[Archive] BFL trolling museum

[bitcoindaddy]

It "should" be easier

That's the problem with BFL. Yes it should be easier, but they don't respond to emails for days/weeks. This causes me to be placed at the end of the ASIC waiting list, even though I placed the order 2 hours after pre-orders went live. It has gotten to the point where I might just cancel my ~$5k worth of orders from BFL. Their customer service is absolute shit.

Your $5k is an insignificant drop in the ocean for them, there are people investing $100k each, and BFL spent atleast $1.5mil designing the ASIC, just because $5k is big for you doesn't mean it is big for them.

Like I said, if it's not about the money. Why don't they just hire somebody that could help them with the customer support?
I just find this all very shady.

They did hire somebody, but it seems like the customer service hasn't noticeably improved since the days when Sonny answered the emails.

[1713557620]

1713557620

[1713557620]

1713557620

[1713557620]

1713557620

[kentrolla]

From what I understand, the difficulty adjusts based on network speed, and the whole point of difficulty is to stop a single node from having too much power over the network ?

Question: Will the ASIC still work once all coins have been generated ?

the whole point of difficulty is to make sure the time it takes to find a block is consistent. it has nothing to do with the 51% attack.

[SgtSpike]

It "should" be easier

That's the problem with BFL. Yes it should be easier, but they don't respond to emails for days/weeks. This causes me to be placed at the end of the ASIC waiting list, even though I placed the order 2 hours after pre-orders went live. It has gotten to the point where I might just cancel my ~$5k worth of orders from BFL. Their customer service is absolute shit.

Your $5k is an insignificant drop in the ocean for them, there are people investing $100k each, and BFL spent atleast $1.5mil designing the ASIC, just because $5k is big for you doesn't mean it is big for them.

Like I said, if it's not about the money. Why don't they just hire somebody that could help them with the customer support?
I just find this all very shady.

They are trying to...
http://www.butterflylabs.com/jobs/

[rjk]

From what I understand, the difficulty adjusts based on network speed, and the whole point of difficulty is to stop a single node from having too much power over the network ?

Question: Will the ASIC still work once all coins have been generated ?

the whole point of difficulty is to make sure the time it takes to find a block is consistent. it has nothing to do with the 51% attack.

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

[dave3]

Just because somebody has a lot of hashing power and could get 51% doesn't necessarily mean they'd do an attack, either.  It's just a threat because they'd be able to if they wanted to.

[crazyates]

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

Huh? A 51% attack could also be called a 51% takeover, and happens when 1 person owns 51% of the hashing power, and takes control of the entire network (rewrites the blockchain). The larger the network is, the harder for 1 person to own 51%. It's a LOT harder to gain 51% of a 200TH/s network than it is a 2TH/s network, for example. A larger network is inherently more resilient to a 51% takeover, you see?

A larger difficulty is just a side effect of that larger network. The difficulty of that 200TH/s network will be 100x that of the 2TH/s network, right? The difficulty itself doesn't actually protect against an attack (or takeover), but the factors that caused the difficulty to raise are what protect us.

[rjk]

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

Huh? A 51% attack could also be called a 51% takeover, and happens when 1 person owns 51% of the hashing power, and takes control of the entire network (rewrites the blockchain). The larger the network is, the harder for 1 person to own 51%. It's a LOT harder to gain 51% of a 200TH/s network than it is a 2TH/s network, for example. A larger network is inherently more resilient to a 51% takeover, you see?

A larger difficulty is just a side effect of that larger network. The difficulty of that 200TH/s network will be 100x that of the 2TH/s network, right? The difficulty itself doesn't actually protect against an attack (or takeover), but the factors that caused the difficulty to raise are what protect us.

Exactly, that's what I was saying.

[Smoovious]

Question: Will the ASIC still work once all coins have been generated ?

Depends on how well they built it. It would probably die before that happens tho.

But, if you mean generally? Yeah, no reason why an asic wouldn't work after the rewards end.

Rewards aren't the purpose of mining, processing transactions is... Rewards are just a reward.

-- Smoov

[kentrolla]

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

Huh? A 51% attack could also be called a 51% takeover, and happens when 1 person owns 51% of the hashing power, and takes control of the entire network (rewrites the blockchain). The larger the network is, the harder for 1 person to own 51%. It's a LOT harder to gain 51% of a 200TH/s network than it is a 2TH/s network, for example. A larger network is inherently more resilient to a 51% takeover, you see?

