THE GAME GETS HARDER

Guest Written by Jason Deane

Bitcoin's First Difficulty Adjustment: A Milestone in Decentralized Consensus

Not long after Bitcoin started operating, a significant technical milestone took place—one that would define the very rhythm of the Bitcoin network and remain a core part of its design to this day. That milestone was the very first mining difficulty adjustment, a mechanism envisioned by Bitcoin’s pseudonymous creator, Satoshi Nakamoto, in the original Bitcoin whitepaper. However, it would take almost an entire year of operation before it would take effect in the way we understand it today.

Why This Matters

Miners are responsible for validating transactions, adding new blocks and issuing new Bitcoin according to the very specific schedule set out in the core protocol. This is a deliberately power intensive process, designed to ensure that transactions are genuine and there has been a ‘proof of work’ in any contribution to the network. In short, a contribution can only be made by expending energy to do so, a natural ‘cost’ to participate in the process. In return, miners are rewarded with new Bitcoin issued on each block processed as well as the total cost of all the transaction fees it contains.

The network also runs to a specific timetable, with each block of transactions being processed every ten minutes. In a closed network with a constant amount of hash power, this works perfectly, but it presents a significant problem if the network is open source and public and anyone can join (or leave) it at any time. Bitcoin, of course, is such a network.

So what would happen, for example, if someone had access to bountiful cheap (or even free) energy and could add lots of processing power at relatively little cost? What if someone developed a machine that could send multiple times more hashes than anyone else? 

In these cases, not only would the security of the system be in question, but the network itself would begin to speed up. After all, more power equals more attempts to solve the blocks and logic dictates they would be solved more quickly. Over time, block times would drop from a ten minute average to far less, inflating the available supply and destabilising the system.

On the other hand, what if someone was contributing significant hashrate to the network, changed their mind and switched off their machines? Without that processing power coming in, the chances are almost certain that block speeds would slow down, causing a different set of problems by jamming up transactions.

Something needed to be done to ensure that the network speed remained constant and couldn’t be adversely affected every time something changed in terms of hash power.

Satoshi’s solution was no less than genius and, as Bitcoin completed its first weeks of operation, it went into action for the very first time.

Mining gets harder

Satoshi proposed a simple but effective rule in the whitepaper: every 2,016 blocks—roughly every two weeks in human terms—the network recalculates the mining difficulty based on how long it took to mine the previous 2,016 blocks. 

If those blocks were mined in less than two weeks, the difficulty (literally, how mathematically difficult it is to find the answer to each block based on probability) goes up. If they took longer, the mathematical difficulty goes down. The new difficulty is exactly proportional to that difference in time, ensuring an autonomous self-correcting mechanism that nudges block production back toward the 10-minute average for the next 2016 blocks.

Although this sounds complicated, the formula is actually quite simple and looks like this:

Expected mining time of 2016 blocks / actual mining time of 2016 blocks (in minutes)

The first number will always be 20160 minutes, but the second will depend on how quickly those blocks are solved. So, if a lot of power has been added to the network in that time and the average time has fallen to 9 minutes, you get:

20160 / 18144 = 1.11

That number is then used to set the new difficulty thus:

Existing difficulty level x 1.11

The first attempted difficulty adjustment occurred as planned at blocked 2016 around January 15^th, 2009, but the recalculation came in at below the limit of 1 set by Satoshi and therefore remained at that level. In fact, this remained true for the next difficulty adjustments as well and it took almost a year for an actual ‘real’ change to take place. That finally occurred at block height 32256 on December 30^th, 2009, when enough new miners had joined the network and block times had started to reduce.

That increase in difficulty was miniscule in numerical terms but represented an 18% increase in difficulty level in relative terms, moving from 1.0 (Satoshi’s baseline) to 1.18. In other words, if this adjustment hadn’t occurred, the increase in hash power would have meant that blocks would have started being solved roughly 18% faster than they should have been.

Thereafter, the difficulty continued to adjust every 2016 blocks and still does to this day, sometimes going down slightly but, more often than not, going upwards as network power continues to grow.

Difficulty Today

Today’s difficulty is exponentially higher than it was back in 2009 and, chances are, it will continue to increase forever as more and more mining power comes on line through new operations, hobbyist miners and better equipment.

Yet the network still works exactly as designed - well over sixteen years after it was first put in motion by Satoshi himself – a true testament to the genius of a simple, yet completely robust, solution to a very particular problem.