I am evaluating the feasibility of a large mine, 10PH+. As for any miner, the main drivers of profitability are low cost of hardware, low costs of power/hosting and difficulty on the network/total network hashing power.
Here is an excel model that incorporates the following components (note: all inputs/outputs are on the "Pro Forma" tab, all inputs that can be changed are in yellow cells):
Main conclusions:
- With a realistic increase in difficulty to 376PH on the network by the end of the year (start difficulty at 16% increase per 14 day step falling to 3% at 0.006% speed), it is hard to see how buying hardware in the fall beyond $0.50/GH will be profitable in the long run,
- With a severe increase in difficulty to 600PH (start difficulty at 20% increase per step falling to 3% at 0.006% speed), it will take hardware at $0.25/GH to make a modest ROI.
Assumptions:
- difficulty is modeled in a way consistent with the quasi-linear decrease in growth observed this year, where the pace of increase can be adjusted to reflect the predictions by insiders (bottom of article),
- reward halving at block 420,000 is taken into account,
- constant BTC price to look at the merits of the mine, not at an investment in BTC,
- capital structure: can include debt at an interest rate set by the user,
- cost of hosting/cost of power: an overall cost is assumed all-in (likely hosting in Central Washington), rather than modeling them separately,
- start date: mid-October (can be changed),
- duration of project: until marginal revenues < marginal costs.
What I am looking for: - Any feedback on the modeling,
- Any comment on the assumptions used, additional revenues, hardware sourcing strategies, and project finance funding.
Thanks!
