A resurfaced Bitcoin forum post from Sept. 7, 2010 shows just how dramatically the economics of mining have changed.
At the time, one early Bitcoin user calculated that producing a single block on their home computer would cost roughly $5.68 in electricity.
The block subsidy was then 50 BTC, but the miner described the setup as “a net loser.”
Hindsight is 20/20
The user, TTBit, said they had connected a Kill-a-Watt electricity meter to their computer and measured power consumption at roughly 140 watts with the monitors switched off.
Their machine was producing around 2,200 kilohashes per second.
They estimated that it would take approximately 14 days and two hours to successfully mine one Bitcoin block based on the difficulty.
That would consume approximately 47.3 kilowatt-hours of electricity. At an electricity rate of 12 cents per kilowatt-hour, the resulting cost worked out to $5.68.
“It would cost me 47.327 kWh to produce a block; at 12 cents per kWh, that’s $5.68 per block, a net loser,” the user wrote.
Yes, you got this right: the user was actually debating whether it was economical to spend less than $6 in electricity for a chance to receive 50 Bitcoin.
Back in the day, the flagship cryptocurrency had very little economic value.
Mining was still largely performed with ordinary CPUs, and forum users were busy arguing whether it made more sense to generate Bitcoin or simply buy it. GPU mining was not really a thing back then.
In the same thread, TTBit noted reports of users achieving around 25,000 kilohashes per second with graphics cards using CUDA software. The user speculated that GPU mining could make Bitcoin profitable again if electricity consumption remained below roughly 1,000 watts.
Another participant reported getting around 33,000 kilohashes per second from an Nvidia GTX 260 while consuming roughly 200 watts. This was far more efficient compared to CPU mining.
The thread even attracted Bitcoin creator Satoshi Nakamoto. Satoshi estimated that a 24-core AMD system could reach around 66,000 kilohashes per second.
TTBit attempted to model future difficulty growth and predicted a level of roughly 6,672 by the end of 2010. Another user thought even that estimate might be conservative.
“If the trend continues, it will be quite difficult to generate coins in the near future,” TTBit wrote.
That prediction proved directionally correct.
