Jimmy2011
|
|
August 31, 2011, 05:11:14 AM |
|
Let's assume we lose almost all the rest of the miners. What's the longest time that can pass until the next difficulty adjustment? We can't really have a discussion unless you give actual numbers. The longest time it can take is infinite, with 0 miners. Don't know if that really answers your question or not. I'm sorry if "almost all" means 0 to you. Let me try some easier questions for you. The FAQ states: it can lower difficulty 400% in 12 hours
First, what does 400% lower difficulty mean given a current difficulty of 18,246? Why hasn't the difficulty lowered in the last 13+ hours? If you have a look at the code, you would understand everything what 400% means. Actually, CoinHunter is just promoting for solidcoin. I think CoinHunter is also an excellent programmer through the thread. Apart from that, I can't see some other excellent points. P.S. "400%" is not invented by SolidCoin. Everyone can see this "400%" is there in Bitcoin. What CoinHunter did was that he promoted it as a special feature belonging to SolidCoin.
|
|
|
|
coblee
Donator
Legendary
Offline
Activity: 1654
Merit: 1351
Creator of Litecoin. Cryptocurrency enthusiast.
|
|
August 31, 2011, 05:29:08 AM |
|
They don't even have to turn off mining 1. Pull a mining cartel attack and fork the chain when another miner finds a block. 2. Start fucking with block timestamps. 3. . 4. Near-infinite number of blocks! Similar things are possible with BTC so not sure of the relevance to only SC? Things need to be done to ensure network balance, both SC and BTC are powerless in that regard. Howeer like I said SCGuild is concerned with such things and I'm sure he would realize he would face consequences if he attacked peoples investments. Other pools, who knows, there has been open talk from people like LukeJR (elegius owner) to run malicious attacks on IXCoin. So obviously this kind of behaviour isn't unheard of when it comes to pools. I talked to ArtForz on IRC and he explained this 51% attack to me. This attack is only possible with this retargeting algorithm introduced in SolidCoin. So it doesn't affect Bitcoin. The gist of it is that you fork the block chain and keep finding blocks on your own without reporting them. You fake the timestamps to make it seem like it took 4 times as long to find blocks. After this period of 240 blocks, the difficulty will be reduced to 25% of what it was. Since the difficulty can only climb by 11% each time, you can do 13 fast periods before the difficulty is back to where it was. Then you introduce this blockchain to the rest of the network. Since you were hashing at >51%, your block chain will be the longest and will be accepted by everyone. ArtForz calculated that you can make ~3.5 times the number of coins that would normally get. The effect can also be compounded. So if you do 2 slow periods and reduce the difficulty by %25^2 (or 6.25%), you can make about 3.5^2 (~12.25) times more coins. While 51% attacks are not likely, they are still a possibility. Especially when the reward can be so huge. If you just do 4 slow periods, you get 150 times the normal output.
|
|
|
|
coblee
Donator
Legendary
Offline
Activity: 1654
Merit: 1351
Creator of Litecoin. Cryptocurrency enthusiast.
|
|
August 31, 2011, 05:36:25 AM |
|
raise by a maximum of 10% every 12 hours
So that's wrong too? There's no 12 hour calculation in the difficulty adjustment period at all? It's simply every 240 blocks exactly, no matter how fast or slow those blocks were found? Exactly. So it could possibly takes months for the difficulty to retarget. But since SolidCoin is set to retarget every 12 hours ideally, it won't be as bad as what we are seeing with namecoin. I0coin tried to fix that by introducing retargeting every X blocks or every 2 weeks (whichever is faster). But there was a coding bug that caused the whole network to crash when the 2 weeks came up.
|
|
|
|
Kartoff
|
|
August 31, 2011, 05:40:29 AM |
|
So when difficulty will be changed ?
|
|
|
|
CoinHunter (OP)
|
|
August 31, 2011, 05:44:21 AM |
|
I talked to ArtForz on IRC and he explained this 51% attack to me. This attack is only possible with this retargeting algorithm introduced in SolidCoin. So it doesn't affect Bitcoin. The gist of it is that you fork the block chain and keep finding blocks on your own without reporting them. You fake the timestamps to make it seem like it took 4 times as long to find blocks. After this period of 240 blocks, the difficulty will be reduced to 25% of what it was. Since the difficulty can only climb by 11% each time, you can do 13 fast periods before the difficulty is back to where it was. Then you introduce this blockchain to the rest of the network. Since you were hashing at >51%, your block chain will be the longest and will be accepted by everyone. ArtForz calculated that you can make ~3.5 times the number of coins that would normally get. The effect can also be compounded. So if you do 2 slow periods and reduce the difficulty by %25^2 (or 6.25%), you can make about 3.5^2 (~12.25) times more coins.
