corruption detection.
We just use SHA-256 instead, which is far stronger and should remain so for the foreseeable future.
Yeah, I was just saying that MD5 uses the avalanche effect aswell.
May 26, 2011, 07:20:06 PM
I think this might be a fair analogy:
1) In your hand you have 10 quarters
2) Toss the quarters into the air
3) Are all 10 of them heads up? If no, try again. If yes, collect 50BTC.
After doing this procedure 10 times, you aren't any "closer" than you are if you do it 100 times. So there's no progress towards it. All that can be done is doing the flips as many times as possible in the shortest amount of time.
The difference is that instead of 10 quarters, you're working with millions, maybe billions of quarters. And you don't need them all to come up heads, only a set number of them (i.e. the difficulty).
1) In your hand you have 10 quarters
2) Toss the quarters into the air
3) Are all 10 of them heads up? If no, try again. If yes, collect 50BTC.
After doing this procedure 10 times, you aren't any "closer" than you are if you do it 100 times. So there's no progress towards it. All that can be done is doing the flips as many times as possible in the shortest amount of time.
The difference is that instead of 10 quarters, you're working with millions, maybe billions of quarters. And you don't need them all to come up heads, only a set number of them (i.e. the difficulty).
due to the way a hash function works (a property called the avalanche effect. google it if interested), every time the nonce gets changed, the hash output changes dramatically to effectively a random number.
Not so different to MD5 if anyone else is wondering.
exactly like MD5. MD5 is a hash function and a popular one. it's not suitable for security use anymore (collisions can be found), though it's just fine for corruption detection.
We just use SHA-256 instead, which is far stronger and should remain so for the foreseeable future.
due to the way a hash function works (a property called the avalanche effect. google it if interested), every time the nonce gets changed, the hash output changes dramatically to effectively a random number.
Not so different to MD5 if anyone else is wondering.
nonce=0
while (block_hash>target)
{
block_hash=hash(hash(transactions), previous_block_hash, nonce, (other stuff not relevant to this discussion))
increment nonce
}
basically, this is what every miner is doing, over and over, millions/billions of times per second. we're not coordinating on this, we're just running this over and over looking for a block with a hash lower than the target. we only talk to others if we find a block hash that meets that criteria or someone tells us that one was found.
due to the way a hash function works (a property called the avalanche effect. google it if interested), every time the nonce gets changed, the hash output changes dramatically to effectively a random number.
while (block_hash>target)
{
block_hash=hash(hash(transactions), previous_block_hash, nonce, (other stuff not relevant to this discussion))
increment nonce
}
basically, this is what every miner is doing, over and over, millions/billions of times per second. we're not coordinating on this, we're just running this over and over looking for a block with a hash lower than the target. we only talk to others if we find a block hash that meets that criteria or someone tells us that one was found.
due to the way a hash function works (a property called the avalanche effect. google it if interested), every time the nonce gets changed, the hash output changes dramatically to effectively a random number.
No, everyone is working on the same block at the same time
Well, everyone is working on the same numbered block, but each person is trying to make a block that pays BTC to them (or their pool).
I don't understand
It's like rolling a giant die with billions of sides (numbered 1 to billions), and trying to get it to land on 42. The chance is the same each roll.
Wining in the lottery. Not a job done.
In other worlds, in theory - you can solve the block from first attempt.
In other worlds, in theory - you can solve the block from first attempt.
No, everyone is working on the same block at the same time, it's just the first person to finish it gets the 50 coins, (if it's a pool, then it gets split up).
AFAIK anyway.
AFAIK anyway.
There's no progress toward a block, so 'finishing' is a bad word to use.
I don't understand
No, everyone is working on the same block at the same time, it's just the first person to finish it gets the 50 coins, (if it's a pool, then it gets split up).
AFAIK anyway.
AFAIK anyway.
There's no progress toward a block, so 'finishing' is a bad word to use.
No, everyone is working on the same block at the same time, it's just the first person to finish it gets the 50 coins, (if it's a pool, then it gets split up).
AFAIK anyway.
AFAIK anyway.
if after you start working on it, others won't be able to work on it, how long can you do that?
If you machine need 10 years to finish that, that means, the block belongs to you for 10 years (others can't working on that?)
If you machine need 10 years to finish that, that means, the block belongs to you for 10 years (others can't working on that?)
Jump to: