Proposed Bitmark (MARKS) v0.9.7.3 - Development Report
Coin Emission Modulation v0.1 has proved succesful, and we believe it can benefit from more dynamic range authority. We propose that CEM v0.2 have authority somewhere between 75% and 80% of the epoch full reward, (instead of just 50%). The look-back period on which peak hashrate is based will be reduced from 1 year to 3 months. We believe this time frame is sufficient and better suited.
A discussion about the merits of merge-mining arose on the Bitmark GitHub wiki (
merge-mining discussion ), and an interesting white paper was highlighted:
Merged Mining: ¿ Curse or Cure ? Based on this, we want to make a distinction between
blocks mined natively or (
n-blocks), and
blocks merge-mined via the AuxPoW mechanism as children of another parent chain or (
mm-blocks). The motivation is to reduce mining centralization, while securing the benefits of the large hashpower of powerful parent chains. Another goal is to produce the right number of coins for the available market demand and stabilize the coin's price. Merge-mined coins tend to be freebies which are sold off on exchanges quickly, while natively mined coins are usually more prized by the miners.
Merge-mined blocks (
mm-blocks) will receive a lighter reward than natively-mined blocks (
n-blocks). Naturally, both the native and merged variants must each therefore have their own blocktime targets and difficulty. Each variant's hashrate will also be individually tracked and this data used by Coin Emission Modulation individually as well. This will create, effectively, a total of 16 "viritual" algos from the present 8 mPoW algos. Each mPoW algo will have a
native and a
merge-mined variant. As per the original Bitmark (MARKS) spec, a block will continue to be produced in an average inter-block time of 2 minutes
(β=120s) and the original coin emission caps are observed.
The reasoning for the lower subsidy for
mm-blocks is that the low marginal cost to obtain an mm-block should be likewise equitably rewarded. If a mining operation did not want to be subject to the lower mm-block reward, they can always choose to mine Bitmark as the parent chain, and get the full CEM-modulated reward of the native variant by generating n-blocks.
Work on this proposed hard fork is happening on @piratelinux's GitHub repo, dev branch:
Bitmark v0.9.7.3 development
