Bitcoin mining sits at the heart of the network's security model, yet most explanations are either too abstract ("miners solve puzzles") or too technical for anyone who hasn't studied computer science. This guide takes a different approach: we'll explain exactly what miners do, why it matters, what the economics look like in 2026, and whether it makes sense for you to mine Bitcoin yourself.
By the end you'll understand proof of work from the ground up, know which hardware actually turns a profit, understand how mining pools work and pay out, see through the cloud mining marketing, and have a realistic picture of what the April 2024 halving did to miner margins.
What's in this guide
What is proof of work?
Bitcoin transactions are grouped into blocks and added to a shared ledger (the blockchain) approximately every 10 minutes. The question that proof of work answers is: who gets to write the next block, and how do we know they're not cheating?
Bitcoin uses a cryptographic hash function called SHA-256. Feed any input into SHA-256 and you get a fixed-length string of 64 hexadecimal characters. Change a single letter of the input and the output looks completely different — there's no way to predict what the output will be without actually running the calculation.
Miners take a candidate block (containing pending transactions), add a random number called a nonce, and run SHA-256. Their goal is to find a nonce that produces a hash starting with a certain number of leading zeros. This target is set by the protocol, and there is no shortcut — you have to try billions of random nonces until one produces a hash that qualifies.
The first miner to find a valid nonce broadcasts the block to the network. Other nodes verify it instantly (running SHA-256 once is trivial), and the winning miner collects the block reward — newly minted Bitcoin — plus all transaction fees in the block. Then the race to find the next block begins.
Why "proof of work" prevents cheating
To rewrite transaction history, an attacker would need to redo the proof-of-work calculation for the block they want to change and every subsequent block — while the honest network keeps adding new blocks on top. Because this requires an enormous, ongoing energy expenditure, attacking the chain is economically self-defeating: the cost of a "51% attack" on Bitcoin would require billions of dollars in hardware and electricity, and a successful attack would destroy the value of the very asset you're trying to steal.
Difficulty adjustment
Every 2,016 blocks (roughly two weeks), Bitcoin automatically recalibrates the target hash. If blocks were arriving faster than every 10 minutes, the target gets harder (more leading zeros required). If blocks slowed down, it gets easier. This keeps the block time remarkably stable across massive swings in mining activity — it's one of the most elegant mechanisms in computer science.
Mining hardware: from CPUs to ASICs
Bitcoin mining hardware has evolved through four generations. Understanding why each generation superseded the last explains why you can no longer mine Bitcoin profitably with a laptop or gaming PC.
Generation 1: CPUs (2009–2010)
When Satoshi Nakamoto launched Bitcoin in January 2009, mining with a standard desktop CPU was perfectly viable — difficulty was near zero and there were almost no competing miners. Satoshi himself mined the genesis block on a standard PC. This era ended within months as more people joined the network and difficulty climbed.
Generation 2: GPUs (2010–2013)
Graphics cards (GPUs) can perform the same mathematical operations as a CPU but in massively parallel fashion — a single GPU runs hundreds of calculation cores simultaneously. A GPU mine Bitcoin roughly 50–100× faster than a CPU. By 2011, anyone mining with a CPU was wasting electricity relative to GPU farms. The competitive edge of the GPU era was similarly short-lived.
Generation 3: FPGAs (2011–2013)
Field-Programmable Gate Arrays are chips that can be reprogrammed after manufacture. Miners configured FPGAs specifically for SHA-256, achieving better performance-per-watt than GPUs. FPGAs were a bridge technology — they offered a hint of what was coming — but were quickly obsoleted by the next generation.
Generation 4: ASICs (2013–present)
Application-Specific Integrated Circuits are chips designed exclusively for Bitcoin mining. They do nothing else, but at SHA-256 they are extraordinarily efficient — modern ASICs are millions of times more efficient than the CPU mining of 2009. Once ASICs appeared, all earlier hardware became economically worthless for Bitcoin mining overnight.
Today, the ASIC market is dominated by a small number of manufacturers:
| Manufacturer | Popular models | Hash rate range | Power consumption |
|---|---|---|---|
| Bitmain (Antminer) | S21 XP, S21 Pro, S19k Pro | 100–300+ TH/s | 3,500–5,500W |
| MicroBT (Whatsminer) | M60S++, M66S | 150–300+ TH/s | 3,000–5,000W |
| Canaan (Avalon) | A1566, A1346 | 150–185 TH/s | 3,200–4,200W |
Efficiency is the key metric
When evaluating an ASIC, the most important number isn't raw hash rate — it's joules per terahash (J/TH). A machine delivering 300 TH/s while consuming 6,000W (20 J/TH) is less efficient than one delivering 200 TH/s at 2,800W (14 J/TH). Lower J/TH means lower electricity cost per Bitcoin mined. The best 2025–26 ASICs achieve 13–16 J/TH; older S19 machines average 23–34 J/TH.
