Solidcoin 2 exploit: Difficulty coaster

[tacotime]

SC2 has a special difficulty algorithm that allows for vast decreases in difficulty per 360 blocks (~200%) but only small increases (~14%).  It seems that this could be exploited by using a script that does the following:
1.) Mines SC2 when ((number of blocks / 360) % 2 == 0), that is, every other difficulty difficulty adjustment period.
2.) Mines BTC when ((number of blocks / 360) % 2 != 0).  This would allow you to use your cards for BTC in the meantime.

If a large number of people use such a script, for example 90% of the miners, the difficulty will quickly drop to what it would be if only 10% of the network were mining.  This would maximize return per watt for people mining SC2, yielding bitcoin when they aren't mining SC2.

[sd]

SC2 has a special difficulty algorithm that allows for vast decreases in difficulty per 360 blocks (~200%) but only small increases (~14%).  It seems that this could be exploited by using a script that does the following:
1.) Mines SC2 when ((number of blocks / 360) % 2 == 0), that is, every other difficulty difficulty adjustment period.
2.) Mines BTC when ((number of blocks / 360) % 2 != 0).  This would allow you to use your cards for BTC in the meantime.

If a large number of people use such a script, for example 90% of the miners, the difficulty will quickly drop to what it would be if only 10% of the network were mining.  This would maximize return per watt for people mining SC2, yielding bitcoin when they aren't mining SC2.

This is an old an very well known attack. Using it this way is an interesting way to compensate for the 5 SC reward per block but it will still take a very long time to get the million SC you need to free the network from coinhunter's talons. Even if you do manage that you are still using a blockchain with a tax.

Give up on SolidCoin, it's too compromised to even be worth abusing.

[Spacy]

SC2 has a special difficulty algorithm that allows for vast decreases in difficulty per 360 blocks (~200%) but only small increases (~14%).  It seems that this could be exploited by using a script that does the following:
1.) Mines SC2 when ((number of blocks / 360) % 2 == 0), that is, every other difficulty difficulty adjustment period.
2.) Mines BTC when ((number of blocks / 360) % 2 != 0).  This would allow you to use your cards for BTC in the meantime.

If a large number of people use such a script, for example 90% of the miners, the difficulty will quickly drop to what it would be if only 10% of the network were mining.  This would maximize return per watt for people mining SC2, yielding bitcoin when they aren't mining SC2.

The difficulty won't drop, because miners with less than 1M SC can't mine trusted blocks. So it hasn't any effect on the actual blocks/time. You can decide to not mine SC, and the difficulty will drop a little bit.

[sd]

The difficulty won't drop, because miners with less than 1M SC can't mine trusted blocks. So it hasn't any effect on the actual blocks/time. You can decide to not mine SC, and the difficulty will drop a little bit.

The difficulty will drop. The difficulty for trusted blocks and normal blocks isn't related.

[Spacy]

The difficulty won't drop, because miners with less than 1M SC can't mine trusted blocks. So it hasn't any effect on the actual blocks/time. You can decide to not mine SC, and the difficulty will drop a little bit.

The difficulty will drop. The difficulty for trusted blocks and normal blocks isn't related.

Yep, they are not related. You can't affect the difficulty by trying to mine/not mine trusted blocks, because you can't solve one withouth 1M SC, it doesn't matter if you stop the miner on even blocks or not. So you can only affect the SC difficulty by mining/not mining odd blocks. But you can stop the miner on even blocks and switch to some other chain until the trusted block is solved.

[BeeCee1]

SC2 has a special difficulty algorithm that allows for vast decreases in difficulty per 360 blocks (~200%) but only small increases (~14%).  It seems that this could be exploited by using a script that does the following:
1.) Mines SC2 when ((number of blocks / 360) % 2 == 0), that is, every other difficulty difficulty adjustment period.
2.) Mines BTC when ((number of blocks / 360) % 2 != 0).  This would allow you to use your cards for BTC in the meantime.

If a large number of people use such a script, for example 90% of the miners, the difficulty will quickly drop to what it would be if only 10% of the network were mining.  This would maximize return per watt for people mining SC2, yielding bitcoin when they aren't mining SC2.

The difficulty won't drop, because miners with less than 1M SC can't mine trusted blocks. So it hasn't any effect on the actual blocks/time. You can decide to not mine SC, and the difficulty will drop a little bit.