A larger difficulty is just a side effect of that larger network. The difficulty of that 200TH/s network will be 100x that of the 2TH/s network, right? The difficulty itself doesn't actually protect against an attack (or takeover), but the factors that caused the difficulty to raise are what protect us.

Exactly, that's what I was saying.

no, its not what you were saying. what i said is completely true. difficulty has nothing to do with the 51% takeover. raising or lowering the difficulty would have 0 effect on the 51% takeover. bottom line is that you need 51% of the network's total hashing power.

[rjk]

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

Huh? A 51% attack could also be called a 51% takeover, and happens when 1 person owns 51% of the hashing power, and takes control of the entire network (rewrites the blockchain). The larger the network is, the harder for 1 person to own 51%. It's a LOT harder to gain 51% of a 200TH/s network than it is a 2TH/s network, for example. A larger network is inherently more resilient to a 51% takeover, you see?

A larger difficulty is just a side effect of that larger network. The difficulty of that 200TH/s network will be 100x that of the 2TH/s network, right? The difficulty itself doesn't actually protect against an attack (or takeover), but the factors that caused the difficulty to raise are what protect us.

Exactly, that's what I was saying.

no, its not what you were saying. what i said is completely true. difficulty has nothing to do with the 51% takeover. raising or lowering the difficulty would have 0 effect on the 51% takeover. bottom line is that you need 51% of the network's total hashing power.

The difficulty directly affects how much power you need to have in order to achieve 51%, correct? Therefore, for a network speed of (say) 12Th/s, you would need just over 12 MORE Th/s in order to be 51% of the network. However, if the difficulty was set for a network speed of 120Th/s, you would then need to have more than 120Th/s of hash power in order to be 51%.

The difference between a 12Th/s network and a 120Th/s network is directly related to the difficulty. If the difficulty were set for a 12Th/s network, and someone showed up with 120Th/s, he would have far more than 51% and the difficulty would rise to meet the hash power he is bringing online. In contrast, if someone threw 12Th/s at a network that was running at 120Th/s, he would be unable to do anything more than get a lot of coins, because he cannot overcome the difficulty to achieve 51%.

I'm not sure where the misunderstanding is. The network becomes more difficult to attack, directly as a result of how much power gets put up against it.

[mrb]

no, its not what you were saying. what i said is completely true. difficulty has nothing to do with the 51% takeover. raising or lowering the difficulty would have 0 effect on the 51% takeover. bottom line is that you need 51% of the network's total hashing power.

The difficulty directly affects how much power you need to have in order to achieve 51%, correct?

Technically, no. Because the difficulty adjusts itself with a certain latency. Right now the difficulty is 1.7M and the hash rate is 12 Thash/s. So you need 12 Thash/s right now to perform a majority attack. But if tomorrow the hash rate drops to 6 Thash/s for whatever reason, then you would only need 6 Thash/s, while the difficulty would still be 1.7M. The difficulty would only drop to 850k after up to 2016 blocks.

[SgtSpike]

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

Huh? A 51% attack could also be called a 51% takeover, and happens when 1 person owns 51% of the hashing power, and takes control of the entire network (rewrites the blockchain). The larger the network is, the harder for 1 person to own 51%. It's a LOT harder to gain 51% of a 200TH/s network than it is a 2TH/s network, for example. A larger network is inherently more resilient to a 51% takeover, you see?

A larger difficulty is just a side effect of that larger network. The difficulty of that 200TH/s network will be 100x that of the 2TH/s network, right? The difficulty itself doesn't actually protect against an attack (or takeover), but the factors that caused the difficulty to raise are what protect us.

Exactly, that's what I was saying.

no, its not what you were saying. what i said is completely true. difficulty has nothing to do with the 51% takeover. raising or lowering the difficulty would have 0 effect on the 51% takeover. bottom line is that you need 51% of the network's total hashing power.

The difficulty directly affects how much power you need to have in order to achieve 51%, correct? Therefore, for a network speed of (say) 12Th/s, you would need just over 12 MORE Th/s in order to be 51% of the network. However, if the difficulty was set for a network speed of 120Th/s, you would then need to have more than 120Th/s of hash power in order to be 51%.

