Wednesday, August 30, 2017

Gridcoin Mandatory v3.6.0.2 - V8 Staking Time Estimate Update


To recap, the V8 stake engine is set to come live in just over 2000 blocks, and will remove mag based stake weight, thus preventing researchers from securing the network. This job will now be solely on the shoulders of investors in a classic POS system. Such a change was required to patch several security vulnerabilities uncovered by the dev team. An unfortunate side effect is that it will now be much harder for miners to get paid for their research by staking a block if they are 'mining' solo (instead of in a pool). The question over the last few weeks has been how much harder.

So, How Much Harder is 'Harder'?

In the last article I used the network stake weight to estimate how often anyone will stake a block on the Gridcoin blockchain, based on their balance. When comparing the numbers with @denravonska's (Ravon's) empirical data, we found a significant gap between the two predictions. Ravon contacted the Blackcoin dev team asking about the network weight, as Gridcoin is based on the Blackcoin architecture, who replied with:
There is no way to observe directly how many coins are staking. The network weight is just an estimator for the unobservable total staking funds. It is calculated from the difficulty. The network weight you see in the GUI is a function of a running average of the difficulty.
What this means is that the number that we have been assuming to be the total number of coins staking, is in fact not that at all. The number of coins staking will be lower. The implications are twofold:
  • We have been vastly overestimating how much of the total GRC in existence is actually staking and securing the network.
  • My estimates provided were overestimates of the time/balance required to stake. We have no way of knowing by how much.
That being said, the upper bound figures provided still hold true. If we assume all coins are actively staking (390 million GRC), then the required balances to stake at frequency X is:
  • 432,000 GRC (USD$17280 of GRC) to stake daily
  • 63,000 GRC (USD$2520 of GRC) to stake weekly
  • 15,000 GRC (USD$600 of GRC) to stake monthly
  • 2400 GRC (USD$96 of GRC) to stake every 6 months
The fact that we have been overestimating the number of coins staking is another good reason to only allow investors to earn interest when their wallets are online. Interest is paid in return for securing the network, and clearly most investors are not securing the network - instead only popping online for a few blocks every blue moon to collect their owed GRC. The side effect of this is that such a change will bring the reality of staking times far closer to the above estimate.

Some Final Notes Before V8

  • Make sure that you upgrade your client to the latest version, which is currently v3.6.0.2. If you do not, you will end up on the abandoned fork of the Gridcoin blockchain when it reaches block 1010001.
  • V8 will drop your stake rate if you are primarily a miner. If you are not staking often enough for your liking, I recommend joining a pool. There is currently only one pool, which charges zero fees aside from a withdrawal fee.
  • While I am critical of the changes affecting miners in V8, I want to reiterate that the change as a whole is good and necessary for the Gridcoin network. There are already ideas being fielded to benefit the miners, such as integrating research payment into superblocks.
Remember that this is the environment all successful crypto investors passed through. If you get in early on a coin, you deal with hiccups such as this, and if you get lucky it pays off. For all intents and purposes, it is still very early days for Gridcoin. Lets see where this journey takes us, and I'm looking forward to the many more valuable contributions we make to science.

Full article here.