Topic: Balanced cryptocoin (Read 623 times)
Exactly that, that will naturally rage limit. so long as at least one transaction goes through in the period before the precompute gets to time, the block will be invalid. I'd like to avoid nxt's pos method if possible. I don't like how the founders get all the coin, it's basically premining. I'm going to start playing with code soon to see what I can do.
There needs to be a method in the code to limit the hash. The system MUST be open source, and must be usable on current hardware. The less power for the hash the better. Ideally, CPU mining would be awesome simply because of the easy availability of potential CPU miners. If I can make a chain that requires virtually no power, adoption should be incredibly feasible.
The idea I had was to eliminate the nonce from the hash, to some extent.
If I remove the nonce and replace it with unix time, I can limit miners to one hash per second, per bitcoin address or block modification.
Difficulty can be scaled to make this fair
You can pre-mine seconds, but the block will only be accepted in the appropriate time frame; probably x-many seconds after the block time. Pre-generating a block will create a block that is only valid for the transactions within.
As such, blocks with the most recent transactions will be the ones accepted by the network over pre-mined blocks.
There are ways to beat the system, but I'm thinking about ways to get around that.
The idea I had was to eliminate the nonce from the hash, to some extent.
If I remove the nonce and replace it with unix time, I can limit miners to one hash per second, per bitcoin address or block modification.
Difficulty can be scaled to make this fair
You can pre-mine seconds, but the block will only be accepted in the appropriate time frame; probably x-many seconds after the block time. Pre-generating a block will create a block that is only valid for the transactions within.
As such, blocks with the most recent transactions will be the ones accepted by the network over pre-mined blocks.
There are ways to beat the system, but I'm thinking about ways to get around that.
Include the nonce to prove that the hashing was done after a certain time period. Still leaves the question of how exactly to rate limit. I was just posting to correct the train carrying your thought.
Apparently NXT is an implementation of the similar sort to what I was suggesting.
There needs to be a method in the code to limit the hash. The system MUST be open source, and must be usable on current hardware. The less power for the hash the better. Ideally, CPU mining would be awesome simply because of the easy availability of potential CPU miners. If I can make a chain that requires virtually no power, adoption should be incredibly feasible.
The idea I had was to eliminate the nonce from the hash, to some extent.
If I remove the nonce and replace it with unix time, I can limit miners to one hash per second, per bitcoin address or block modification.
Difficulty can be scaled to make this fair
You can pre-mine seconds, but the block will only be accepted in the appropriate time frame; probably x-many seconds after the block time. Pre-generating a block will create a block that is only valid for the transactions within.
As such, blocks with the most recent transactions will be the ones accepted by the network over pre-mined blocks.
There are ways to beat the system, but I'm thinking about ways to get around that.
The idea I had was to eliminate the nonce from the hash, to some extent.
If I remove the nonce and replace it with unix time, I can limit miners to one hash per second, per bitcoin address or block modification.
Difficulty can be scaled to make this fair
You can pre-mine seconds, but the block will only be accepted in the appropriate time frame; probably x-many seconds after the block time. Pre-generating a block will create a block that is only valid for the transactions within.
As such, blocks with the most recent transactions will be the ones accepted by the network over pre-mined blocks.
There are ways to beat the system, but I'm thinking about ways to get around that.
Centralization of mining servers is indeed a thread to the stability, openness & security of the currency.
As for your proposal, it is possible to limit hash rate in the software, but as a open source, there is no way to stop anyone from modifying the code to disable the limitation or create miner program that do not have any limit on hash rate.
What may be possible is that to place a cool down period for mining reward to the same address. That will, at least, lower the incentive to build big mining farms.
Regards.
As for your proposal, it is possible to limit hash rate in the software, but as a open source, there is no way to stop anyone from modifying the code to disable the limitation or create miner program that do not have any limit on hash rate.
What may be possible is that to place a cool down period for mining reward to the same address. That will, at least, lower the incentive to build big mining farms.
Regards.
Interesting thought shared by a few forum members.
The problem:
Cryptocoin mining is nearly unfeasible for those with a low amount of buy in, or anyone who has less than 20 GHs. This is only going to get worse as more asicminers enter the field and companies like (hypothetically) BFL bring on large farms of ridiculous hashrates.
Why is this a problem?
The network is becoming more centralized all the time. As server farms become major forces in the mining community, we lose security. The network is most secure when everyone is solo mining at the same hash.
A potential solution:
A hash function that is somehow rate limited. Some sort of function that runs at the same rate no matter your hardware.
I have no idea how to implement this. It seems nearly impossible, and may very well be for both mathematical and practical reasons. I'd like to give it a try, though.
Anyone have ideas or comments?
The problem:
Cryptocoin mining is nearly unfeasible for those with a low amount of buy in, or anyone who has less than 20 GHs. This is only going to get worse as more asicminers enter the field and companies like (hypothetically) BFL bring on large farms of ridiculous hashrates.
Why is this a problem?
The network is becoming more centralized all the time. As server farms become major forces in the mining community, we lose security. The network is most secure when everyone is solo mining at the same hash.
A potential solution:
A hash function that is somehow rate limited. Some sort of function that runs at the same rate no matter your hardware.
I have no idea how to implement this. It seems nearly impossible, and may very well be for both mathematical and practical reasons. I'd like to give it a try, though.
Anyone have ideas or comments?
Jump to: