Please run a full node

[franky1]

Well, they would realize that all merchants are simply locked out of all transactions, unless merchants use the new rule software on their full node, or connect their wallet to a miner node.  Because there aren't any other transactions in accepted blocks.

So yes, miners can't cash out their coins, as long as ALL USERS decide not to upgrade, and accept being locked out of bitcoin, their funds and their transactions at the same time.

1. pools are not going to waste 16 hours to double prove what they learn in 3-seconds to 10 minutes.. which is they cannot spend funds
2. pools are not going to waste days/weeks/month to really trying to push it in the HOPE that nodes download a MD5 hashing rule implementation..
after all even core suggest it would take a year to get clear node acceptance..

reality is..
if there was some secret cartel of "lets make md5 blocks for the next 6 months to force nodes to be less secured"
initially 20 pools have that motive.. but soon enough a pool gets greedy and jumps back to sha256 and wins every block.

[1715242059]

1715242059

[1715242059]

1715242059

[1715242059]

1715242059

[dinofelis]

so pools wont continue making md5 hashed blocks for 16 hours because 1 pool will see a nice easy income by switching back to sha256 and win every block that is spendable to the merchants and have no competition

Nope, it won't win anything more than if it were keeping his agreement with other miners, because it has only a fraction of the hash rate, and can hence only provide just as many blocks as it would when with his peers.  In other words, that pool is betting on making a very short chain, with much less PoW than the rest of his peers, and breaking his agreement.

In other words, if this pool has 10% of all the hash rate, when his peers have mined 90 blocks on the new chain, he will have mined 10 blocks on the old chain.  Most merchants and exchanges will not see their transactions on this small chain because the blocks are too full.  So merchants and exchanges would still be locked out of bitcoin for 90% - while they would be running entirely NORMALLY if they simply upgrade their node to the majority hash rate of the miners' rules, and see all their transactions.

And as I told you, you are now considering a hard fork, so in the end, you are considering the power game of a hard fork, not the power which you pretended, came from full nodes, so you are now building an argument for a statement that is not what you pretended.  But moreover, who is going to bet on this small chain with less PoW ?  Who is going to be confident in the transactions on this short chain with minority hash power ?

[dinofelis]

if there was some secret cartel of "lets make md5 blocks for the next 6 months to force nodes to be less secured"
initially 20 pools have that motive.. but soon enough a pool gets greedy and jumps back to sha256 and wins every block.

Your example of MINERS changing the hash algorithm is of course somewhat strange, because they would kill their own hardware.  But if you want to prove the power of full nodes, you will have to leave aside hard forking.  Hard forking will be decided in the market ; NOT by full nodes.  So this is a straw man argument.

Try rather to argue how full nodes would impose MD5 mining upon miners who continue to use SHA-256. THEN you will prove the power of full nodes over miners !

[franky1]

Nope, it won't win anything more than if it were keeping his agreement with other miners, because it has only a fraction of the hash rate, and can hence only provide just as many blocks as it would when with his peers.  In other words, that pool is betting on making a very short chain, with much less PoW than the rest of his peers, and breaking his agreement.

In other words, if this pool has 10% of all the hash rate, when his peers have mined 90 blocks on the new chain, he will have mined 10 blocks on the old chain.  Most merchants and exchanges will not see their transactions on this small chain because the blocks are too full.  So merchants and exchanges would still be locked out of bitcoin for 90% - while they would be running entirely NORMALLY if they simply upgrade their node to the majority hash rate of the miners' rules, and see all their transactions.

you have no clue..
now you are meandering into trying to argue about hash power.. yet you dont even understand hashpower either

you are presuming if it takes pool A 10minutes.. then it would take pool B 20 minutes, pool C 30 minutes.
that is not the case.
pools B and C could have found a block just SECONDS later.. but because there is only 1 winner. no one cares about the runners up timing.

if you take 1 pool away its not going to take 20 minutes to make a block. it can still take 10 minutes average block, just less competition so that the runners up now become winners more often, without affecting the average time much

i think its time you go to a shop and get some dice.. and some friends and family and play out some scenarios of randomness.. and see some real world scenarios play out.. it will surprise you

take your 10% of network hash scnario for instance
buy 100 dice..