He's not talking about mining every other block, hes suggesting mining every other difficulty adjustment period.  So you mine the user blocks for one 360 block period then you go away for the next 360 block period.  If over 14% of the mining power does this you will come out ahead (on a per cpu cycle spent mining solidcoin) of where you would be if you mined solidcoin full time.

[dree12]

SC2 has a special difficulty algorithm that allows for vast decreases in difficulty per 360 blocks (~200%) but only small increases (~14%).  It seems that this could be exploited by using a script that does the following:
1.) Mines SC2 when ((number of blocks / 360) % 2 == 0), that is, every other difficulty difficulty adjustment period.
2.) Mines BTC when ((number of blocks / 360) % 2 != 0).  This would allow you to use your cards for BTC in the meantime.

If a large number of people use such a script, for example 90% of the miners, the difficulty will quickly drop to what it would be if only 10% of the network were mining.  This would maximize return per watt for people mining SC2, yielding bitcoin when they aren't mining SC2.

The difficulty won't drop, because miners with less than 1M SC can't mine trusted blocks. So it hasn't any effect on the actual blocks/time. You can decide to not mine SC, and the difficulty will drop a little bit.

He's not talking about mining every other block, hes suggesting mining every other difficulty adjustment period.  So you mine the user blocks for one 360 block period then you go away for the next 360 block period.  If over 14% of the mining power does this you will come out ahead (on a per cpu cycle spent mining solidcoin) of where you would be if you mined solidcoin full time.

Remember that this 14% won't be coordinated, so you won't come ahead at all... there needs to be 14% more miners doing one side than the other. This is an useless exploit IMO - even worse than powerblock-only mining.

[tacotime]

Remember that this 14% won't be coordinated, so you won't come ahead at all... there needs to be 14% more miners doing one side than the other. This is an useless exploit IMO - even worse than powerblock-only mining.

Well... if everyone does it pretty much everyone wins in terms of getting more for less.  Alternatively if you represent 50% of the network hash rate, it allows you to mine more than double the amount of coin.

With the difficultly algorithm the way it is (as was explained on BTC-e earlier), for every amount you decrease it the increased return in coin is exponential.  All it takes is one of the large scale miners currently doing tens or hundreds of gh/s of BTC to use the script as described and the whole SC2 network will be hijacked.  I was writing the script earlier today but got snagged in some python bugginess; it shouldn't be too hard to implement though.  Then all people need to do is mine together using the same script.  To use the script makes more sense in terms of profit/watt, so it seems obvious that people would coordinate with it.  As soon as someone does to a large extent it's likely others will follow.

[dree12]

Remember that this 14% won't be coordinated, so you won't come ahead at all... there needs to be 14% more miners doing one side than the other. This is an useless exploit IMO - even worse than powerblock-only mining.

Well... if everyone does it pretty much everyone wins in terms of getting more for less.  Alternatively if you represent 50% of the network hash rate, it allows you to mine more than double the amount of coin.

With the difficultly algorithm the way it is (as was explained on BTC-e earlier), for every amount you decrease it the increased return in coin is exponential.  All it takes is one of the large scale miners currently doing tens or hundreds of gh/s of BTC to use the script as described and the whole SC2 network will be hijacked.  I was writing the script earlier today but got snagged in some python bugginess; it shouldn't be too hard to implement though.  Then all people need to do is mine together using the same script.  To use the script makes more sense in terms of profit/watt, so it seems obvious that people would coordinate with it.  As soon as someone does to a large extent it's likely others will follow.

How do you prevent half the people mining odd difficulty adjustments, and half mining even ones? People are stubborn. This won't ever work.

Why don't people coodinate to elect one party this election and the other the next? This will evidently be good for the people. Same thing with pressuring companies to improve: if 90% of people switched away from Windows, the next version of Windows is guarenteed to be awesome.

[dree12]

I'm willing to bet the next emergency release will have Coinhunter with the ability to "Manually" adjust the difficulty.

Believe it or not, I'm willing to counter-bet. How about these terms:

An "emergency release" is defined as a release that meets ALL these conditions:

• Coinhunter indicates that all clients must upgrade to the release in a forum post somewhere
• An alert is posted on solidcoin.info if it still exists. In the event solidcoin.info no longer exists, whatever website is promoted as the Solidcoin website by Coinhunter will replace solidcoin.info. In the event Coinhunter does not promote any website for solidcoin, no "emergency release" can exist.
• The release occurs before 2011-01-01 00:00:00 UTC

If the first "emergency release" after 2011-11-20 00:00:00 UTC adds explicitly in the source code code which allows CoinHunter or other control nodes to manually adjust difficulty, assuming difficulty still exists, you have won the bet. Otherwise, I have won. If SolidCoin is changed drastically and difficulty or a derivative of its concept no longer exists, the bet is void.

