3

I've seen and heard about issues where a coin's network would get 'stuck' at a specific block. I understand why it would get stuck due to high difficulty and low hash, but some coins seem to be unable to mine past a block with low diff and more than enough hash power on the network.

The two examples I can think of are Thorcoin currently stuck at block #85745 and more recently Suncoin at block #180561.

What exactly causes this and how does it get fixed?

  • Interesting question. I would ask two things: 1. If a difficulty adjustment has just occurred, the target difficulty for the next block might be much greater than the previous; are you sure this is not the case? 2. How do you know how much hash power is on the network? Maybe a high-powered miner left the network (for another more lucrative coin). – Nate Eldredge May 12 '14 at 6:15
  • The Suncoin case is the best example I can provide. It took more than 12 hours to generate block 180562: explorer.suncoin.biz/block/… Yet the difficulty was supposedly very low at the time of mining (0.816). However now that I look at the block, I see a difficulty of 262. I think now I am confused even more :) – Tuaris May 12 '14 at 15:12
1

A few elements of answer:

  • Difficulty readjustment can be done for each block on some altcoins.
  • Mining is a highly random process, if you keep on flipping a coin all day, you will get long series of faces. Similarly, the miners may get unlucky. Even in bitcoin you see gaps of over an hour.
  • The above is compounded for altcoins as miners switch from one to another depending on profitability.

So nothing really unusual. As for a fix, bitcoins' testnet uses a rule where difficulty drops to 1 when no block has been found for 20 min. But the security implications should be thought through.

  • Your second point needs to be interpreted with some care. Yes, mining is random and it is possible that bad luck causes a block to take much longer than normal. But we can quantify this possibility. Finding blocks is well described by the Poisson process and times between blocks are exponentially distributed. So if the average time to find a block (given the amount of mining power on the network) is s, the probability of a block taking at least time t is e^(-t/s). – Nate Eldredge Jun 11 '14 at 21:02
  • For Bitcoin, with a target time of 10 minutes, the probability of a block taking at least an hour is e^(-6) ~= 0.002 or about 1 in 403. We should expect to see this happen every couple of days. The Suncoin example is in a different league: with a target time of 1 minute, the probability of taking at least 12 hours is e^(-720) or about 10^(-313); we should expect to see that happen about once in the lifetime of a googol googol googol universes. So that's extremely strong evidence that this phenomenon is not due simply to chance or bad luck. – Nate Eldredge Jun 11 '14 at 21:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.