get 10 people and give them 10 dice each. and ask them all to keep rolling until they get a total of lets say 600(adding each roll)
at very luckiest 1 person MIGHT get it in 10 rolls.. at unluckiest 1 person might get it in 60 rolls.

what you wont find is that if after playing the game for 2 weeks you found out the average took 20 rolls .. does not mean
if you took one person away it would average 40 rolls,
took one person away it would average 60 rolls,
took one person away it would average 80 rolls,
took one person away it would average 100 rolls
took one person away it would average 120 rolls
took one person away it would average 140 rolls
took one person away it would average 160 rolls
took one person away it would average 180 rolls
took one person away it would average 200 rolls

you would see the average would still be ~20 rolls. but now there are less people competing. so other people win more often.. and getting the result just miliseconds/couple roll variance before the other

[dinofelis]

Nope, it won't win anything more than if it were keeping his agreement with other miners, because it has only a fraction of the hash rate, and can hence only provide just as many blocks as it would when with his peers.  In other words, that pool is betting on making a very short chain, with much less PoW than the rest of his peers, and breaking his agreement.

In other words, if this pool has 10% of all the hash rate, when his peers have mined 90 blocks on the new chain, he will have mined 10 blocks on the old chain.  Most merchants and exchanges will not see their transactions on this small chain because the blocks are too full.  So merchants and exchanges would still be locked out of bitcoin for 90% - while they would be running entirely NORMALLY if they simply upgrade their node to the majority hash rate of the miners' rules, and see all their transactions.

you have no clue..
now you are meandering into trying to argue about hash power.. yet you dont even understand hashpower either

you are presuming if it takes pool A 10minutes.. then it would take pool B 20 minutes, pool C 30 minutes.
that is not the case.
pools B and C could have found a block just SECONDS later.. but because there is only 1 winner. no one cares about the runners up timing.

if you take 1 pool away its not going to take 20 minutes to make a block. it can still take 10 minutes average block, just less competition so that the runners up now become winners more often, without affecting the average time much

You have visibly a fundamental misunderstanding about mining blocks.  

If you have hash power that is so that, with a given difficulty, on average, you find a good block, say, every hour, which means that you have about 1/6 of the total hash power *when the difficulty was determined*, then it doesn't matter whether others are mining or not, you will win, on average, one block every hour - minus those few seconds that you were mining on the wrong block each time.

What does it mean, having hash rate that gives you a good block every hour on average ?   It means that you need to hash about 60 minutes on average, to find a hash that satisfies the difficulty.  Whenever that happens, you win a block - unless this happens during those few seconds when your block will get orphaned, which means that you've wasted a few seconds of hash rate.

Your chances don't increase when you've been mining for 20 minutes on the same block.  It doesn't matter on which block you mine.  You could change every 20 seconds the block on which you're mining, and you would STILL be finding a block, on average, every 60 minutes.

So if all other miners shut down, the only thing that happens is that THEIR blocks don't appear any more.  Yours will.  On average, every 60 minutes.

Other miners disappearing doesn't make you win more blocks (except those few that were orphaned).  Other miners not being there any more will make you win more blocks AFTER DIFFICULTY GETS LOWERED, but not before.

If you have hash rate to win a block every 60 minutes with a given difficulty, that's what will happen, *independent of others*.  The fact that others are NOT mining, doesn't make you, by miracle find solutions faster.  It is not a competition of the BEST solution: it is a competition of finding A solution.  If another one found a solution after 10 minutes and you didn't, it means that you, well, didn't.  So if he wasn't there, you still wouldn't have a solution.

To take the example of your dice, if the game is: you have 5 dice, and you need to throw at least 4 times a 6 (difficulty) to win a point, then the number of times you will have to try on average before winning a point, is independent of the number of players you play with.

[aesma]

What is the size of the blockchain currently ?

