We recently published three articles describing staking on the Fetch.AI ledger. These outlined:
- The key features of our staking program and the role of validators
- The factors that influenced the design of our staking model
- How our innovative consensus protocol will transform the user experience
The articles contain a wealth of information and answer many of the questions you may have on the topic. Below is an additional list of FAQs. It is a live document and will be updated as we continue to countdown towards the launch of our main network later this year. If you would like to see a question answered in this article, let us know in our Telegram channel or tweet us @Fetch_ai.
What is staking?
Staking is similar to receiving interest payments on a bank account. The main difference is that interest is paid in FET tokens in exchange for agreeing to lock your stake for a certain amount of time.
What is a validator node?
Validator nodes are computers that run copies of the ledger. The operators of validator nodes lock their stake to gain entry to the Proof-of-Stake consensus. This then gives them the right to earn rewards if they conform to the protocol. This requires them to be online and submit valid blocks during the staking period.
What are the conditions for becoming a validator?
The minimal stake to take part in the competition is 200,000 FET tokens. Stakes will be locked for a duration of approximately one month. In total, 200 validator slots will be made available via the process of an auction. The highest 200 bids will be awarded staking slots for the price offered by the lowest winning bidder, any difference in bid price is immediately returned.
Will it be profitable to run a node?
We designed the consensus to ensure profitability for validators even under quite pessimistic assumptions. One of the benefits of PoS-uD is that the profitability can also be tuned by changing the number of slots for validators in the auction, the amount that is allocated to block rewards or the duration of the staking period. One of the reasons for running the auctions prior to mainnet launch is that we gain an understanding of the auction process so that the incentives we provide for validators are correct.
What are the returns of staking Fetch.AI tokens?
Approximately 7,500 FET tokens per month will be awarded to each of the winning 200 bidders participating in the staking auction.
What will the block rewards be?
Currently, we plan for the yearly FET reward to increase linearly with unlocked token supply, so it will increase over time, as described in this article.
When the mainnet launches later this year, monthly rewards will be approximately 7,500 FET tokens per month. By mid 2020 the monthly rewards will have risen to around 10,000 FET per node and will total around 12,000 FET by the end of 2020.
The per-block reward will be adjusted so that
block_reward = monthly_reward x 200 / blocks_per_month.
The reason for this design is that it scales security with token supply and is also less susceptible to token centralisation. There is some discussion of this topic in this publication. The validators will also earn transaction fees.
What happens to the FET tokens that are bid in the auction?
If you are bidding to be a validator and you win the auction, your stake gets locked for one month and you cannot access your funds during this time. You also have the right to act as a validator and earn rewards for doing that. After the time is completed you get your funds back, plus any rewards you have earned.
If you initially bid more than the sale price during the auction, you will receive the difference between your bid and the sale price returned to you. If you do not get into the top 200 validators, your funds remain unlocked and you carry on as before.
I don’t have the 200,000 FET necessary to enter the auction. Can I still stake my tokens?
In addition to the auction to become a validator node, you will also be able to delegate your stake for smaller amounts. Further details on how to delegate your stake will be announced in due course.
When can we expect the number of validator nodes to increase beyond 200?
We have not specified when this will happen but certainly not before the launch of our mainnet in Q4 2019. Our roadmap lays out a development plan which will enable us to increase the number of nodes that can participate in our consensus protocol.
If there are less then 200 nodes, or if some nodes do not collect the rewards at the end of each month cycle (due to bad actions), what will happen to those ‘lost’ rewards?
Most likely these rewards will be reallocated towards the pool of staking rewards. 15% of the total supply is currently allocated for this purpose.
What other actions would cause rewards to be denied to node operators?
Any type of malicious activity such as publishing multiple different blocks at the same height are likely to result in penalties for the nodes. We’re currently modelling what form and how strong these penalties need to be.
What are the hardware requirements to run a node?
The ledger is scalable so hardware requirements are likely to depend on the transaction load. We anticipate it being relatively low in the early operation of the ledger but that it will increase over time. We’ll publish minimal hardware requirements and information about cloud deployments in the near future.
Will running a node require active scaling of hardware to accommodate the network load, or will a baseline hardware requirement be sufficient?
The blockchain is a work-in-progress and in the longer term, it will support high scalability at the same time as low hardware requirements for many types of node. This capability will be added in the year after launch of the main network, but at launch all nodes will be master nodes and have greater hardware requirements.
If the network of a node operator fails, does the individual lose their stake?
The stake would not be lost in this situation and penalties have not yet been determined. Currently, downtime only leads to nodes being ineligible for block rewards and transaction fees during the offline period but that might change if we find that stronger incentives for being online improve efficiency.
