Cleanup: I'll attack some coins - I owned APEXcoin for 90 blocks

[inBitweTrust]

Start with Litecoin Testnet
If you are right, many honest coins will pay you a tax for your work.

Cheers

Litecoin would be too expensive to attack. Not likely.

https://bitcoinwisdom.com/litecoin/difficulty

1,282 GH/s

[Rent_a_Ray]

The solution is to raise the diff for POS Coins? (different algo?)
I am thinking about a rollback system with wallet voting.

[Maestro1]

And a warning: most POW coins are not safe! it's easy to cheaply rent enough hashrate to attack a lot of coins with low to medium hashrate. Only the biggest ones are good.

The difference for PoW is that it is costly to even try to do such an attack. For PoS, there's only the initial cost of the coins. There's no running cost for tries, retries, etc.

[inBitweTrust]

The difference for PoW is that it is costly to even try to do such an attack. For PoS, there's only the initial cost of the coins. There's no running cost for tries, retries, etc.

...or the effort taken to steal the coins, or the effort taken to borrow the coins(bank, payment processor, exchange, ponzi, ect..), or the effort made to compromise the wallet of a whale, or the effort made to take many loans out....

Some of the above can be applied to PoW as well, but if you compromise a mining pool operator and you use their miners than there will be an instant reaction and correction to stop such an attack from re-occurring or happening in the first place because there is large amounts of investments and expenses to be paid in mining.

With an attack on a PoS , the transactional history is tainted and any rollbacks ruin the fungibility or anonymity of the currency.

[achimsmile]

Haven't read all the posts so it may be mentioned before. Current PoS implementations use some sort of wallet weight as a way to handle competitions for blocks. Wallet weight is basically the amount of coins in the wallet. So if I have a big wallet with very small coin blocks (i.e. I transfer the coins in thousands of small transactions), I have the chance of staking a lot of new blocks continuously with the full weight of my wallet each time. That's the problem.

I have 1% of all stake in 1 wallet. My chance to generate a block is 1%
I have 1% of all stake in 1000 wallets. My chance to generate a block is 1%

I see no problem here. (Nxt)

edit: I guess what you meant is:
I have 1% of all stake in my wallet, and I transferred it there in 1000 small transactions.
I still don't see a problem for PoS.

Yes.

Your chance of generating a block is 1%, next block again 1%, next again 1%. That's a problem. The high weight of your whole coins are used for each individual small stake, increasing your chance every single time which makes you a lot more likely to find consecutive blocks at zero cost.

Ideally your chance shouldn't be 1%, the secure way is 0.001% since you transferred the money in 1000 small transactions. You should either have a big block of coins (1 big transaction), so you have a 1% chance ONCE, or you have 1000 small blocks, so you will have 1000 chances of 0.001%. You should not have a 1% chance, 1000 times.

Sorry, I don't get what you mean.
1000 x 0.001% = 1 x 1%
In PoS your stake matters, not your amount of transactions.

[inBitweTrust]

Sorry, I don't get what you mean.
1000 x 0.001% = 1 x 1%
In PoS your stake matters, not your amount of transactions.

Yes, I am no expert on NxT , but from what I understand a higher weight is given to large stakes in a single account vs many small ones as well.

[bitcoinpaul]

Sorry, I don't get what you mean.
1000 x 0.001% = 1 x 1%
In PoS your stake matters, not your amount of transactions.

Yes, I am no expert on NxT , but from what I understand a higher weight is given to large stakes in a single account vs many small ones as well.

Indeed. The gain of weight with one big account is small but exists.

[achimsmile]

but if you compromise a mining pool operator and you use their miners than there will be an instant reaction and correction to stop such an attack from re-occurring or happening in the first place

Unless you have a time machine, you can't tell which mining pools will attack before it happened.

(only if you independently verify you have found a block, but the pool does not broadcast it. This is unlikely, because most miners check the stats of their miner on the pool website, which could be faked.)