[-ck]

What is the size of the blockchain currently ?

About 140GB but you can run a full node in pruned form and specify how much space to use with 550MB being the lowest setting. However a pruned node isn't as useful to the network as a full one since it can't serve old blocks to the rest of the network.

[franky1]

You have visibly a fundamental misunderstanding about mining blocks.  

If you have hash power that is so that, with a given difficulty, on average, you find a good block, say, every hour, which means that you have about 1/6 of the total hash power *when the difficulty was determined*, then it doesn't matter whether others are mining or not, you will win, on average, one block every hour - minus those few seconds that you were mining on the wrong block each time.

your not getting it at all!!

ok try this..

imagine the olympics 100m

5 guys.. they all run
average is 10 seconds to get to the other end, and only 1 guy wins

now then, you ask the 5 guys to run the length 6 times..

out of those 6 times there is a guy that only wins once..

does that mean if he was the only runner it would take him 60 seconds.. no

because the game is run for 10 seconds and reset.
all the guys run within milliseconds of each other but there is only one winner, the timings of the other 4 guys per race are not multiples of the first guy, but miliseconds behind the first guy.

[dinofelis]

You have visibly a fundamental misunderstanding about mining blocks.  

If you have hash power that is so that, with a given difficulty, on average, you find a good block, say, every hour, which means that you have about 1/6 of the total hash power *when the difficulty was determined*, then it doesn't matter whether others are mining or not, you will win, on average, one block every hour - minus those few seconds that you were mining on the wrong block each time.

your not getting it at all!!

Really, you are mistaken on how mining works.  No point in discussing further until this is cleared out.

For a given difficulty level, mining is a Poisson process with an average probability to win a block in a given time window dT, which equals

dT / 10 minutes * (your hash rate / hash rate corresponding to the difficulty level)

The rate of winning blocks is independent of others winning blocks as long as the same difficulty level is maintained.

But all this has nothing to do with your claim that full nodes can enforce a protocol (or a protocol chain) on the set of miners if these are agreeing amongst themselves on a different protocol and do not fork off.

[dinofelis]

You have visibly a fundamental misunderstanding about mining blocks.  

If you have hash power that is so that, with a given difficulty, on average, you find a good block, say, every hour, which means that you have about 1/6 of the total hash power *when the difficulty was determined*, then it doesn't matter whether others are mining or not, you will win, on average, one block every hour - minus those few seconds that you were mining on the wrong block each time.

your not getting it at all!!

ok try this..

imagine the olympics 100m

5 guys.. they all run
average is 10 seconds to get to the other end, and only 1 guy wins

This is simply wrong, because the mining process is not a cumulative work towards a solution.  Every hash is a random trial, independent of other trials.  It is not because you have been hashing for 20 minutes on a block, that your probability of finding a solution in the next second is higher than if you just started hashing on that block or any other one.  It is a Poisson process, not a cumulative calculation.

The process is much closer to:  everyone has a certain number of dice (= hash rate ; you have 7 dice say).  The difficulty is: in one throw, you have to have at least 4 six.  You (just as anyone else) throws every second the dice he has).  Each time someone has 4 six, he wins the round.  Most of the time, when someone wins, nobody else wins (orphaning rate is low).  Your rate of winning rounds is essentially unaffected by the fact that others play too, because most of the time when you throw 4 six, nobody else does.  Whether the others play or not, your RATE of winning a round is independent of that.
In fact, in each round, you have a chance of 1/56.7 to win, so you will win about one round every 57 rounds.  As long as you are much less than 57 players, what the other players do has not much influence on your rate of winning.

[-ck]

You have visibly a fundamental misunderstanding about mining blocks.  

If you have hash power that is so that, with a given difficulty, on average, you find a good block, say, every hour, which means that you have about 1/6 of the total hash power *when the difficulty was determined*, then it doesn't matter whether others are mining or not, you will win, on average, one block every hour - minus those few seconds that you were mining on the wrong block each time.

your not getting it at all!!

ok try this..

imagine the olympics 100m

5 guys.. they all run
average is 10 seconds to get to the other end, and only 1 guy wins

This is simply wrong, because the mining process is not a cumulative work towards a solution.  Every hash is a random trial, independent of other trials.  It is not because you have been hashing for 20 minutes on a block, that your probability of finding a solution in the next second is higher than if you just started hashing on that block or any other one.  It is a Poisson process, not a cumulative calculation.