The difference between a 12Th/s network and a 120Th/s network is directly related to the difficulty. If the difficulty were set for a 12Th/s network, and someone showed up with 120Th/s, he would have far more than 51% and the difficulty would rise to meet the hash power he is bringing online. In contrast, if someone threw 12Th/s at a network that was running at 120Th/s, he would be unable to do anything more than get a lot of coins, because he cannot overcome the difficulty to achieve 51%.

I'm not sure where the misunderstanding is. The network becomes more difficult to attack, directly as a result of how much power gets put up against it.

No, not at all.

If the difficulty was set to match a hashing power of 12 TH/s, then suddenly tomorrow, 50% of the hashing power dropped off the face of the earth, it would require only slightly more than 6 TH/s of hashing power to overcome the rest of the network.

What matters is the hashing power NOW.  The difficulty attempts to retroactively match the hashing power, but even that only happens every 2 weeks.  It is basically the most accurate indicator of total hashing power, second only to looking at how often blocks are being found, but still isn't a perfect indicator by any means.  And it doesn't mean that anyone who has hashing power greater than the difficulty could stage a 51% attack - it is a lot more involved than that.

[rjk]

no, its not what you were saying. what i said is completely true. difficulty has nothing to do with the 51% takeover. raising or lowering the difficulty would have 0 effect on the 51% takeover. bottom line is that you need 51% of the network's total hashing power.

The difficulty directly affects how much power you need to have in order to achieve 51%, correct?

Technically, no. Because the difficulty adjusts itself with a certain latency. Right now the difficulty is 1.7M and the hash rate is 12 Thash/s. So you need 12 Thash/s right now to perform a majority attack. But if tomorrow the hash rate drops to 6 Thash/s for whatever reason, then you only need 6 Thash/s, while the difficulty would still be 1.7M.

OK I can see that happening in the face of a sudden dramatic drop in network hash power, which at this point I find unlikely. Taking your example further, we could say that for a diff of 1.7m, a drop to 3Th/s would need only 3 more Th/s to 51%, however it would be extremely boring because the network would be functioning at half speed and everything would take twice as long to occur.

[molecular]

Besides the fact that a higher difficulty will make a 51% attack that much harder, yeah. No really.

A 51% attack happens when blocks are being found way faster than the difficulty allows. Raise the difficulty, and bang - 51% is out of commission, at least until he can get his speed up again.

Huh? A 51% attack could also be called a 51% takeover, and happens when 1 person owns 51% of the hashing power, and takes control of the entire network (rewrites the blockchain). The larger the network is, the harder for 1 person to own 51%. It's a LOT harder to gain 51% of a 200TH/s network than it is a 2TH/s network, for example. A larger network is inherently more resilient to a 51% takeover, you see?

A larger difficulty is just a side effect of that larger network. The difficulty of that 200TH/s network will be 100x that of the 2TH/s network, right? The difficulty itself doesn't actually protect against an attack (or takeover), but the factors that caused the difficulty to raise are what protect us.

Exactly, that's what I was saying.

no, its not what you were saying. what i said is completely true. difficulty has nothing to do with the 51% takeover. raising or lowering the difficulty would have 0 effect on the 51% takeover. bottom line is that you need 51% of the network's total hashing power.

The difficulty directly affects how much power you need to have in order to achieve 51%, correct? Therefore, for a network speed of (say) 12Th/s, you would need just over 12 MORE Th/s in order to be 51% of the network. However, if the difficulty was set for a network speed of 120Th/s, you would then need to have more than 120Th/s of hash power in order to be 51%.

The difference between a 12Th/s network and a 120Th/s network is directly related to the difficulty. If the difficulty were set for a 12Th/s network, and someone showed up with 120Th/s, he would have far more than 51% and the difficulty would rise to meet the hash power he is bringing online. In contrast, if someone threw 12Th/s at a network that was running at 120Th/s, he would be unable to do anything more than get a lot of coins, because he cannot overcome the difficulty to achieve 51%.

I'm not sure where the misunderstanding is. The network becomes more difficult to attack, directly as a result of how much power gets put up against it.

You said: "A 51% attack happens when blocks are being found way faster than the difficulty allows", which is simply wrong:

• blocks can be found way faster then the difficulty "allows" and no 51%-attack is happening (noone having more than 50% of the hashpower), for example when many individual miners join the network because they get their jalapenos delivered.
• a 51% attack can happen even when block are being found according to current difficulty (not faster than "allowed"). The attacker could have build up the hashing power slowly in the past.

