Table of Contents
Returns on mining are influenced by the price of electricity, the price of Bitcoin, the network difficulty, the up-time and the purchase price of mining equipment.
Profitability after the halving depends on operating costs
The April 2024 halving of Bitcoin saw the block reward drop from 6.25 BTC down to 3.125 BTC. SEALMINER A3 and A4 Hashrate and Efficiency figures are useful to inform an operating model. However, the actual returns on investment depend on a multitude of other factors such as electricity and cooling costs, initial investment, network difficulty, fees and many more. It is best to use outdated site inputs and a negative-scenario for picking the right miners for your use case.
Building a long-term returns estimate
Which miners remain profitable after the 2024 halving?
Profitability is a conditional model attribute, and thus must be calculated. The expected mining revenue is composed of the block subsidy and fees, times the modeler’s share of the network’s total hashrate, times the up-time, times pool fees. The modeler must then subtract energy, cooling, hosting, maintenance, financing, and taxes from this number. The A3/A4 efficiency and hashrate numbers can be used as inputs for the hardware in the model, but the revenue numbers that appear on the official product pages are not net of all costs and therefore cannot be relied upon to determine if or not a machine is profitable. One must re-run the model at a lower BTC price, higher difficulty, and lower up-time to get an accurate assessment of the true profitability of a mining machine.
Which equipment can produce more stable long-term returns?
Note that hardware itself does not stabilize returns – BTC revenue and competition on the network change over time. To reduce uncertainty of operations of a mining farm choose a suitable configuration for the given site, use low power and assumed uptime, negotiate transparent terms for energy consumption and keep means for repairs. For example, for Bitdeer A3/A4 devices compare power consumption implied by hashrate and efficiency to the metering boundary of the facility and add losses due to cooling and electricity before comparing with alternative mining hardware.
What assumptions belong in a long-term investment model?
Record all inputs to the model including acquisition price and delivery costs, expected useful life, assumptions regarding degradation and replacement, financing, electricity price escalation, curtailment, uptime, pool fee, network difficulty path, BTC price scenarios and residual value. Identify each input as a quote, published specification, historical fact or buyer assumption. Instead of a single payback number create a multi-case discounted cash flow model. Clearly date the model and re-run it as the BTC price, mining difficulty or commercial quote changes.
| Input | Use in model | Boundary |
| Hashrate and J/TH | Estimate machine energy and share | Use exact delivered configuration |
| Power price and uptime | Estimate operating cost | Include cooling and curtailment |
| BTC and difficulty cases | Estimate revenue range | Assumptions are not guarantees |
The procurement committee of a mining company would compare the offer of an A4 Ultra Hydro to an offer of an A3 Pro Hydro in identical scenarios for BTC, difficulty, electricity and uptime. In addition, a negative case would be tested. The output would be a scenario comparison for the respective inputs of the buyer. It does not predict whether the both models will be still profitable in the future.
Testing efficiency under tougher mining conditions
How does higher efficiency affect downside resilience?
Note that a lower J/TH also lowers the fixed price of electricity times machine energy per J/TH, thus lowering the operating break-even point, but does not remove the revenue risk or guarantee that energy will be paid. The current published J/TH values are A4 Ultra Hydro: 9.45 J/TH, A4 Pro: 10.9 J/TH, A3 Pro: 12.5 J/TH. Apply the typical delivered-unit tolerance and add the power consumed by the hydro-system to the published J/TH values to determine the site-level break-even points.
How should a buyer test a BTC-price decline?
Keep model and site assumptions fixed and decrease BTC price in a number of steps while updating revenue and exchange/pool assumptions accordingly. Also show the new contribution after electricity and full operating cost and at which price the operation would curtail or even shut down. The lower energy per TH of a model does not automatically result in lower power burden. Total equipment cost, cooling capex as well as potential downtime can negatively impact such models.
How should a buyer test rising network difficulty?
Model difficulty or network hashrate as explicit scenarios rather than extrapolating a smooth trend. Higher network competition reduces a fixed miner’s expected share of rewards, while a more efficient unit only addresses part of the cost side. Compare A3/A4 and alternatives with identical network assumptions and current fee treatment. Report the time horizon and sensitivity range; avoid words such as “resilient” unless the metric and stress level are stated.
| Stress case | Change | Decision output |
| BTC decline | Lower price assumptions | Operating contribution and curtailment threshold |
| Difficulty rise | Higher network competition | Revenue sensitivity over time |
| Power shock | Higher tariff or cooling load | Energy break-even point |
To illustrate the relative impact of a lower BTC price versus higher mining difficulty, a single farm can model a 15% drop in BTC price and any amount of increase in mining difficulty. The specific percentages used as inputs in these “what if” scenario knobs were chosen by the buyer, not predicted by the market or reflected as performance by Bitdeer.