It's incredible that you have to explain something so basic about mining to this moron who claims to be so knowledgeable that he posts hundreds of times a day, purely to discredit core, blockstream, segwit, whatever isn't BU. Do yourself a favour and put him on ignore; he feigns some kind of smarts that look like he's creating a counterargument when in fact it's a load of bollocks. He's probably laughing about how he pretends to answer the question while attempting to ridicule real progress, development and intelligent discussion as a shill... or more likely he's just a moron.

[franky1]

You have visibly a fundamental misunderstanding about mining blocks.  

If you have hash power that is so that, with a given difficulty, on average, you find a good block, say, every hour, which means that you have about 1/6 of the total hash power *when the difficulty was determined*, then it doesn't matter whether others are mining or not, you will win, on average, one block every hour - minus those few seconds that you were mining on the wrong block each time.

your not getting it at all!!

ok try this..

imagine the olympics 100m

5 guys.. they all run
average is 10 seconds to get to the other end, and only 1 guy wins

This is simply wrong, because the mining process is not a cumulative work towards a solution.  Every hash is a random trial, independent of other trials.  It is not because you have been hashing for 20 minutes on a block, that your probability of finding a solution in the next second is higher than if you just started hashing on that block or any other one.  It is a Poisson process, not a cumulative calculation.

YOU WERE saying by taking people away makes the time double, triple quadruple.. not me

but now you are switching so now you are proving my point
my point is that everyone is independant and there is only one winner.. i never said its cumulative.. it was you that said it was cumulative by suggesting take 90% away and people will be waiting hours..

they wont..
AGAIN
have an olympic 100m race of 10 guys.. the winner gets to the end in 10 seconds..
now imagine if he was shot..
the runner up WONT!!!!!! have got to the other end in 20 seconds.. he would have got there in about 10seconds (plus a few miliseconds)

shoot all 9 runners so there is only 1 runner.. that last runner. again would still reach the end point in ~10 seconds.

based on YOUR scenario of everyone having 10% of hashrate. ill show you what i mean
YOUR (wrong) scenario:
"In other words, if this pool has 10% of all the hash rate, when his peers have mined 90 blocks on the new chain, he will have mined 10 blocks on the old chain. "

AGAIN only winning 1 block an hour does not mean your X* slower than anyone else.. it just means your competing against X number of people and 1 of X times you happen to be milisecond faster that anyone else.

you wont find runner 1 =10seconds
you wont find runner 2 =20seconds
you wont find runner 3 =30seconds
you wont find runner 4 =40seconds
you wont find runner 5 =40seconds
you wont find runner 6 =40seconds

you will find they are all close by each other yes a difference in hashrate can make a difference .. BUT so can LUCK of the randomness of the solution

take your example again
"In other words, if this pool has 10% of all the hash rate, when his peers have mined 90 blocks on the new chain, he will have mined 10 blocks on the old chain. "
if 9 pools went off and mined on chain X and 1 pool remained on chain A

after say 900 minutes
pool1chainX =~10 blocks taking ~10 minutes EACH block
pool2chainX =~10 blocks taking ~10 minutes EACH block
pool3chainX =~10 blocks taking ~10 minutes EACH block
pool4chainX =~10 blocks taking ~10 minutes EACH block
pool5chainX =~10 blocks taking ~10 minutes EACH block
pool6chainX =~10 blocks taking ~10 minutes EACH block
pool7chainX =~10 blocks taking ~10 minutes EACH block
pool8chainX =~10 blocks taking ~10 minutes EACH block
pool9chainX =~10 blocks taking ~10 minutes EACH block
(totalling 90 blocks with average 10minutes)

and
pool1chainA =~90 blocks in ~900 minutes
pool1chainA has no competition to lose by, by seconds

check out the orphan timing
https://blockchain.info/orphaned-blocks
the runner ups are SECONDS behind each other not minutes/hours


465722    
Timestamp    2017-05-10 08:19:11
Number Of Transactions    2752
Relayed By    Bixin
   
Timestamp    2017-05-10 08:19:10
Number Of Transactions    2726
Relayed By    GBMiners

[dinofelis]

YOU WERE saying by taking people away makes the time double, triple quadruple.. not me

Last try to make you understand the basics of bitcoin mining.