While 51% attacks are not likely, they are still a possibility. Especially when the reward can be so huge. If you just do 4 slow periods, you get 150 times the normal output.
Well he is right that it introduces some more "fun" you can have with a chain at 51%, the point is if you're at 51% it's not healthy regardless whether you're on SC or BTC. At 51% and having a malicious attacker both chains are dead. Discussing the new ways you could die is kinda irrelevant. Do you understand what I'm saying?
|
|
|
|
CoinHunter (OP)
|
|
August 31, 2011, 05:49:58 AM |
|
If you have a look at the code, you would understand everything what 400% means. Actually, CoinHunter is just promoting for solidcoin. I think CoinHunter is also an excellent programmer through the thread. Apart from that, I can't see some other excellent points.
P.S. "400%" is not invented by SolidCoin. Everyone can see this "400%" is there in Bitcoin. What CoinHunter did was that he promoted it as a special feature belonging to SolidCoin.
Yes Bitcoin has a 4x limit on increasing or decreasing. However it is far too long, every 2016 blocks. Bitcoin also desires a "magic number" for it's blocks, and that is 10 minutes. SolidCoin on the other hand tries to stay _UNDER_ 3 minutes. To reduce variance and provide a fast network. As such the people thinking "coin generation goes too fast" are kinda missing the point, it is supposed to go fast, it isn't supposed to always be at 3 minutes and if the system is "gamed" by a miner or two to provide "artificially" higher speeds then that is also "ok". However I also stress network balance is very important and having SCGuild so powerful on the network (like deepbit is on btc) isn't good, so I'll be looking at ways to solve this going forward, even though I think the current situation will resolve itself soon.
|
|
|
|
coblee
Donator
Legendary
Offline
Activity: 1654
Merit: 1351
Creator of Litecoin. Cryptocurrency enthusiast.
|
|
August 31, 2011, 05:51:21 AM |
|
I talked to ArtForz on IRC and he explained this 51% attack to me. This attack is only possible with this retargeting algorithm introduced in SolidCoin. So it doesn't affect Bitcoin. The gist of it is that you fork the block chain and keep finding blocks on your own without reporting them. You fake the timestamps to make it seem like it took 4 times as long to find blocks. After this period of 240 blocks, the difficulty will be reduced to 25% of what it was. Since the difficulty can only climb by 11% each time, you can do 13 fast periods before the difficulty is back to where it was. Then you introduce this blockchain to the rest of the network. Since you were hashing at >51%, your block chain will be the longest and will be accepted by everyone. ArtForz calculated that you can make ~3.5 times the number of coins that would normally get. The effect can also be compounded. So if you do 2 slow periods and reduce the difficulty by %25^2 (or 6.25%), you can make about 3.5^2 (~12.25) times more coins.
While 51% attacks are not likely, they are still a possibility. Especially when the reward can be so huge. If you just do 4 slow periods, you get 150 times the normal output.
Well he is right that it introduces some more "fun" you can have with a chain at 51%, the point is if you're at 51% it's not healthy regardless whether you're on SC or BTC. At 51% and having a malicious attacker both chains are dead. Discussing the new ways you could die is kinda irrelevant. Do you understand what I'm saying? At 51%, bitcoin will not be dead. They can just do some double spending and maybe reverse some transactions. They can create near infinite amount of coins. But solidcoin would be dead, because they can create near infinite amount of coins for themselves.
|
|
|
|
CoinHunter (OP)
|
|
August 31, 2011, 05:52:02 AM |
|
So when difficulty will be changed ?
About 4-6 hours.
|
|
|
|
CoinHunter (OP)
|
|
August 31, 2011, 05:56:36 AM |
|
At 51%, bitcoin will not be dead. They can just do some double spending and maybe reverse some transactions. They can create near infinite amount of coins. But solidcoin would be dead, because they can create near infinite amount of coins for themselves.
A little education doesn't go astray . At 51% you are guaranteed to be able to always produce the longest chain and hence you can not just "do a few double spends" you can write the chain pretty much however you like. So yes, bitcoin would be dead because that "group" would effectively devalue the entirely currency as they plundered it. Mtgox would be bankrupt in a few hours/days. No business could any longer trust transactions unless there were massive amounts of confirmations, likely reaching days (even then they are still vulnerable). >51% and malicious = death. Bitcoin isn't designed to protect against it, and one of the only known ways is centralization.
|
|
|
|
Jimmy2011
|
|
August 31, 2011, 06:04:32 AM |
|
If you have a look at the code, you would understand everything what 400% means. Actually, CoinHunter is just promoting for solidcoin. I think CoinHunter is also an excellent programmer through the thread. Apart from that, I can't see some other excellent points.
