One of the very much important characteristics of our money is that it’s decentralized. Having nobody at the top deciding who can send or receive money is key, nobody would want to voluntarily store their time and effort in something that can be taken away on a whim. To achieve such decentralization, we shared the work of running our network between nodes and miners. But for that idea to work as we intended, we need to ensure all of them remain decentralized.
Nodes are easier to decentralize because running a node is cheaper, and if you use the money yourself, you’re already incentivized to have your own copy of the record that says who owns what. You want that list to be in as many places as possible, it’s your money after all.
But miners are a bit more tricky. Someone might mine to make money and not to use money. They might be altruistic and all warm and fuzzy inside, but their incentives are to maximize profit. When you want to maximize profit you tend to do two things: get better and get bigger. One of them is solvable, the other is… let’s say combatible.
Remember how miners were basically using computers to try to guess a very difficult number? Well, it happens that if the math they have to do is always the same, you can get better at it. Just like someone that doesn’t know how to play piano if you challenge them to play a particular song. They will start to press keys randomly until they make kinda sense, they’re ordered from bass to treble, left to right. They hum the song and try one key after another, and eventually with a lot of trial and error, they play it. What’s going to happen if you ask them to play a different song next? You can bet they’re faster than the first time. They got better at it. The same happens with mining. If we use the same algorithm to guess the next lucky number, and people are making money out of it, then you can bet people are going to invent better chips to do that. Those newer chips would be optimized for that particular algorithm. The maker of those chips would be a bit faster than anyone else at getting a reward. With that reward they can make more optimized machines, and charge a premium for them. Eventually, the only remaining miners are the profitable ones, and they’re a small bunch of very rich participants. Now you have a minority deciding which transactions are worthy of being written, who gets to use our money, and above all else: what’s the true money.
RandomX was developed by Tevador and many others to attack this problem. It randomly changes the mining algorithm. It’s as if every time our novice piano player managed to unlock a new tune, we took all the keys out and scrambled them. It’d be very, very hard to get better at piano like this. Since RandomX changes all the time, the best chip at it would be a chip that could tackle all those different problems, and that happens to be a CPU. The brain of your computer is a multi-purpose tool. RandomX demands that you use a swiss army knife for everything, it may not be the best tool for any particular need, but if you can carry only one tool for an unknown need, it’s your best bet.
There’s a secondary outcome of this approach that is useful to combat miners getting bigger. You cannot avoid participants that have more money, buy more machines, have bigger deals with electric companies, have dedicated employees, etc to outperform you at this. That’s called economies of scale. But we can minimize its effects with something like RandomX. Since it works best in all-purpose machines, it’s more likely that more people with have them because they need them for other things. So they’d already own the machines, and you’re only throwing a cherry on top by allowing them to also mine our money. The more miners the more our network decentralizes, and therefore the better money we have.
Let’s take this brief moment to bask in the warm afternoon light beams hitting our smiling face and enjoy knowing we designed a pretty damn good money.
The moment flies by, a small cloud covers the sun and we open our eyes. A tiny insidious question fades in coming fast toward us from the back of our mind like a exit sign on a night highway: Why doing all this?