Re: [ANN] ArabianChain - DubaiCoin (DBIX) - Smart Contracts from Dubai to the World
Block time will become normal since the moment actual net hashrate will be equal to hashrate shown in stats and stay constant. Fluctuations are main problem. I have no idea how to do it (may be total brain surgery))) You should read about auto difficulty adjust and how algo works in general. Big hashes comes when difficulty low, catch fast blocks, rise difficulty up and go out. You stay in and mine this high difficulty blocks and so on.
If developers decrease this delay, then big hashes will mine a lot of coins without any useful for the project. Then value of token dumped to zero. Investors come to developers and ask "where is my f...ing money?" And so on..
If developers decrease this delay, then big hashes will mine a lot of coins without any useful for the project. Then value of token dumped to zero. Investors come to developers and ask "where is my f...ing money?" And so on..
ubiq has solved this problem with new difficulty algorithm:
Quote
https://github.com/ubiq/go-ubiq/releases
This is a hard fork release so you will need to upgrade before block 8000.
The previous hard fork was activated without any issue but the Digishield difficulty algorithm is proving to be inadequate. We have coded up a new solution called the Flux Difficulty Algorithm. This is a minor adjustment to Digishield so that it implements throttling which will eliminate the peaks and troughs of the difficulty chart. This works by making an extra comparison with the previous block time and throttling the rate of change. The current Ethereum difficulty algorithm works using the approach of looking at the time of the previous block.
We have been testing this for a few days in our local Testnet and the results look good. We do not anticipate any issues and we are in close conversation with our pool operators.
Thank you for your support and understanding.
This is a hard fork release so you will need to upgrade before block 8000.
The previous hard fork was activated without any issue but the Digishield difficulty algorithm is proving to be inadequate. We have coded up a new solution called the Flux Difficulty Algorithm. This is a minor adjustment to Digishield so that it implements throttling which will eliminate the peaks and troughs of the difficulty chart. This works by making an extra comparison with the previous block time and throttling the rate of change. The current Ethereum difficulty algorithm works using the approach of looking at the time of the previous block.
We have been testing this for a few days in our local Testnet and the results look good. We do not anticipate any issues and we are in close conversation with our pool operators.
Thank you for your support and understanding.
Could help?
https://github.com/ubiq/go-ubiq/commit/d8bb1eeded866926dce347b268e7103f4e95ebab
Anyway, I have a lot of confidence in the arabianchain dev...good job guys!
I consulted with the Ubiq dev and got some great advice specific to this issue early on, including an offer to handle the Hard Fork. Those guys Rock!!
the difficulty issue is something with addressed with ubiq at the beginning, with the introduction of our flux difficulty algorithm. our target time is 88seconds
so theres a few easy options for difficulty, minor tweaks to eth difficulty (like EXP) howver you will likely see a constant ~30% drift from target time with that approach. Or replace with diff adjustment all together, adding flux from ubiq. with flux, you will still see some odd blocks times, fast, slow, but over time, everything averages out almost perfectly on target time
either option would require a hard fork release
However the choice was made above, as a team to allow the current ArabianChain programmer to handle the Hard Fork himself. I believe it to be a fair decision that was made. I have spent some time chatting with the ArabianChain dev and I believe he is highly qualified and capable of the task which he has completed and which was fully implemented in the Hard Fork which began at Block 39k. I believe that the issue has a really excellent chance of being solved through the Hard Fork. I think that we all need to support the ArabianChain dev and give this the proper amount of time needed in order to resolve itself through the modifications made in the Hard Fork in order for it to reasonably reach the the desired 90 seconds Block Timed emissions or as close to the 90 second average as possible since obviously DBIX is not the only Ethereum Clone to experience this issue.