Mining pools vs solo mining
Given that the global network hash rate sits above 700 EH/s (700 × 1018 hashes per second), a single home miner running a 200 TH/s machine controls approximately 0.00000003% of the network. The expected time between winning a block solo: over 10,000 years. This is why mining pools exist.
How pools work
A mining pool aggregates the hash power of thousands of participants. When any pool member finds a valid block, the reward is divided among all members proportionally to the work they contributed. Instead of a massive payout once per decade, you receive small, regular payments that collectively add up to roughly the same amount (minus pool fees of 0%–2%).
Pool payout methods
| Method | How it works | Variance |
|---|---|---|
| PPS (Pay Per Share) | Pool pays you for every valid share submitted, regardless of whether the pool finds a block. Pool absorbs variance risk. | Very low |
| PPLNS (Pay Per Last N Shares) | Rewards distributed based on shares submitted in the window before each block find. Rewards loyal miners. | Medium |
| FPPS (Full Pay Per Share) | Like PPS but also includes a proportional share of transaction fees, not just block subsidy. | Very low |
| Solo pool | Pool coordinates mining but pays full reward only to the member who finds the block. Retains solo-mining economics. | Very high |
Major mining pools in 2026
Pool market share shifts frequently. The top pools by hash rate typically include Foundry USA (largest US pool, popular with North American miners), AntPool (run by Bitmain), F2Pool, ViaBTC, and MARA Pool (operated by listed miner Marathon Digital). You can join most pools without any approval process — just point your ASIC at the pool's stratum address with your wallet address as the worker ID.
Mining economics & profitability
Bitcoin mining profitability is determined by five variables that interact with each other constantly. Ignoring any of them produces misleading numbers.
The five variables
- Bitcoin price — revenue is denominated in USD but earned in BTC. Higher prices mean more dollars per block mined.
- Network difficulty — as more hash power competes, each individual miner earns a smaller share. Difficulty climbs whenever hash power grows faster than price.
- Electricity cost ($/kWh) — typically the single biggest expense. Industrial miners in Kazakhstan or Texas pay $0.03–$0.05/kWh; residential UK or US customers often pay $0.15–$0.30/kWh.
- Hardware efficiency (J/TH) — determines how much electricity is consumed per unit of work. Newer, more efficient machines produce more BTC per dollar of electricity.
- Hardware capital cost — a new top-tier ASIC costs $3,000–$8,000. This upfront cost must be amortised over the machine's lifespan (typically 3–5 years before it becomes uncompetitive).
A realistic profitability example
Assume you're running a single Antminer S21 XP (270 TH/s, 3,645W) in 2026. At $0.08/kWh electricity, $90,000 BTC price, and current network difficulty:
- Daily revenue: approximately $5–$8 in BTC
- Daily electricity cost: approximately $6–$7
- Daily profit margin: razor-thin to slightly negative
These numbers change constantly. At $0.04/kWh (industrial rate), the same machine generates meaningful positive margins. This is why electricity cost is the determining factor between profitable and unprofitable mining for 99% of operators.
The halving and what it means for miners
Bitcoin's protocol includes a supply-reduction mechanism called the halving: every 210,000 blocks (approximately four years), the block subsidy paid to miners is cut in half. This continues until the final Bitcoin is mined around 2140.
| Halving date | Block height | Block reward | BTC price ~6 months later |
|---|---|---|---|
| November 2012 | 210,000 | 50 → 25 BTC | ~$260 (from ~$12) |
| July 2016 | 420,000 | 25 → 12.5 BTC | ~$650 (from ~$650) |
| May 2020 | 630,000 | 12.5 → 6.25 BTC | ~$17,000 (from ~$9,000) |
| April 2024 | 840,000 | 6.25 → 3.125 BTC | ~$65,000–$95,000+ |
| ~2028 | 1,050,000 | 3.125 → 1.5625 BTC | Unknown |
The April 2024 halving was historic for another reason: the Bitcoin Ordinals craze had pushed on-chain transaction activity to new highs, meaning the first blocks after the halving included over 37 BTC in transaction fees alone — temporarily more than the block subsidy. This pointed to what Bitcoin's long-term security model relies on: fees replacing the subsidy as the primary miner incentive.
What the halving means for your margins
The halving doesn't eliminate revenue — it halves one component of it. If price rises enough to compensate (as it has done historically), profitable miners continue to profit. What the halving does do is accelerate the competitive pressure: less-efficient machines that were marginally profitable before the halving often become unprofitable immediately after, until they're turned off and difficulty adjusts downward.