Realizing your pool attacks and switching miners to a different pool is not instant.

[Maestro1]

The difference for PoW is that it is costly to even try to do such an attack. For PoS, there's only the initial cost of the coins. There's no running cost for tries, retries, etc.

...or the effort taken to steal the coins, or the effort taken to borrow the coins(bank, payment processor, exchange, ponzi, ect..), or the effort made to compromise the wallet of a whale, or the effort made to take many loans out....

Some of the above can be applied to PoW as well, but if you compromise a mining pool operator and you use their miners than there will be an instant reaction and correction to stop such an attack from re-occurring or happening in the first place because there is large amounts of investments and expenses to be paid in mining.

With an attack on a PoS , the transactional history is tainted and any rollbacks ruin the fungibility or anonymity of the currency.

The point I was making was that attacking PoW has a running economical cost while attacking PoS does not.

[Maestro1]

Haven't read all the posts so it may be mentioned before. Current PoS implementations use some sort of wallet weight as a way to handle competitions for blocks. Wallet weight is basically the amount of coins in the wallet. So if I have a big wallet with very small coin blocks (i.e. I transfer the coins in thousands of small transactions), I have the chance of staking a lot of new blocks continuously with the full weight of my wallet each time. That's the problem.

I have 1% of all stake in 1 wallet. My chance to generate a block is 1%
I have 1% of all stake in 1000 wallets. My chance to generate a block is 1%

I see no problem here. (Nxt)

edit: I guess what you meant is:
I have 1% of all stake in my wallet, and I transferred it there in 1000 small transactions.
I still don't see a problem for PoS.

Yes.

Your chance of generating a block is 1%, next block again 1%, next again 1%. That's a problem. The high weight of your whole coins are used for each individual small stake, increasing your chance every single time which makes you a lot more likely to find consecutive blocks at zero cost.

Ideally your chance shouldn't be 1%, the secure way is 0.001% since you transferred the money in 1000 small transactions. You should either have a big block of coins (1 big transaction), so you have a 1% chance ONCE, or you have 1000 small blocks, so you will have 1000 chances of 0.001%. You should not have a 1% chance, 1000 times.

Sorry, I don't get what you mean.
1000 x 0.001% = 1 x 1%
In PoS your stake matters, not your amount of transactions.

Yes, either 1000 x 0.001% or 1 x 1%. It's not the number of transactions, it's the number of coin blocks. Each block stakes once until it matures again (8 hours or whatever), if you have 1000 small blocks in your wallet, you can stake 1000 times right after the other and current implementations give you the full weight of your wallet in the competition for staking, giving you a great chance of finding consecutive blocks.

So basically in simple terms, current implementations give you 1000 x 1%. That's the problem.

[achimsmile]

Sorry, I don't get what you mean.
1000 x 0.001% = 1 x 1%
In PoS your stake matters, not your amount of transactions.

Yes, I am no expert on NxT , but from what I understand a higher weight is given to large stakes in a single account vs many small ones as well.

yes, but Maestro1 was talking about:

I have 1% of all stake in my wallet, and I transferred it there in 1000 small transactions.

[inBitweTrust]

The point I was making was that attacking PoW has a running economical cost while attacking PoS does not.

Unfortunately, for Bitcoin , there are some scenarios where there is only a cost in effort to perform an attack.
2 compromised mining pools could perform the attack right now.

Realizing your pool attacks and switching miners to a different pool is not instant.

Instant in the sense that 2-3 double spends would occur before being caught.


Bitcoin has its on set of security problems. Still more secure than any other coin but insecure nonetheless.

When judging the security of a coin there are many other aspects to consider such as the amount of peer review, the amount of nodes,
the amount of implementations and stacks, the amount of researchers auditing, the amount of developers, the total market cap which
 incentivizes would be attackers from exploiting the currency, ect....