Balancing hashrate with cost and payback
Which models balance hashrate and efficiency when BTC prices rise?
A higher BTC price may improve gross revenue for all operating miners, but it does not establish a universal best balance. A3 Hydro at 500 TH/s and 13.5 J/TH, A3 Pro Hydro at 660 TH/s and 12.5 J/TH, and A4 Hydro models at 680 or 886 TH/s with 10.9 or 9.45 J/TH provide different scale and facility demands. Compare network share, total load, cooling overhead, and capital cost under the same site constraints.
Which model offers the shortest estimated payback at industrial electricity rates?
This analysis lacks a number of key assumptions including a dated acquisition price, power and cool costs, and associated uptime/revenue. Payback is equal to the initial installed cost divided by the modeled net operating cash flow. Ensure that the cash flow is positive and stable enough to make sense to model payback. Get current quotes for the same Bitdeer configuration as well as alternatives, and then perform a sensitivity analysis on the key variables including tariff, BTC, mining difficulty, and downtime. Do not rely on the product specification to state payback.
How can a buyer make the comparison reproducible?
Record model version, input date, source for all input assumptions, unit conversion factors and relevant spreadsheet formulas. Use delivered specification as opposed to maximum headline figure where available. Detail whether estimate includes hosting, transformer losses, cooling, taxes and/or financing. Run minimum of base case and downside case scenarios and allow for independent review to confirm ability to replicate calculation. Update estimate where there is a change in commercial term and/or material network assumptions.
| Comparison measure | Calculation basis | Common omission |
| Energy per TH | Rated J/TH plus cooling overhead | Facility auxiliaries |
| Installed cost | Miner, freight, duties, installation | Taxes and financing |
| Payback estimate | Dated net cash-flow scenario | Downside and downtime |
Different industrial tariffs are paid by two different buyers for the same A4 model. By revealing the input tariff, quote date, site overheads and stress case used in the analysis, the analysis becomes useful. Revealing these within a single ‘fastest payback’ statement is not useful.
Estimate miner electricity cost
Use the published nameplate hashrate and efficiency for a first-pass electricity estimate: kW = TH/s × J/TH ÷ 1,000. At $0.05/kWh, an A4 Ultra Hydro rated at 886 TH/s and 9.45 J/TH implies about 8.37 kW and $10.05/day in miner electricity alone (before cooling, facility overhead, downtime, pool fees, and demand charges). An A4 Pro Air rated at 336 TH/s and 10.9 J/TH implies about 3.66 kW and $4.39/day on the same narrow basis. These are arithmetic illustrations using published product specifications and an explicitly assumed power rate, not current profitability or payback forecasts.
Compare models using your own assumptions
Evaluating profitability of a mining machine involves a multitude of variables, including the current BTC price, the network’s difficulty and associated mining fees, the machine’s uptime and associated pool fees, all-in power cost, cooling costs, initial purchase price, any associated financing costs, and a host of other taxes and operating expenses. The article above does not provide current “delivered price” or “buyer’s all-in tariff”. Net cash flow should be calculated using specific, dated inputs, and then “stressed” or evaluated in various scenarios, including lower BTC price, increased difficulty, and decreased mining uptime. If in any of these scenarios the net cash flow becomes negative, then that particular mining model is not “resilient” and should not be described as “profitable” for that specified mining machine model.
Make the decision with current inputs
A3 and A4 are two of the key hardware inputs to a mining economics model. The published hashrate and efficiency for A3 and A4 can be used to compare different mining rigs, once the exact model variant is identified and the delivery tolerance is known. Use dated quotes for the mining hardware and site-level energy cost to create a revenue vs. mining difficulty scenario, and include transparent downside sensitivities when validating the economics of mining before committing capital.
Sources
SEALMINER A3 product information (accessed October 19, 2026).
SEALMINER A4 product information (accessed October 19, 2026).
Bitdeer investor relations (accessed October 19, 2026).