P.S. "400%" is not invented by SolidCoin. Everyone can see this "400%" is there in Bitcoin. What CoinHunter did was that he promoted it as a special feature belonging to SolidCoin.
Yes Bitcoin has a 4x limit on increasing or decreasing. However it is far too long, every 2016 blocks. Bitcoin also desires a "magic number" for it's blocks, and that is 10 minutes. SolidCoin on the other hand tries to stay _UNDER_ 3 minutes. To reduce variance and provide a fast network. As such the people thinking "coin generation goes too fast" are kinda missing the point, it is supposed to go fast, it isn't supposed to always be at 3 minutes and if the system is "gamed" by a miner or two to provide "artificially" higher speeds then that is also "ok". However I also stress network balance is very important and having SCGuild so powerful on the network (like deepbit is on btc) isn't good, so I'll be looking at ways to solve this going forward, even though I think the current situation will resolve itself soon. Yes, lowering retargeting period make sense at a certain time range whether it is 3 minutes or 4 minutes. Variance reduction as you said maybe considered as the absolute variance, however, reduced variance in your retargeting period can't make it more stable from what is happening, which seems to be a sin(t) function I have said last day. You also realized it as a cycle, which is the stability problem, and at this moment I can't guess whether it will be more stable as time is going. Sometimes network balance is out of one's control, so we can just improve the algorithm to be more resistant to network unbalance.
|
|
|
|
Anonymous
Guest
|
|
August 31, 2011, 06:45:15 AM |
|
Solidcoin won't fail. Coinhunter has done a really good job setting this game up . The same people talking crap now will be shooting themselves in the head in 6 months knowing they could have bought at these prices. I think all cryptography will have a place in the future, and anyone thinking there is only room for one is a fucking moron. Pardon my french... See now i like you lol.
|
|
|
|
joulesbeef
Sr. Member
Offline
Activity: 476
Merit: 250
moOo
|
|
August 31, 2011, 07:27:43 AM |
|
the diff is about to change.. i wanted to get a little btc into the market before it did. Well this is one of the things i like better about sc over btc. I sent my btc to the exchange when solid had 20 blocks til the diff change.. I am sitting here at 5 confirms and sc has only 6 blocks left... gahhhh why you take so long to confirm bitcoin. I know some people think the faster block times arent a good thing but from where I am standing, I really wish these btc were scs.. when sending them to the market.. i could have been in bed 30 minutes ago
|
mooo for rent
|
|
|
coblee
Donator
Legendary
Offline
Activity: 1654
Merit: 1351
Creator of Litecoin. Cryptocurrency enthusiast.
|
|
August 31, 2011, 07:47:57 AM |
|
At 51%, bitcoin will not be dead. They can just do some double spending and maybe reverse some transactions. They can create near infinite amount of coins. But solidcoin would be dead, because they can create near infinite amount of coins for themselves.
A little education doesn't go astray . At 51% you are guaranteed to be able to always produce the longest chain and hence you can not just "do a few double spends" you can write the chain pretty much however you like. So yes, bitcoin would be dead because that "group" would effectively devalue the entirely currency as they plundered it. Mtgox would be bankrupt in a few hours/days. No business could any longer trust transactions unless there were massive amounts of confirmations, likely reaching days (even then they are still vulnerable). >51% and malicious = death. Bitcoin isn't designed to protect against it, and one of the only known ways is centralization. So I take that back about bitcoin not being dead with a 51% attack. With 51% and malicious, you can hurt bitcoin, and the whole bitcoin economy. You can effectively kill the bitcoin economy by not accepting any transactions into your blocks and not accepting blocks found by other miners. But you still cannot "write the chain pretty much however you like." You cannot spend money that doesn't belong to you. And you cannot earn more coins than you would otherwise. So your only incentive to gain 51% of the network hash is if you wanted to kill bitcoin, because you cannot benefit from it. So bitcoin's argument against someone doing a 51% attack is that they would benefit more to work with bitcoins (and earn 51% of the found coins) as oppose to against it. Unless of course, the government wants to shut down bitcoin and performs a 51% attack to kill it. But that's another story altogether. On the other hand with solidcoins, you can benefit yourself and not kill off solidcoin by performing a 51% attack. Sure the solidcoin economy will take a hit (mtgox/mybitcoin level hit), but it will likely recover. And the attacker would have created hundreds of thousands of solidcoins for himself during that attack and no one could do anything about it.