[BeeCee1]

SC2 has a special difficulty algorithm that allows for vast decreases in difficulty per 360 blocks (~200%) but only small increases (~14%).  It seems that this could be exploited by using a script that does the following:
1.) Mines SC2 when ((number of blocks / 360) % 2 == 0), that is, every other difficulty difficulty adjustment period.
2.) Mines BTC when ((number of blocks / 360) % 2 != 0).  This would allow you to use your cards for BTC in the meantime.

If you simply mine every other difficulty adjustment period you won't be taking advantage of the asymmetric difficulty adjustment.  You'd just be sending a big pulse of mining power for one period when difficulty is low then going away the next one when it is high.  You can do this on any blockchain out there.  This happened to NameCoin earlier this summer before merged mining, lots of people left which caused a huge difficulty drop then everyone rushed back in and completed two weeks worth of blocks in less than two days.

Now that there are lots of block chains out there someone is sure to come up with a script (or a pool) that switches to the most profitable one at any given time, be it bitcoin, litecoin, or solid coin.  They may stay for one (power)block or they may stay for weeks.  Lots of people will use scripts/pools with similar rules so you'll see large pulses of mining power jump among the systems. Pulsed mining will soon be taken for granted.

Where solid coin's asymmetric difficulty adjustment comes in is when 1) you have over 14% of the mining power, and 2) even after an adjustment it is still more profitable than other chains.   Number 2 is more likely with solidcoin than other blockchains since the difficulty doesn't adjust up as fast.  This happened right when it launched when there was a huge influx of miners and difficulty didn't adjust up very fast.

[elggawf]

Isn't SC2 scrypt, so you're not mining it with GPUs anyway?

Still though, you could use the attack to shut off your EC2 nodes every so often and save some cash.

Solidcoin development: "Like watching a bunch of retards trying to fuck a doorknob." It's all the retard of Bitcoin, but without the fact that the underlying protocol is quite brilliant.

[terrytibbs]

Isn't SC2 scrypt, so you're not mining it with GPUs anyway?

No. SolidCoin uses a custom algorithm invented by RealSolid to "equalize performance between GPUs and CPUs".

[tacotime]

If you simply mine every other difficulty adjustment period you won't be taking advantage of the asymmetric difficulty adjustment.  You'd just be sending a big pulse of mining power for one period when difficulty is low then going away the next one when it is high.  You can do this on any blockchain out there.  This happened to NameCoin earlier this summer before merged mining, lots of people left which caused a huge difficulty drop then everyone rushed back in and completed two weeks worth of blocks in less than two days.

You will if you have a large percent of the network, e.g. 50%.  You will effectively mine at the same difficulty as if the network was only at half of the hash power.  The algorithm SC2 uses for difficulty also increases block solvation exponentially with linear decreases in difficulty, so your return is much higher switching between difficulty periods.

[2112]

• The release occurs before 2011-01-01 00:00:00 UTC

This is what I would call a perfectly rigged bet. 

[tacotime]

that was interesting -- network hash rate just increased 3 fold with last difficulty change. Is someone implementing this now?

[dree12]

that was interesting -- network hash rate just increased 3 fold with last difficulty change. Is someone implementing this now?

I doubt it. Solidcoin prices have also increased 3-fold, so that may have been a factor.

[tacotime]

that was interesting -- network hash rate just increased 3 fold with last difficulty change. Is someone implementing this now?

I doubt it. Solidcoin prices have also increased 3-fold, so that may have been a factor.

It went back down in the morning...  It was 0.14 last night on allchains and then this morning 0.04.  Very strange.

[dree12]

that was interesting -- network hash rate just increased 3 fold with last difficulty change. Is someone implementing this now?

I doubt it. Solidcoin prices have also increased 3-fold, so that may have been a factor.

It went back down in the morning...  It was 0.14 last night on allchains and then this morning 0.04.  Very strange.

My data says it's 0.18 right now...

[k9quaint]

Measuring an increase in hashrate on SC2 is pointless since it most likely reflects how many new people have access to a good GPU miner.