Consider 5 mining pools, each with 20% of the hash rate and with "difficulty in equilibrium" which means that ON AVERAGE there's 1 block every 10 minutes in all, and each pool wins a block every 50 minutes on average (1/5 of the blocks).

Up to here, I hope you are with me, OK ?

Now, consider that the first 4 pools switch off, but the difficulty level remains unchanged (it will not change for the next 2000 blocks).

Well, the fifth pool will still make his blocks every 50 minutes, and that's it.

So:

1) the fifth pool won't see much of a difference in his rate of winning blocks (namely about 1 every 50 minutes on average, with exponential probability distribution)

2) the total chain now being made by only this pool, the whole chain only wins 1 block every 50 minutes on average.

This is a very elementary notion in bitcoin mining.  If you don't agree, you've seriously misunderstood something.  Go ask elsewhere if you don't believe me.  Until this point is cleared up, there's no point in discussing other aspects of the bitcoin system.

[spartacusrex]

YOU WERE saying by taking people away makes the time double, triple quadruple.. not me

Last try to make you understand the basics of bitcoin mining.

Haha! Join the Club Dino!! we've all tried that.. (never works..)

This is a very elementary notion in bitcoin mining.  If you don't agree, you've seriously misunderstood something.  Go ask elsewhere if you don't believe me.  Until this point is cleared up, there's no point in discussing other aspects of the bitcoin system.

Boom.. Have you met Franky1 ? Let me introduce you.. 

[dinofelis]

YOU WERE saying by taking people away makes the time double, triple quadruple.. not me

Last try to make you understand the basics of bitcoin mining.

Haha! Join the Club Dino!! we've all tried that.. (never works..)

This is a very elementary notion in bitcoin mining.  If you don't agree, you've seriously misunderstood something.  Go ask elsewhere if you don't believe me.  Until this point is cleared up, there's no point in discussing other aspects of the bitcoin system.

Boom.. Have you met Franky1 ? Let me introduce you.. 

I hope it is not that bad...

[dinofelis]

I need to add something, to lift a potential confusion: the orphaned blocks.

One may erroneously think that there are, say, 5 miners *in competition* and that it takes them *about exactly* 10 minutes of computing to get a block, but that sometimes, it takes the first one only 9 minutes and 59 seconds, and the second one, 10 minutes and 1 second, and they were all almost within a few seconds "in time".

This would be the case if mining a block were a cumulative effort: that the calculation took 10 minutes (and sometimes a few seconds less, and sometimes a few seconds more).

But, as pointed out before, this is not the case.  Mining a block is calculating hashes, and only one out of X hashes is an acceptable one, but calculating more hashes doesn't change the *probability* for the next hash to be good or bad.

As such, at each hash you calculate, you have a probability of 1/X to have a good hash and win the block.  This means that ON AVERAGE you have to calculate about X hashes, before you can hope ON AVERAGE to have a good one.  But this good one could be the first, or it could only come after 3X trials.  The probability distribution of the number of hashes needed to find the first good hash, is an exponential, with an average of X.

This means that miners find blocks at random times.  Sometimes, you find it directly.  The first hash you tried was the good one.  Sometimes, you have to calculate 3X hashes.  It is not that there is a block EVERY 10 minutes.  There is a block at RANDOM TIMES, and the interval is ON AVERAGE 10 minutes of all miners combined, but has an exponential distribution.
Each miner has an exponential distribution of "winning blocks", with an AVERAGE time that is given by

T_av = 10 minutes * (1/fraction of hash rate needed for 10 minute difficulty)


This means that MOST OF THE TIME, when a miner finds a block, HE'S THE ONLY ONE.  So MOST OF THE TIME, a miner wins all the blocks that he can solve, at an average time of T_av.