[achimsmile]

Yes, either 1000 x 0.001% or 1 x 1%. It's not the number of transactions, it's the number of coin blocks. Each block stakes once until it matures again (8 hours or whatever), if you have 1000 small blocks in your wallet, you can stake 1000 times right after the other and current implementations give you the full weight of your wallet in the competition for staking, giving you a great chance of finding consecutive blocks.

So basically in simple terms, current implementations give you 1000 x 1%. That's the problem.

Which specific PoS implementation are you talking about?

[cynicSOB]

And a warning: most POW coins are not safe! it's easy to cheaply rent enough hashrate to attack a lot of coins with low to medium hashrate. Only the biggest ones are good.

Start with Litecoin Testnet
If you are right, many honest coins will pay you a tax for your work.

Cheers

really? meh.... why whould they pay?

Yes, I am no expert on NxT , but from what I understand a higher weight is given to large stakes in a single account vs many small ones as well.

Indeed. The gain of weight with one big account is small but exists.

This would mitigate the attack, but it's not a good thing as it benefits the "rich" over the "poor". If I'm not mistaken, in nxt the difference is so small that it doesn't matter.
BTW: you say "accounts" but is it accounts or tx outputs that are used to forge? in PPC you don't mint with your addresses, it's unspent outputs that are used to mint.

I just made my "cynicSOB" account at nxtforums, so I have the same user name there.

[cynicSOB]

OP doesn't need cheer he needs BTC. It takes money to own a coin

hehe.. well said cheers don't hurt though 

[cynicSOB]

Litecoin would be too expensive to attack. Not likely.

https://bitcoinwisdom.com/litecoin/difficulty

1,282 GH/s

Not litecoin, but litecoin testnet: more like 11 MH/s
but is it worth it?
there's no need to prove anything by attacking, just go to hash rental site and see how much it costs to rent that for an hour.

edit: I felt attacking POS was needed to prove the point, because the point has been made and ignored many times.

I'll say it again just in case: you don't need 51% of the supply, you need 51% of the staking weight (actively forging stake, or whatever you call it). And for most coins, that's about 5% of the supply. There's no secret attack. Only splitting the staking weight so it's not affected by each block you mine/mint/forge/whatever.

[Rent_a_Ray]

really? meh.... why whould they pay?

Because of the safety and the future of CryptoCurrencies.
If you are able to attack LTC testnet I would spend some bucks. (If you tell us how to reconstructing the attack)

Cheers,
Ray

[inBitweTrust]

Not litecoin, but litecoin testnet: more like 11 MH/s
but is it worth it?
there's no need to prove anything by attacking, just go to hash rental site and see how much it costs to rent that for an hour.

edit: I felt attacking POS was needed to prove the point, because the point has been made and ignored many times.

I'll say it again just in case: you don't need 51% of the supply, you need 51% of the staking weight (actively forging stake, or whatever you call it). And for most coins, that's about 5% of the supply. There's no secret attack. Only splitting the staking weight so it's not affected by each block you mine/mint/forge/whatever.

There are already many examples of both SHA256 and Scrypt coins being attacked so wasting time on the litecoin testnet where coins with the exact same algo have been attacked is not needed.

With PoS there are different variations where they make wild claims like impossible to be attacked and despite you proving yourself with Apex they will merely deflect by claiming their algo is different. Better to focus on PoS.

[YarkoL]

This would mitigate the attack, but it's not a good thing as it benefits the "rich" over the "poor".

I hardly think that would deter anyone who has
investment in the coin. Just the same as with centralized
checkpointing. Philosophy drops when money talks.

[Rent_a_Ray]

This would mitigate the attack, but it's not a good thing as it benefits the "rich" over the "poor".

I hardly think that would deter anyone who has
investment in the coin. Just the same as with centralized
checkpointing. Philosophy drops when money talks.

Checkpointing is always centralized. Owners of the private master key (like theymos) are able to set checkpoints on demand.

Cheers,
Ray