Bitcoin's kryptonite: The 51% attack.

[cartman]

would it be possible to detect such an attack and "vote the attacker out of the network"?

just random thoughts... like upgrading to a client that limits haspower within a certain region or ip-range...

[1713919045]

1713919045

[andes]

would it be possible to detect such an attack and "vote the attacker out of the network"?

just random thoughts... like upgrading to a client that limits haspower within a certain region or ip-range...

The first point would be wonderful if feasible.

Regarding the second point, take into consideration that the whole paradigm of security/honesty of Bitcoin is based on brute hashing power. Reducing hashing power means getting weaker to attacks. What you say would require changing the whole security paradigm of Bitcoin. Not saying its not possible, but would need some of out of the box thinking IMO. But then again, I am just a newbie here.

[sketchman]

Also small miners can more easily deal with or during the winter benefit from the waste heat of mining.

https://i.imgur.com/pkMU6.png

[Transisto]

IMO, Seeing this network grow, learning and profiting from it would be better to these groups that attacking it.

The "current" network could only protect itself if these superpower start competing at the same time.

Given the resources available and the small ratio of ASIC/GPU making up the total, I would not expect an attack to be imminent.  We're nowhere near parity with electrical cost on 40 nm ASIC.

[amincd]

One mistake I have read, is saying that Bitcoin is as resilient as Napster, by being p2p. We cannot equate Bitcoin with Napster. Napster has no single point of failure;

Napster was not a pure p2p program. It relied on a central server, and thus had a single point of failure. Bitcoin does NOT have a central point of failure. You literally need more hashing power than all of the honest nodes to disrupt the network.

[andes]

Related topic

Inevitable development into mining elite?
http://forum.bitcoin.org/index.php?topic=15345.0

[AntiVigilante]

One mistake I have read, is saying that Bitcoin is as resilient as Napster, by being p2p. We cannot equate Bitcoin with Napster. Napster has no single point of failure;

Napster was not a pure p2p program. It relied on a central server, and thus had a single point of failure. Bitcoin does NOT have a central point of failure. You literally need more hashing power than all of the honest nodes to disrupt the network.

Storage is decentralized.

Mining is centralized. Entry is as well.

Trading is centralized. Decision making is also.

There's your vulnerabilities.

[Drone_Fragger]

Nah you're all wrong, turns out bitcoins realkryptonite is a nerd making a post on an internet forum about how the price of bitcoins is crashing and how all the greasy neckbearded libertarians who believe in their currency based on cheeto dust should bail out of it before they're reduced to penniless hobos.

[Warren]

Couldn't the 51% be raised, to say 95% instead?

There is no way that any single entity/organization could ever have 95% of the computing power of the network...

[andes]

Couldn't the 51% be raised, to say 95% instead?

There is no way that any single entity/organization could ever have 95% of the computing power of the network...

By design, Bitcoin's "true" block chain is not determined by voting (say 95%) but by "proof of work", the chain with more "proof of work" wins. This is a probabilistic issue. Another way to see it, is as a competition issue. The team that dominates 50%+ of the power can outperform the rest.

If you own more than 50% of the hashing power, you will probably be able to produce the longest chain measured in CPU power (you will beat the rest more times in the mining lottery). If you are not honest, you will mess the whole network, until the honest guys regain control. That does not mean you will own the chain, or that you can do a long term damage to the chain, but you can cause a lot of trouble in the meantime, especially if you can sustain the attack for a long time. Think about if you could not trust the confirmations you get after doing a transaction. This would damage the confidence in the network.

That is an unresolved weakness in the Bitcoin paradigm, in my opinion.



EDIT: More related links:

If an attacker gets more than 50 % of mining power
http://forum.bitcoin.org/index.php?topic=24996.0;all

POLL: What are the most likely things that may cause bitcoin to fail ? (merged thread)
http://forum.bitcoin.org/index.php?topic=25026.0

My Response to Ben Laurie’s ‘Last Word’ on Bitcoin
http://forum.bitcoin.org/index.php?topic=25760.0

[markm]

My noob self has to agree...the central "what happens if a pool/operator gets 51% and double spends" hasn't been touched in many of these replies.

Seems like a good test to see if you actually understand btc.

What happens if someone commits fraud / computer fraud / theft / computer theft / wire fraud / wire theft, in other words?

I think there might be some legal rulings or precedents already on the books in various jurisdictions about such activities.

Without getting into specifics, I can say that in general they do seem to me to tend to frown upon such activities.