|
|
|
|
CoinHunter (OP)
|
|
August 31, 2011, 08:00:14 AM |
|
So I take that back about bitcoin not being dead with a 51% attack. With 51% and malicious, you can hurt bitcoin, and the whole bitcoin economy. You can effectively kill the bitcoin economy by not accepting any transactions into your blocks and not accepting blocks found by other miners. But you still cannot "write the chain pretty much however you like." You cannot spend money that doesn't belong to you. And you cannot earn more coins than you would otherwise. It's actually irrelevant about not being able to spend others coins, because you can continue to spend your own, you simply give the vendor the amount of confs they want, then take the chain back to a point in time before the TX and rewrite history. Because no one is going to beat your length there is nothing they can do about your rewritten history. The higher over 50% you are the quicker you can repeat this process. ie MtGox would be cleaned out or would be forced to shutdown, freezing the entire economy. Bitcoin dies. On the other hand with solidcoins, you can benefit yourself and not kill off solidcoin by performing a 51% attack. Sure the solidcoin economy will take a hit (mtgox/mybitcoin level hit), but it will likely recover. And the attacker would have created hundreds of thousands of solidcoins for himself during that attack and no one could do anything about it.
Yes you could manipulate the speed of the network to increase coins. But it's kinda irrelevant when you can just spend whatever you want over and over again. Either way there is utter chaos and doesn't end well for the chain. In both instances the malicious team have to work fast to get wealth out of the system before it is noticed, so in both instances it will be very obvious something is happening. It is why I don't understand why people support pools like SCguild or deepbit when they are too powerful for their respective networks. SCGuild at least will be moved down soon as other large pools join but it's just as likely deepbit will do the same to SC as it is now to BTC.
|
|
|
|
coblee
Donator
Legendary
Offline
Activity: 1654
Merit: 1351
Creator of Litecoin. Cryptocurrency enthusiast.
|
|
August 31, 2011, 08:08:57 AM |
|
So I take that back about bitcoin not being dead with a 51% attack. With 51% and malicious, you can hurt bitcoin, and the whole bitcoin economy. You can effectively kill the bitcoin economy by not accepting any transactions into your blocks and not accepting blocks found by other miners. But you still cannot "write the chain pretty much however you like." You cannot spend money that doesn't belong to you. And you cannot earn more coins than you would otherwise. It's actually irrelevant about not being able to spend others coins, because you can continue to spend your own, you simply give the vendor the amount of confs they want, then take the chain back to a point in time before the TX and rewrite history. Because no one is going to beat your length there is nothing they can do about your rewritten history. The higher over 50% you are the quicker you can repeat this process. ie MtGox would be cleaned out or would be forced to shutdown, freezing the entire economy. Bitcoin dies. On the other hand with solidcoins, you can benefit yourself and not kill off solidcoin by performing a 51% attack. Sure the solidcoin economy will take a hit (mtgox/mybitcoin level hit), but it will likely recover. And the attacker would have created hundreds of thousands of solidcoins for himself during that attack and no one could do anything about it.
Yes you could manipulate the speed of the network to increase coins. But it's kinda irrelevant when you can just spend whatever you want over and over again. Either way there is utter chaos and doesn't end well for the chain. In both instances the malicious team have to work fast to get wealth out of the system before it is noticed, so in both instances it will be very obvious something is happening. It is why I don't understand why people support pools like SCguild or deepbit when they are too powerful for their respective networks. SCGuild at least will be moved down soon as other large pools join but it's just as likely deepbit will do the same to SC as it is now to BTC. I don't think you got my point. Let me lay it out for you. Bitcoins: 51% attack => Make some money from cheating mtgox => Keep doing it and MtGox can't function and stops withdrawal and bitcoin dies. You can't make a lot of money, so no monetary incentive to do 51% attack. Solidcoins: 51% attack => Mint a lot of solidcoins for yourself => Stop doing it and solidcoin recovers. => Convert those coins you minted for Bitcoin/USD. You can make a lot of money, so there is monetary incentive for 51% attack.
|
|
|
|
CoinHunter (OP)
|
|
August 31, 2011, 08:13:57 AM |
|
I don't think you got my point. Let me lay it out for you.
