Idea for Improving Proof of Work

[MicroGuy]

I have an idea for improving proof of work to make it more sustainable.

What if we made a variable sliding threshold (a limiter) that could be reached whereby a miner would no longer have an advantage based on computational power. This would operate like the governor on a combustion engine. A predetermined setpoint could be calculated by an algo based on network conditions.

The sliding scale would adjust in correlation to network hashrate. Once the setpoint is tripped, all miners beyond the threshold would have an equal chance of finding the block.

~~

This is a raw idea that will need more contribution. But perhaps it's enough to qualify as a spark of an idea.
Let's save New Zealand from Goldcoin.

[cryptosolar]

I have an idea for improving proof of work to make it more sustainable.

What if we made a variable sliding threshold (a limiter) that could be reached whereby a miner would no longer have an advantage based on computational power. This would operate like the governor on a combustion engine. A predetermined setpoint could be calculated by an algo based on network conditions.

The sliding scale would adjust in correlation to network hashrate. Once the setpoint is tripped, all miners beyond the threshold would have an equal chance of finding the block.

~~

This is a raw idea that will need more contribution. But perhaps it's enough to qualify as a spark of an idea.
Let's save New Zealand from Goldcoin.

- would be nice to get a rough example for this
- i am not sure but i believe Golden River Algo is solving this problem.

[Stouse49]

This governor concept would be hard to implement and may be easy to undo (since its open source software).

The client can provide the current network hashrate and the miner can measure its own hashrate. If we said that 10% was the max any miner or pool could have in terms of the total hashrate, then mining software would have to limit its speed to 10% of the network total. This would be a special modification to the miner software, but nothing in the GoldCoin protocol requires that a particular mining software is to be used.

If we released an official mining software, then miners and pools could use the ones currently available, like cgminer.  If the Goldcoin client was modified to govern the speed, then it could use the Network Hashrate and calculate the mining hashrate based on how often blocks are solved and then reject blocks that are above the amount that could be solved by 10% of the network in a given period of time. The local client for the miner would have to keep track of the blocks that it solved.

Based on my knowledge, the rest of the network cannot tell which miner solved which blocks. So, a miner could modify the source to turn off this feature, no other peers could tell the difference and determine that a particular miner wasn't playing by the rules (limiting block solving beyond the current rule for the 51% defense). The 51% Defense works because all nodes can verify that every block follows those rules, because they have all the necessary information to do so. However, if a miner is submitting more than his allowed 10% of the blocks, the other nodes cannot verify this because they don't know where the block was first mined, nor what hashrate that miner has.