Possibly the main problem would be ensuring that by the time you are able to convert whatever recompense you might be awarded by the courts back into bitcoin it does actually come out as about the amount it was intended by the court to amount to.

How anonymous, in reality, are people who control >50% of the hashing power, typically?

-MarkM-

[PatrickHarnett]

so, when this thread was arguing the possible flaws of one person generating all the blocks, did anyone bother to notice that 1/3 of the coins are already in existence?

[JoelKatz]

The last time I did the "back of the envelope" calculations (about two weeks ago), it would cost about $12,000,000 to reach 50%. That's the cost to design, contract, receive, and hookup fully-custom ASICs to build a hashmonster.  I believe it was 200 blades, each with 24 ASICs. Each ASIC would have 32 2xSHA256 cores running at 350Mhz, each with the hashes fully-implemented as custom gates, pipelined. (So while we're doing round 3 of the 3rd nonce, we're doing round 2 of the 2nd, and so on.) Each core, once its pipeline was full, could churn out one doublehash every 8 clock cycles. The total hashing power: 6.7THash/s. (That's from memory.)

However, this would only give you half if it was already half. Which it might be, for all we know.

[manuelgar]

Andes, to have control of 50% of the network you need to deploy 101% processing power (you´ll end up having 50,5%)

But, why thinking there will only be 1 dishonest entity?  If there is one, there could be more, each one of them would have their own self-interest on controlling the network.   If a lot of "dishonest" entities with great processing power try to control the network, they would make the 51% attack less probable themselves.

[hugolp]

The last time I did the "back of the envelope" calculations (about two weeks ago), it would cost about $12,000,000 to reach 50%. That's the cost to design, contract, receive, and hookup fully-custom ASICs to build a hashmonster.  I believe it was 200 blades, each with 24 ASICs. Each ASIC would have 32 2xSHA256 cores running at 350Mhz, each with the hashes fully-implemented as custom gates, pipelined. (So while we're doing round 3 of the 3rd nonce, we're doing round 2 of the 2nd, and so on.) Each core, once its pipeline was full, could churn out one doublehash every 8 clock cycles. The total hashing power: 6.7THash/s. (That's from memory.)

However, this would only give you half if it was already half. Which it might be, for all we know.

But for the situation to be realistic we have to assume that the honest miners would already be using the same technology. Otherwise your scenario is only valid in the concrete moment when the technology is evolving.

[JoelKatz]

The last time I did the "back of the envelope" calculations (about two weeks ago), it would cost about $12,000,000 to reach 50%. That's the cost to design, contract, receive, and hookup fully-custom ASICs to build a hashmonster.  I believe it was 200 blades, each with 24 ASICs. Each ASIC would have 32 2xSHA256 cores running at 350Mhz, each with the hashes fully-implemented as custom gates, pipelined. (So while we're doing round 3 of the 3rd nonce, we're doing round 2 of the 2nd, and so on.) Each core, once its pipeline was full, could churn out one doublehash every 8 clock cycles. The total hashing power: 6.7THash/s. (That's from memory.)

However, this would only give you half if it was already half. Which it might be, for all we know.

But for the situation to be realistic we have to assume that the honest miners would already be using the same technology. Otherwise your scenario is only valid in the concrete moment when the technology is evolving.

Honest miners are not using this technology today because they can't afford to develop and build it. The return on investment would come from the attacks, not from mining revenue. You can't afford tech like this for honest mining. This plan only works on an economy of scale. You can't do it with FPGAs.

And, by the way, we have no way to know whether or not dishonest miners already have this technology and are using it. They could be feeding into any number of pools and running as independents as we speak -- making us think the network is much more secure than it really is. (I don't think this is actually happening, but it can't be ruled out.)

You couldn't even start on something like this with less than $2,000,000.

[Michele1940]

Hi, as a newbie I want to move a question I put recently in the middle of another thread, because I think it deserves its own discussion. Sorry if this thing has been answered before, I have been reading the forum for several hours so far but did not find the answer yet. If it was answered elsewhere, please post the link.

Is decentralized mining power important for the security and long term independence of bitcoin?

I have read a lot about transactions being decentralized as a built-in feature of bitcoin, but what about decentralized block creation? The bitcoin architecture does not guarantee decentralized mining at all. In fact, the network could in theory work "as well" with nothing more than one powerful miner, or pool of miners. Am I right?