Bitcoins: 51% attack => Make some money from cheating mtgox => Keep doing it and MtGox can't function and stops withdrawal and bitcoin dies. You can't make a lot of money, so no monetary incentive to do 51% attack. Solidcoins: 51% attack => Mint a lot of solidcoins for yourself => Stop doing it and solidcoin recovers. => Convert those coins you minted for Bitcoin/USD. You can make a lot of money, so there is monetary incentive for 51% attack.
No I get you you're just overstating the impact here. If someone did start messing with timestamps/speed/etc it would be detectable. It's not like a ninja attack in the night. It's the same with any >51% attack, they are very visible and you need to take advantage of them quickly before people realize. In both instances confidence in the network will be shattered.
|
|
|
|
Xenland
Legendary
Offline
Activity: 980
Merit: 1003
I'm not just any shaman, I'm a Sha256man
|
|
August 31, 2011, 08:21:42 AM |
|
Just have 100+ confirmations NO RISK!
|
|
|
|
HostFat
Staff
Legendary
Offline
Activity: 4256
Merit: 1208
I support freedom of choice
|
|
August 31, 2011, 08:23:19 AM Last edit: August 31, 2011, 08:42:12 AM by HostFat |
|
it seems that CoinHunter is trying to avoid the point of the discussion Bitcoin: only governmens will have interests to make this attack and it's not easy to do it. SolidCoin: It is easier to make the attack and everyone will have interests to do it. Good luck!
|
|
|
|
coblee
Donator
Legendary
Offline
Activity: 1654
Merit: 1351
Creator of Litecoin. Cryptocurrency enthusiast.
|
|
August 31, 2011, 08:28:47 AM |
|
No I get you you're just overstating the impact here.
If someone did start messing with timestamps/speed/etc it would be detectable. It's not like a ninja attack in the night. It's the same with any >51% attack, they are very visible and you need to take advantage of them quickly before people realize. In both instances confidence in the network will be shattered.
So this is all in theory, but it's good to think about it to help keep these coins secure. If you have >51% of the network, and want to perform the solidcoin attack I mentioned. You can work in secret and keep all your blocks to yourself and only announce them after you've successfully completed the attack. So no one would know that an attack is taking place until you release your longest blockchain and wipe out the other chain. After that, people will see that their confirmations got totally wiped out and their transactions reversed. And they will see that someone performed a 51% attack and mucked with the timestamps/speed/etc. Yes, confidence will take a big hit afterwards. I'm guessing on the order of magnitude as the mtgox hack or the mybitcoin theft. But I'm thinking it won't kill solidcoin similar to how mtgox/mybitcoin did not kill bitcoin. Let say this attack happened when there are 7 million solidcoins, and each solidcoin is worth $10, and this guy minted 150,000 coins. These numbers are equivalent to current number of bitcoins and mybitcoin theft. So what do you do then to save solidcoin? Only about 2% inflation, and no one really lost their coins. So this guy made $1.5 million. Do you: A) rewrite the blockchain to start at before the hack took place and try to kickstart the economy at that point? B) accept that there was this hack and fix the algorithm to prevent the hack in the future? If you do (A), that might totally kill solidcoin b/c how can one trust it anymore if you can just rewrite history. So you will probably do (B) and this guys comes away with $1.5 million.
|
|
|
|
3phase
Sr. Member
Offline
Activity: 313
Merit: 251
Third score
|
|
August 31, 2011, 08:49:06 AM |
|
All this discussion about someone making profit with Solidcoin is with a 51% attack is pointless:
If something like this happens, think of all the SC deposited at bitparking in the period where the attacker was building his alt chain. Would they become invalid or not? If doublec was sleeping at the time the new chain is released and accepted, all those who had deposited in that period would be able to trade non-existing balances, thus buying REAL BTCs with FAKE SC.
Unless bitparking checks the blockexplorer before every transaction (which I don't think it does), that would f**k it up completely.
When doublec (who BTW has been a really hard working guy and his contribution to the SC community is definitely underestimated) wakes up and finds the mess, don't you think he would close the exchange for good, having lost nearly a fortune?
What then would the attacker do with his funny Solidcoins? Sell them to whom? Where?
All in all, a 51% attack as you (or ArtForz originally) have described it would destroy Solidcoin just the same as could happen to Bitcoin with a 51% attack. The difference is of course in the incentive to exploit the network majority for profit. In the case of Solidcoin, I think you understand that this is not possible. In the case of Bitcoin, it certainly makes sense.
I guess doublec if you read this, you might not be sleeping properly.
The plot thickens...
|
|
|
|
|