Topic: Please run a full node - page 7. (Read 6669 times)

See: now you got it directly from him  


franky1, really, you have a serious, serious misunderstanding of the basics of bitcoin mining.

You are EXACTLY committing the error I pointed out above:


I think it is because you do not understand the difference between a very peaked distribution around 10 minutes intervals, and an exponential distribution with AVERAGE 10 minutes.

lol your reading too much reddit
seems you are already reading the scripts of the doomsday FUD buzzwords


so far other nodes brands lik BU have only dropped at most 420ish
 on march 17th CORE dropped 560 nodes in on go.


also core is not perfect
https://github.com/bitcoin/bitcoin/issues?q=is%3Aissue+is%3Aopen+label%3ABug
and thats just the bugs they want to reveal. they also have a policy to not publicly divulge certain bugs for atleast a month after a fix
as commented here
https://github.com/bitcoin/bitcoin/issues/10364

oh and BTCC (core/blockstream/dcg cartel pool) also does empty blocks
https://blockchain.info/block-height/465117

What full node should i run ? chinese nodes aka BUg nodes ?
that is using an exploit called ASICBOOST that only mines an empty block.
Or should i run a core devs nodes that maintained the welfare of bitcoin through all this years ?

wrong.. he has no competition so although his average times of maybe say 10.026.. he is not fighting off competition

ok

imagine it this way...

there are 2 100m tracks
track A
5 runners
they do their runs and after 100 races
on average each racer wins 20 races (give or take as some racers are slightly faster) but each race has a winner somewhere around 10 seconds


track B
1 runner
he runs, and after 100 races
he wins 100 races no competition
each race take around 10 seconds give or take a few miliseconds

he does not take 50 seconds.. he still only takes ~10 seconds but because he has no competition he wins every time

here il display it better


remember only the green numbers matter as thats the "fastest win", and there is no second place.. hense why no one cares about how long it takes the rest, which is where you wrongly think it takes the others much much much longer than reality(you imply he only gets 1 block every ~50 seconds)

as you can see.. on a track of 5 racers 1 guy only wins once out of 5 races..
but on his own he wins every race.. obviously.. and also it does not mean that if he only wins 1 race out of 5 with competition. that it would take 50 seconds without competition

ut 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.

This is what I'm trying to point out.  I don't know if it is "lying" or "indoctrinated" or whatever.
But the argument that bigger blocks would lead to centralization, while the centralization already took place, always left me astonished: no-body ever reacted to that.  It is not so much that I absolutely want bigger blocks or whatever - it is that the argumentation is fallacious, and I have a hard time believing that the experts saying so, can't make the same obvious reasoning than I did here, and did several times, just to be greeted with the counter argument "but everyone knows that full nodes matter / keep the network honest / .... " and if that doesn't work, that "I'm a paid shill " or something of the kind.  So, are all these people self-deluded ; or do some of them know this but don't want it to be acknowledged ?

I have difficulties with fallacious arguments from experts.

That said, full nodes are not totally useless, but their only use is for *their owner* who is the only one who can *check for himself*.  But with that knowledge, he can do nothing else but acknowledge "that he has been had" or "that he hasn't been had", but that's about it.  The other advantage of a full node, for his owner, is that his owner can send out his own transactions, and nobody can know that HE was the one sending that transaction, as, being a full node, he would also send all transactions of light wallets connected to him.  So there is some kind of deniable anonymity of the IP address that sent out a transaction.

But that's about it.  These can be sufficiently good reasons to run a full node, BTW.

Of course, it is true that full nodes DID HAVE power of filtering "bad blocks" when mining nodes were connected only to the P2P network to other mining nodes, and didn't talk to themselves directly.  Then the P2P network could stop them from building a chain with which the full nodes didn't agree, because they would not receive one-another's blocks.

But as I explained several times, a miner pool would be crazy to wait for other miners' blocks through the P2P network, while he can just configure his node to connect directly to the other miner pool's node and get it faster, wasting less hash rate.  As this is mutually beneficial, I don't see why mining pools wouldn't do so.

When I look at the LOW orphan rate, I arrive at the conclusion that miners know one another's blocks in about 1.5 seconds on average, which indicates that those blocks cannot really hop from node to node in between.  1 MB blocks, transmitted and verified within 1.5 seconds seems impossible to do over a random P2P path, where each node receives the block, checks it, and sends it out again.

I agree with all you say. However, the situation is even more dire. Turns out we have been lied to all this time. The reality is that no entity is a node, if that entity does not mine. Indeed, this is the true, original, and proper definition of a node.

We need to reevaluate who 'sold us a bill of goods'. We have abdicated our authority in the network. In reality, non-mining nodes are irrelevant. Indeed, in the original wallet, nodes mine - period.

from https://github.com/trottier/original-bitcoin/blob/92ee8d9a994391d148733da77e2bbc2f4acc43cd/src/main.h#L795

GitHub
trottier/original-bitcoin
original-bitcoin - This is a historical repository of Satoshi Nakamoto's original bitcoin sourcecode

(emphasis added)

I don't know when the definition of 'node' became corrupted to include non-mining entities. I don't know who introduced this lie. Though I must admit to propagating it. For years. Sorry. I was ignorant.

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.

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.

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.

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.

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.

I hope it is not that bad...

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


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

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.

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

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.

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.

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.