Pearl Mining Kit · guide
What actually decides your income, which settings keep the card alive, and when mining is not worth it.
Your GPU computes hashes. The network adjusts difficulty so blocks arrive at a steady rate; when a block is found, the pool splits the reward between miners in proportion to submitted shares. Hashrate is your share. Everything else is a detail.
Income = block reward × your share − pool fee − electricity. Price and difficulty move constantly, so any daily number is an estimate.
| Setting | Value |
|---|---|
| Power limit | 190 W (~76% of stock) |
| Core / memory offset | 0 / 0 |
| Fan curve | 60% below 70 °C, 80% above |
| Result | ~124 TH/s, 74 °C, ~80% fan |
| Thermal guard | throttle 78 °C, stop 85 °C |
Cutting power is not linear. Losing 25% of watts often costs only 5–10% of hashrate, so hashrate per watt — the number that pays the bill — goes up.
A hot card throttles, gets loud and wears faster. Two temperatures matter: where you throttle and where you stop. Make both automatic, because you will not notice a problem at 2 a.m.
PROP pays in proportion to shares in the current round — simple and common. PPLNS pays over the last N shares and rewards miners who stay. PPS pays per share with steady income and a higher fee.
A 0% pool with expensive withdrawals is not free. Compare net payout per TH/s over a full week, after fees, not one lucky day.
The seed phrase is the wallet. Whoever has those 12–24 words owns the coins.
Watts × hours ÷ 1000 = kWh. Multiply by your price. A 200 W rig runs about 144 kWh a month. Count the whole system and measure at the wall, not in software.
Then: daily income − daily cost. At $0.06/kWh a 5070 at 190 W nets roughly $2/day in my experience — and that number moves with price and difficulty. Expensive power, an old inefficient card, or no time to maintain it can make the honest answer “no”.
It is the boring part done for you: installer, miner, wallet, safe presets, fan curve, thermal guard, signed auto-updates, and a phone dashboard with alerts.