Bitcoin 100x less secure than commonly believed

[mariano76]

In terms of prime numbers that don't look ridiculous or small, 25 and 33 look a hell of a lot better.

Niether of those numbers are prime.

haha you are right

[r0ach]

Niether of those numbers are prime.

Meant two odd numbers.  I read someone talking about 17 and the first two numbers that don't seem minuscule off the top of my head were 25 and 33.  31 is ok for prime numbres.

[wuyanren]

I haven't seen BM for a few days. I miss him so much.

[monsterer]

I've just realised something interesting. Either:

1. No mining pools are aware of this exploit
2. Pools are already actively engaged in it
3. It doesn't work in practice

2 is particularly interesting - bytemaster's claim that all pools would be foolish not to engage in this attack against other pools leads to the implication that on the limit, all pools will already be doing this to all other pools. This includes bytemaster's evil subsidy pool, which would be vulnerable to the same attack, so in the end this attack is a zero sum game.

[sittingduck]

Statistically each member of a pool only finds a winning hash once every few years


Sent from my iPhone using Tapatalk

[Troglodactyl]

A negative miner is a miner who adds an additional filter onto their mining software so that it submits only hashes that are between the pool difficulty and the chain difficulty, instead of all hashes above the pool difficulty.  This will make very little difference in the miner's payouts from the pool, but it means that the miner is contributing no value at all to the pool.

Understood. Wouldn't this kind of thing stand out statistically, though for any pool doing analysis on share submission?

Not really.  A major reason for using pools is because you don't have enough hash power to get your own blocks regularly, so accumulating a large enough sample size to be statistically significant could take longer than the time frame of the attack.  Even if it was detectable eventually (which is unlikely) the negative mining could be done anonymously with new accounts and proxies each time one was burned.  There would be false positives from unlucky honest miners to sort out also.

[monsterer]

A negative miner is a miner who adds an additional filter onto their mining software so that it submits only hashes that are between the pool difficulty and the chain difficulty, instead of all hashes above the pool difficulty.  This will make very little difference in the miner's payouts from the pool, but it means that the miner is contributing no value at all to the pool.

Understood. Wouldn't this kind of thing stand out statistically, though for any pool doing analysis on share submission?

[Troglodactyl]

A negative miner is one who joins a pool, collects their share, but never returns a wining hash

Aren't pool rewards paid out per share? One share is a solved POW, so if you don't submit a winning hash, you don't have a share in the first place.

To calculate hash rate, pool miners turn in all hashes that meet a significantly lower difficulty level than the mined chain difficulty level to the pool.  Pool payouts are based on how many of these easier hashes each miner submits, as they represent the amount of hash power each miner is contributing.  Miners submitting every single hash regardless of its score would be horribly inefficient; they only submit hashes above the pools set difficulty level, so the vast majority of hashes submitted to the pool are above the pool difficulty but below the blockchain's current difficulty level.

A negative miner is a miner who adds an additional filter onto their mining software so that it submits only hashes that are between the pool difficulty and the chain difficulty, instead of all hashes above the pool difficulty.  This will make very little difference in the miner's payouts from the pool, but it means that the miner is contributing no value at all to the pool.

[monsterer]

A negative miner is one who joins a pool, collects their share, but never returns a wining hash

Aren't pool rewards paid out per share? One share is a solved POW, so if you don't submit a winning hash, you don't have a share in the first place.

[wallace]

BTC is wasting energy, that's right.

but why the marketing can support the price? just the fee of electricity? obsolutely not. we should understand that. there're a lot of interest groups backed behind. miners, traders, some related company are part of the groups. it's a benefits chain, they will let the bitcoin system safe.

[mint chocolate chip]

In terms of prime numbers that don't look ridiculous or small, 25 and 33 look a hell of a lot better.

Niether of those numbers are prime.

[liondani]

All miners are currently paying an electric bill of  $6000 every 10 minutes and earning BTC worth $6300 for a net profit of $300.

You make the wrong assumption that all miners together pay $6000 every 10 minutes!!! How do you know that?
I say  that many miner's steal electricity and they mine for free! So the real average cost of miners could be in reality something like  $3000 for example or  whatever....
And the victims can not stop mining (indirectly) since they don't know they are victims.... (even we can be victims cause we are mining bitcoins because of a virus mining software installed on our computers)

I personally have no doubts that electricity costs are for many miners near zero....

[r0ach]

In terms of odd numbers that don't look ridiculous or small, 25 and 33 look a hell of a lot better.

[Empirical1.2]

I knew it was wasteful, but didn't realize the numbers were that tight. What's your sense of the halving ~ August 2016? Is the market going to price that in or will it ultimately make mining Bitcoin unprofitable?

I know you asked BM. With the same level of average new demand but half the mining cost the price could in theory double.

If we look at the recent Litecoin halving. The price averaged circa $60 million for the first 6 months of the year. After their halving, for the last 6 weeks or so the price has averaged $120 million.


Which is what I kind of expected.

Regards LTC. It settled in the $50-60 million range for the majority of this year. With 30% annual inflation that implies an average of $1-1.5 Million in new demand every month to offset sales of newly generated coins. All else being equal after the halving LTC could sustain twice the price, $120 million with the same level of demand.

However all else is not equal, there has been the greek crisis, Chinese market problems, BTC problems and also an LTC ponzi which has increased demand and influenced the price positively.

So in terms of how high LTC can go in the short term it's highly influenced by external events but I think it can sustain at least $120 million with the same level of average new demand that it had in the first half of 2015.

[donkeypong]

Exactly, by paying the miners NOT to hash you remove them from the competition, then you PAY them to hash for you.       

So, you buy their hashing power... how is this not just paying for majority hash?

Assuming 5% profit margin:

All miners are currently paying an electric bill of  $6000 every 10 minutes and earning BTC worth $6300 for a net profit of $300.

It is commonly believed that an attacker would have to spend $6001 every 10 minutes to get 50.01% away from these miners.

Technically speaking the attacker only needs to pay the existing miners $1 more every 10 minutes and the miners could defect for a higher pay rate and the attacker gives them a long-term contract for $301 every 10 minutes for their hash power.  The long-term contract protects the miners against any volatility in BTC price as a result of the attack resulting in the attacker having total costs of $1 +  $300*PERCENT_DECLINE_IN_BTS_PRICE as a result of the attack.

Miners have a business model with ever-decreasing profit margins so their job positions have ever decreasing salaries.   Therefore there is no long-term upside to supporting BTC for miners and they would be foolish to not sell out to someone offering them a better long-term income.

Negative mining is just an approach to push pools out of business and concentrate power in the hands of large mining farms which you can then negotiate with.

A large mining farm would be foolish to do anything other than negative mining on the public pools.   Large public pools serve to allow more competition and thus lower overall margins.   

Hence if I owned 10% of all hashing power, I would use it entirely for negative mining until the pools collapsed and then switch to solo-mining.   This would maximize my long-term profits and secure my position of control.

I knew it was wasteful, but didn't realize the numbers were that tight. What's your sense of the halving ~ August 2016? Is the market going to price that in or will it ultimately make mining Bitcoin unprofitable?