Bitcoin Forum

Alternate cryptocurrencies => Altcoin Discussion => Topic started by: nesic1 on July 11, 2013, 11:39:35 PM



Title: is dgc 51% attack?
Post by: nesic1 on July 11, 2013, 11:39:35 PM
Network hashrate: 986.14 MH/s Pool hashrate: 730.81 MH/s
so? when other coin it at it all of you scream, what is now?


Title: Re: is dgc 51% attack?
Post by: mustyoshi on July 11, 2013, 11:40:43 PM
It's only an attack if they do something malicious with that power.


Title: Re: is dgc 51% attack?
Post by: seleme on July 12, 2013, 12:22:51 AM
It's only an attack if they do something malicious with that power.

this


Title: Re: is dgc 51% attack?
Post by: erk on July 12, 2013, 12:28:38 AM
Network hashrate: 986.14 MH/s Pool hashrate: 730.81 MH/s
so? when other coin it at it all of you scream, what is now?

They pool hash rates are an estimate calculated from the block numbers and the timestamps, they don't really tell you might about what is going on with the network.

The shorter the confirmation time, the less time available for an attacker to calculate valid blocks to subvert the chain.




Title: Re: is dgc 51% attack?
Post by: nearmiss on July 12, 2013, 12:57:14 AM
nevertheless, not a great state for any coin to be in.....


Title: Re: is dgc 51% attack?
Post by: Scrappy Do on July 12, 2013, 01:21:39 AM
nevertheless, not a great state for any coin to be in.....

 Not a bad one either. This coin has challenged folks to find flaws, and to date, is flawless. Also, they are growing by leaps and bounds. For this said 51%, one would have to hold the coins. and as every big rig guy joins 41kh/s there is another joining as well. This is not to say they could not send coins to each other, but why would you waste the power to kill a coin? 41 kh/s a sec is around 1200 a month power here and my rate is .078 per. Do the math.


Title: Re: is dgc 51% attack?
Post by: seleme on July 12, 2013, 01:33:22 AM
nevertheless, not a great state for any coin to be in.....

That for sure.. never a good thing to have a single party with that amount of network hashrate


Title: Re: is dgc 51% attack?
Post by: DeathAndTaxes on July 12, 2013, 01:40:12 AM
The shorter the confirmation time, the less time available for an attacker to calculate valid blocks to subvert the chain.

Nonsense.  Where do people come up with this junk?


Title: Re: is dgc 51% attack?
Post by: usahero on July 12, 2013, 01:53:39 AM
The shorter the confirmation time, the less time available for an attacker to calculate valid blocks to subvert the chain.

Nonsense.  Where do people come up with this junk?

In the lol-land of wishful thinking...




Title: Re: is dgc 51% attack?
Post by: erk on July 12, 2013, 02:02:16 AM
The shorter the confirmation time, the less time available for an attacker to calculate valid blocks to subvert the chain.

Nonsense.  Where do people come up with this junk?


http://we.lovebitco.in/bitcoin-paper/#ch11


Title: Re: is dgc 51% attack?
Post by: DeathAndTaxes on July 12, 2013, 02:46:33 AM
The shorter the confirmation time, the less time available for an attacker to calculate valid blocks to subvert the chain.

Nonsense.  Where do people come up with this junk?


http://we.lovebitco.in/bitcoin-paper/#ch11

Yeah that cite doesn't prove what you said.  It doesn't because what you said was wrong.

Difficulty and block window are arbitrary.  The amount of hashpower that the attacker has relative to the global hashpower is all that matters.

If the attacker has 10% of the hash power he has a 10% chance of solving the next block.
He has a 10% chance of solving the next block if the block window is 600 seconds or 1 second.

The risk with many of these yet-another-copycoins is the low hashpower on the chain.  That makes it far easier for an attacker to have the resources necessary for an attack.



Title: Re: is dgc 51% attack?
Post by: groll on July 12, 2013, 02:51:44 AM
The shorter the confirmation time, the less time available for an attacker to calculate valid blocks to subvert the chain.

Nonsense.  Where do people come up with this junk?


http://we.lovebitco.in/bitcoin-paper/#ch11

ok you need to put that in perspective this is catchup, 51% don`t have to be in catchup. And 75% can catch up easily for 5-10 blocks over 20

In recent 51% the attacker seems to have use it in it`s favor so he try to get some block before the honest get one (as choosen block for the fork usually are when the honest chain have bad luck in the few sample I have looked at.) then make its false transaction in honest chain that he want to revert continue to race until honest is long enough for the double spend to be accepted, since he was already in front and have more power is chain is longer and then he release it invalidating 5-7 block in honest chain with a longer chain. so this can easily be automated to start when you have the advantage(fast block don't help as it should be miliseconds to have impact).

i intentionnaly remove variance and some probability to just keep average to make it easy to understand in the following explanation(simplified but still valid for around 50% chance):

Faster block means nothing in itself as difficulty to find block need to be factored to balance the real difficulty to make the catchup at 51% hash power starting 1 minutes later will take the same time in average for block every seconds the blocks every minutes. in first one he start 1 block back in seconds he start 60 block back. but difficulty to find a block in the 1 seconds is 60 times less then the one for 1 minutes. The number of block for 1 minutes should be around 50 so after 50 minutes he should catchup. for 1 seconds blocks it will be after 50*60 to catchup the 60 blocks yes the number is way bigger but the hashpower is the same the difficulty to find a block is 60 times less so still 50 minutes.

what the paper explain is that catching up 60 block for the 1 minutes will takes a lots of time 50 hours (50*60)

But faster block because hash power is high makes the attack more difficult as 51% require more hash power from the attacker. so faster then target in correct or high difficulty is a good sign. slow block the target and low difficulty means vulnerable (very slow and high diff also means vulnerable)

in fact script with less then 1Gh/s are subject to 51%, above 2Gh/s should be not a bad spot.


Title: Re: is dgc 51% attack?
Post by: erk on July 12, 2013, 03:13:41 AM
Here is a paper on the theory of a bitcoin double spend attack

https://bitcoil.co.il/Doublespend.pdf


Quote
7 Conclusions
In this paper we have explained the basic structure of the Bitcoin blockchain, the protection
it gives against double-spending, and the ways in which this protection can be undermined.
We have derived the probability for a successful double-spend, and tabulated it in various
ways. We have also brie y discussed the conditions in which a double-spending attack can
be economical, and hence likely. In so doing we have dispelled some popular myths, such as
the absolute security believed to be granted by waiting for 6 con rmations, or the length of
time waited (as opposed to the number of con firmations in terms of discrete blocks) as an
allegedly major factor in determining security.

So the quicker you can get to a high number of confirmations, the safer you are.


This is most likely why during the recent FTC attack, it was carried out when the FTC confirmation time had blown out to over half an hour.





Title: Re: is dgc 51% attack?
Post by: groll on July 12, 2013, 03:46:33 AM


So the quicker you can get to a high number of confirmations, the safer you are.

mostly true. at same difficulty it's true. But 10 confirm at 0.01 diff at 10 seconds each  is way easier to revert then 5 confirms at diff 87 at 1.5 minutes each

~50kh/s vs 4Gh/s for the 51% attack

but for same diff 10 is safer then 5 for sure


Quote

This is most likely why during the recent FTC attack, it was carried out when the FTC confirmation time had blown out to over half an hour.



hash power was very low so the attacker had >70% and is likely to have used the method described in the paper with 1 block before start so he had 0% chance to fail