Home mining vs industrial mining
The economics of home mining and industrial mining are fundamentally different. Here's what that means in practice.
Home mining realities
- Noise: A single ASIC runs at 70–80 dB — roughly as loud as a vacuum cleaner, running 24/7. Most residential setups require a dedicated, sound-insulated space (basement, shed, outbuilding).
- Heat: An ASIC generating 3,500–5,500W of compute also produces 3,500–5,500W of heat. Without adequate ventilation, ambient temperatures climb rapidly, throttling performance and shortening hardware lifespan.
- Electrical infrastructure: Most ASICs require a 240V 20–30A dedicated circuit. Standard home outlets won't work. Factor in electrician costs if you're retrofitting.
- Electricity rates: Residential rates in the UK average £0.25+/kWh; US averages $0.13/kWh but varies widely by state. Home mining at these rates is extremely difficult to justify financially — it's closer to a hobby than a business.
Where home mining can make sense
Home mining is viable when electricity is unusually cheap: rural areas with cheap renewable power, homes with solar and batteries, properties with old industrial electricity contracts, or regions where natural gas is abundant. Some hobbyist miners use their ASIC as a combined space heater and Bitcoin accumulator during winter months — a creative use case where "waste" heat displaces other heating costs.
Industrial mining
Professional mining operations lease or build dedicated facilities near cheap power sources — hydroelectric dams in Paraguay, natural gas flare sites in Texas and North Dakota, geothermal in Iceland, or large-scale solar farms. They negotiate power rates of $0.02–$0.05/kWh and run thousands of ASICs with professional cooling (immersion or direct liquid cooling at scale). Public companies like Marathon Digital (MARA), Riot Platforms, and CleanSpark operate at this scale. For most individuals, the alternative to home mining is simply buying Bitcoin directly — the economics are typically more favourable.
Cloud mining: mostly a scam
Cloud mining services offer to rent you hash power from a remote data centre in exchange for a contract payment. The appeal is obvious: no hardware, no noise, no electricity bills — just passive Bitcoin income. The reality is almost universally disappointing.
- Guaranteed daily returns (e.g., "earn 1% per day")
- No verifiable data centre address or proof of hardware
- Referral programmes with high commissions (multi-level marketing structure)
- Contracts that magically "expire" before they become profitable
- Withdrawal conditions that activate only after you recruit others
Why even legitimate cloud mining underperforms
Even from providers that genuinely operate mining hardware, the economics rarely favour the buyer. Consider: an industrial mining company can sell hash power to you at a profit, or mine with it directly and keep all the Bitcoin. They will only sell contracts when doing so is more profitable than mining themselves — which typically means the contract terms are bad for you. You're paying retail for hash power they can deploy at industrial efficiency.
There are a handful of transparent, long-running cloud mining providers, but even these see contract buyers typically paying more for their Bitcoin than they would by just buying spot. If you want to mine Bitcoin without the hardware, the honest alternative is to buy Bitcoin directly and accept that you're not mining — because financially, you're usually in the same or better position. Read our full Bitcoin scam guide for more patterns to watch for.
Bitcoin mining and energy
No guide to Bitcoin mining is complete without addressing the energy debate. Bitcoin's network consumes roughly 150–200 TWh per year — comparable to a mid-sized country. This is a real figure and worth taking seriously.
The nuanced version of the debate:
- Energy mix matters: Studies from the Bitcoin Mining Council estimate that roughly 50–60% of Bitcoin mining runs on sustainable energy sources (hydro, wind, solar, nuclear, geothermal). Mining is uniquely location-flexible — hardware can be shipped to wherever power is cheap, which is often where renewable capacity is stranded and can't be exported on the grid.
- Comparison to alternatives: The traditional banking system (branches, ATMs, data centres, gold mining, armoured transport) consumes significantly more energy than Bitcoin by most estimates. This doesn't make Bitcoin's consumption acceptable; it provides context for the comparison.
- Stranded and flared gas: Some of the most interesting mining deployments use natural gas that would otherwise be flared (burned wastefully at oil wells). Mining this "waste" gas converts it to electricity and then to Bitcoin — arguably a better outcome than venting methane into the atmosphere.
Reasonable people disagree sharply about whether Bitcoin's energy use is justified by its utility. What's clear is that the network's energy mix is improving over time, and that miners have strong economic incentives to find the cheapest (often cleanest) power available.
Frequently asked questions
Continue your Bitcoin journey
This article is general educational content and does not constitute financial, investment, or legal advice. Bitcoin mining involves significant financial risk including hardware depreciation, electricity costs, and price volatility. Always conduct thorough due diligence before purchasing mining equipment or contracts.