But it can be that during the small interval of time when the miner was calculating on the wrong (old) block, he happens to find a block JUST after another miner already found a block.  Essentially, the probability of this happening is dT / T_av, where dT is the kind of propagation time of a block from other miners to him.  In that case, our miner didn't know he was mining on the wrong (old) block, and publishes his block, to see that in fact, he was late.  That's orphaning. 

The frequency of orphaning can be estimated as follows:

Let us say in our artificial example that there are 5 mining pools with each, 20% of the hash rate.  This means that for all of them, T_av = 3000 seconds.

They all produce a block according to an exponential distribution, with an average of 3000 seconds each of them.

Suppose that it takes 2 seconds for a block to get propagated and checked by the other miners.  This means that during these 2 seconds, a miner hasn't yet seen and accepted the block of his peer that was published.  He has a probability of 2 s / 3000 s to find a block exactly during this time.  So once out of 1500, he will also publish a block, which will be orphaned.  So every miner will have a block orphaned once out of 1500.  Because there are 5 miners, it means that about every 300 blocks, a block is orphaned.

[jonald_fyookball]

YOU WERE saying by taking people away makes the time double, triple quadruple.. not me

Last try to make you understand the basics of bitcoin mining.

Consider 5 mining pools, each with 20% of the hash rate and with "difficulty in equilibrium" which means that ON AVERAGE there's 1 block every 10 minutes in all, and each pool wins a block every 50 minutes on average (1/5 of the blocks).

Up to here, I hope you are with me, OK ?

Now, consider that the first 4 pools switch off, but the difficulty level remains unchanged (it will not change for the next 2000 blocks).

Well, the fifth pool will still make his blocks every 50 minutes, and that's it.

So:

1) the fifth pool won't see much of a difference in his rate of winning blocks (namely about 1 every 50 minutes on average, with exponential probability distribution)

2) the total chain now being made by only this pool, the whole chain only wins 1 block every 50 minutes on average.

This is a very elementary notion in bitcoin mining.  If you don't agree, you've seriously misunderstood something.  Go ask elsewhere if you don't believe me.  Until this point is cleared up, there's no point in discussing other aspects of the bitcoin system.

I believe you guys are just arguing semantics.   Franky never implied it wasn't a poisson process.  

In other news, I find it hilarious in this thread that Satoshi's vision is now considered "blasphemy".  Funny how that
never happened before Blockstream came on the scene.

[dinofelis]

I believe you guys are just arguing semantics.   Franky never implied it wasn't a poisson process.  

Of course he did.  He thinks that if there are 5 mining pools, with each of them 20% of the hash rate, and 4 of them switch off, the 5th one will continue make blocks every 10 minutes.

This is the argument he used for a mining pool to leave the agreement he has  with his peers to remain on their mutual protocol, and to switch to the protocol the full nodes want to impose on the miners, because "then he is alone and will make blocks every 10 minutes, pleasing the full nodes and reaping in all the rewards".

My point was that
1) this was not the Gedanken experiment that needed to prove that full nodes can force their protocol onto miners
2) the betraying node is not winning anything, because he's not making blocks at any faster pace than if he remained faithful to the other miners and their agreed-upon protocol.

[jonald_fyookball]

I never interpreted anything in this thread that he implied that (I actually think he tried to argue exactly what you are arguing -- that the percentages don't change based on the other pools) but I guess he can answer to that point.

[mindrust]

Why would anyone run a full node?

First and more importantly, it doesn't earn me any money.

Second and moderately importantly, as it is not generating any money; it costs me money It takes ram, cpu power, hdd space, electricity, internet bandwidth >> money to run a full node. What do i get in return? Nothing. There are many people who run it already no need for more.

If there weren't enough people on the other hand, i would gladly run one.