The other things you said in subsequent posts are correct as far as I can tell. EDIT: maybe not, which began being debated while I wrote my post.

[mrb]

OK I can see that happening in the face of a sudden dramatic drop in network hash power, which at this point I find unlikely. Taking your example further, we could say that for a diff of 1.7m, a drop to 3Th/s would need only 3 more Th/s to 51%, however it would be extremely boring because the network would be functioning at half speed and everything would take twice as long to occur.

What I am saying is true even if the hash rate suddenly increases by 2x. If it does, then you would need 24 Thash/s to perform a majority attack, while the difficulty would still be 1.7M.

There. I gave you 3 scenarios where you would need either 6 Thash/s, or 12 Thash/s, or 24 Thash/s, to perform an attack, and in all 3 cases the difficulty would still be 1.7M for up to 2016 blocks.

The bottom line is that the difficulty factor is an average of the last full chunk of 2016 blocks, whereas a majority attack only needs to be executed for a handful of blocks (depending on how many confirmations are required by the recipient you are planning to double spend), and the global hash rate can vary either up or down quickly between these 2 intervals.

[imsaguy]

OK I can see that happening in the face of a sudden dramatic drop in network hash power, which at this point I find unlikely. Taking your example further, we could say that for a diff of 1.7m, a drop to 3Th/s would need only 3 more Th/s to 51%, however it would be extremely boring because the network would be functioning at half speed and everything would take twice as long to occur.

What I am saying is true even if the hash rate suddenly increases by 2x. If it does, then you would need 24 Thash/s to perform a majority attack, while the difficulty would still be 1.7M.

There. I gave you 3 scenarios where you would need either 6 Thash/s, or 12 Thash/s, or 24 Thash/s, to perform an attack, and in all 3 cases the difficulty would still be 1.7M for up to 2016 blocks.

The bottom line is that the difficulty factor is an average of the last full chunk of 2016 blocks, whereas a majority attack only needs to be executed for a handful of blocks (depending on how many confirmations are required by the recipient you are planning to double spend), and the global hash rate can vary either up or down quickly between these 2 intervals.

That is part of the reason that people are told to wait for at least 6 confirmations.  If people actually follow that (the major exchanges do), it means the 51% has to drop at least that many blocks in a row to be successful.

[Smoovious]

Difficulty, is simply a reflection of how much hash power there was on average during the last cycle, in order to attempt to normalize the block found rate over the long term.

Difficulty is not a function of how much power is needed to do what, merely a reflection of it. (in this particular case)

Raw hash power is all that matters when it comes to a 51% attack. That, and the ability to sustain a 51% proportion of the total hash power over the period of time the attacker needs to succeed in whatever goal he has. What the difficulty actually is at any given time, doesn't play a part in this.

-- Smoov

[btharper]

Higher difficulty makes a 51% attack harder because it means there's generally more people (and hash power) in the network, and it's generally harder to find a block. But no matter what the difficulty is at having 51% of the hashing power in the network still lets you do the same attack.

The 51% attack involves mining the block chain the way you want it, if you pull it off you can exclude any transaction you want, so you could for example do the following: Exclude all transactions and stop spending, exclude Alice's payment to Bob (so Alice could not pay for Bob's services), possibly double spend coins, create large number of orphaned blocks. There's likely other nasty things you can do, but it involves being able to create the longest chain and not letting everyone else work together to outperform you.

[Vorksholk]

Something I'd like to point out though...

The thing with a 51% attack is that that person has to get 51% or more of the total hashing power of the network. Some of the difficulty of getting this 51% is the price of the products needed to reach this. A while ago with CPU mining, the hash rate was way lower, but the hardware available at the time was way slower than GPU/FPGA/ASIC. With the move to GPU, the network hash rate goes way up, but so does the amount of bang for your buck, so it becomes proportionally about as easy to 51% (except for the general increasing adoption trend, this ignored for now). With the move the FPGA, it wasn't so crazy as before from CPU to GPU, but hardware did go down in price (especially for electricity to run them), but the hash rate would increase as well.

What's not exactly so important is that it will be 200TH, but how hard it would for someone to get their hands on another 202TH.

[TheHarbinger]

What's not exactly so important is that it will be 200TH, but how hard it would for someone to get their hands on another 202TH.

4 to 6 weeks hard.


Sorry, couldn't resist.