If concentration of mining-power increases (because of bitcoin difficulty increasing faster than moore's law, leading to bigger hardware investments needed to be in the game, profitability decreasing, and economies of scale kicking in) ^(Note1), can a few miners produce all the blocks in the network without compromising the security and independece of the project? Is it possible to avoid excesive concentration of mining power? I dont see how in the current configuration of the system.

I read somewhere that Bitcoin assumes never a 50%+ of the mining power will be concentrated in one hand or in one cartel. That's the principle behind the honesty validation of the longest chain by "proof of work". Correct me if my newbie understanding is wrong on this. To assume that this concentration of computing power will never happen is ludicrous to my current level of understanding of bitcoin and human behavior.

This raises some further questions. As difficulty changes every 2 weeks, what happens if a Google-like company with bad intentions gets into the game suddenlly with 10x the total combined power of current miners? Could this sudden change of rules endanger bitcoin? Destroy it? I mean lets consider this wild posibilities. For big corporations this move would be peanuts. Powerful states overthrow smaller goverments all the time, big corporations eat small corporations all the time.

Hope to hear some thoughts from the experts out there.


(Note1): Thinking about the issue of increasing bitcoin difficulty, let's remember that  by design difficuly increases when mining power increases, in order to keep the creation rate at 10 minutes per block. So, any powerful organization that wanted to gain control of Bitcoin, could do it easily by injecting enormous amounts of mining power to the network, and by doing so, effectively reducing the rest of the miners relative power, and at the same time putting them out of business, because the difficulty would be so high, that mining would be generated below cost (subsidy). Knowing the enormous level of concentration of economic resources in the current world, this hypothesis seems in fact the most likely outcome. Predatory competition is a reality in todays market. I predict honest miners will be subjected to predatory competition if powerful economic powers decide to take control of Bitcoin.

Following this line of thought, I see Bitcoin could never become what it promises: a descentralized and free currency, if it is left alone in the wild "free market". I hope someone can find flaws in my arguments, or present ideas to correct this flaw. By the way, I have a decent amount of money put in this project, so I feel sad to become aware of this potential vulnerability. If people agree this is a serious vulnerability, lets get into "troubleshooting mode".

As a Newbie, I have also had the same thought like you had. And on june 24th, started this thread in the newbie section : Google $ CO. Mining ? http://forum.bitcoin.org/index.php?topic=21905.0

At the end of the story, no conclusions at all. Now the thread is buried....Have you managed to get to any conclusion in your thread ? This will save me a lot of reading...Thanks.

[terroh8er]

I think the only people we have to worry about are angry governments. They could set up an undercover pool and offer some incentive to miners who join it, whether it is lower fees or whatever. This would divert hashing power away from honest pools and create a starting point so it wouldn't be as costly. I'd imagine that with good marketing, one could get at least 25-30% hashing power like this. After that, they could buy the rest of the hardware for a few million and cause the market to crash. The Wall Street types would say "I told you so" and most people would lose interest in this type of currency. The reason I don't think greedy individuals would be interested in this is because Bitcoins would be worth next to nothing after the attack (assuming it would be detectable).

If it is ever revealed that militant groups are holding large amounts of Bitcoins, Mossad and the CIA would easily justify whatever costs are necessary to diminish it's value. Because it is inevitable that some powerful person's enemy will use Bitcoin, I think this is something to be concerned about.

[manuelgar]

As a Newbie, I have also had the same thought like you had. And on june 24th, started this thread in the newbie section : Google $ CO. Mining ? http://forum.bitcoin.org/index.php?topic=21905.0

At the end of the story, no conclusions at all. Now the thread is buried....Have you managed to get to any conclusion in your thread ? This will save me a lot of reading...Thanks.

Hi Michele,

I insist in my previous post. Why thinking there will only be 1 dishonest entity?  If there is one, there could be more, each one of them would have their own self-interest on controlling the network.   

If a lot of "dishonest" entities with great processing power try to control the network, they would make the 51% attack less probable themselves.

[lettucebee]

As a Newbie, I have also had the same thought like you had. And on june 24th, started this thread in the newbie section : Google $ CO. Mining ? http://forum.bitcoin.org/index.php?topic=21905.0

At the end of the story, no conclusions at all. Now the thread is buried....Have you managed to get to any conclusion in your thread ? This will save me a lot of reading...Thanks.

That's because putting "newbies" into a special, quarantine cage is a terrible idea for so many reasons, not least of which is that no one on the outside ever goes there and so any great ideas, which new people often bring to the party, are dropped on the floor forever!