Here’s how it works: Say Alice wants to transfer one bitcoin to Bob. First Bob sets up a digital address for Alice to send the money to, along with a key allowing him to access the money once it’s there. It works sort-of like an email account and password, except that Bob sets up a new address and key for every incoming transaction (he doesn’t have to do this, but it’s highly recommended).
Generally speaking, every bitcoin miner has a copy of the entire block chain on her computer. If she shuts her computer down and stops mining for a while, when she starts back up, her machine will send a message to other miners requesting the blocks that were created in her absence. No one person or computer has responsibility for these block chain updates; no miner has special status. The updates, like the authentication of new blocks, are provided by the network of bitcoin miners at large.
That’s all transactions are—people signing bitcoins (or fractions of bitcoins) over to each other. The ledger tracks the coins, but it does not track people, at least not explicitly. Assuming Bob creates a new address and key for each transaction, the ledger won’t be able to reveal who he is, or which addresses are his, or how many bitcoins he has in all. It’s just a record of money moving between anonymous hands.
With bitcoin, on the other hand, the supply is tightly controlled by the underlying algorithm. A small number of new bitcoins trickle out every hour, and will continue to do so at a diminishing rate until a maximum of 21 million has been reached. This makes bitcoin more attractive as an asset – in theory, if demand grows and the supply remains the same, the value will increase.
On paper, the Mid-Columbia Basin really did look like El Dorado for Carlson and the other miners who began to trickle in during the first years of the boom. The region’s five huge hydroelectric dams, all owned by public utility districts, generate nearly six times as much power as the region’s residents and businesses can use. Most of the surplus is exported, at high prices, to markets like Seattle or Los Angeles, which allows the utilities to sell power locally at well below its cost of production. Power is so cheap here that people heat their homes with electricity, despite bitterly cold winters, and farmers have been able to irrigate the semi-arid region into one of the world’s most productive agricultural areas. (The local newspaper proudly claims to be published in “the Apple Capital of the World and the Buckle on the Power Belt of the Great Northwest.”) And, importantly, it had already attracted several power-hungry industries, notably aluminum smelting and, starting in the mid-2000s, data centers for tech giants like Microsoft and Intuit.
Bitcoin’s popularity has undeniably been its number one advantage over the numerous other cryptocurrencies. By gaining a large number of adopters and users, Bitcoin has achieved a network effect that attracts even more users. Users who would otherwise be more apprehensive investing in a relatively unknown and unproven digital currency are reassured by Bitcoin’s performance over time, its growing community, and the fact that people they know are adopting cryptos.
That constraint is what makes the problem more or less difficult. More leading zeroes means fewer possible solutions, and more time required to solve the problem. Every 2,016 blocks (roughly two weeks), that difficulty is reset. If it took miners less than 10 minutes on average to solve those 2,016 blocks, then the difficulty is automatically increased. If it took longer, then the difficulty is decreased.
All mining ASICs, Bitmain’s included, are performing essentially the same computation—the SHA-256 hashing algorithm—even if they go about it a bit differently. The standard algorithm takes 64 steps to complete, but in Bitcoin it is run twice for each block header, meaning a full round requires 128 steps that are heavy on integer addition. “That’s what dominates the whole design,” says Timo Hanke, the chief cryptographer at String Labs, a cryptography-focused incubator in Palo Alto, Calif. “So, if somebody was to optimize it, they have to optimize the adders. That’s where most of the work is.”
Still, even supporters acknowledge that that glorious future is going to use a lot of electricity. It’s true that many of the more alarming claims—for example, that by 2020, bitcoin mining will consume “as much electricity as the entire world does today,” as the environmental website Grist recently suggested—are ridiculous: Even if the current bitcoin load grew a hundredfold, it would still represent less than 2 percent of total global power consumption. (And for comparison, even the high-end estimates of bitcoin’s total current power consumption are still less than 6 percent of the power consumed by the world’s banking sector.) But the fact remains that bitcoin takes an astonishing amount of power. By one estimate, the power now needed to mine a single coin would run the average household for 10 days.
The Bank for International Settlements summarized several criticisms of bitcoin in Chapter V of their 2018 annual report. The criticisms include the lack of stability in bitcoin's price, the high energy consumption, high and variable transactions costs, the poor security and fraud at cryptocurrency exchanges, vulnerability to debasement (from forking), and the influence of miners.[187][188][189]
The utilities’ larger challenge comes from the legitimate commercial operators, whose appetite for megawatts has upended a decades-old model of publicly owned power. The combined output of the basin’s five dams averages around 3,000 megawatts, or enough for the population of Los Angeles. Until fairly recently, perhaps 80 percent of this massive output was exported via contracts that were hugely advantageous for locals. Cryptocurrency mining has been changing all that, to a degree that is only now becoming clear. By the end of 2018, Carlson reckons the basin will have a total of 300 megawatts of mining capacity. But that is nothing compared to what some hope to see in the basin. Over the past 12 months or so, the three public utilities reportedly have received applications and inquiries for future power contracts that, were they all to be approved, could approach 2,000 megawatts—enough to consume two-thirds of the basin’s power output.

But due to the volatility of bitcoin, it’s impossible to predict the annual revenue of a mining farm. On my flight from China back to the United States, the price of bitcoin crashed 25 percent, from $2,400 to $1,800. In no time at all the operation I visited was bringing in $50,000 less per day. Within a week it was back up, and approaching an all-time high.

Though transaction fees are optional, miners can choose which transactions to process and prioritize those that pay higher fees.[67] Miners may choose transactions based on the fee paid relative to their storage size, not the absolute amount of money paid as a fee. These fees are generally measured in satoshis per byte (sat/b). The size of transactions is dependent on the number of inputs used to create the transaction, and the number of outputs.[3]:ch. 8