by Gene Michael Stover
created Sunday, 2026 August 2
updated Thursday, 2026 August 6
original at cybertiggyr.com/mining_vs_staking.html
(I wrote this to answer a friend's question.)
In proof-of-work, the right to propose the next block is won by performing an enormous amount of computation. It's a competition between nodes that are trying to win the right to propose the next block.
>In proof-of-stake, the protocol selects a validator from participants who have placed cryptocurrency at risk as collateral. In the long run, each of the participants will get their turn, though from block to block, it's more like a lottery.
A couple key features of cryptocurrencies are that they are (a) decentralized, peer-to-peer systems, & (b) the peers don't trust each other.
By the way, this description is all at the hand-wavy, conceptually correct (I hope) level of detail. A true cryptocurrency geek could surely find inaccuracies in it; I'm try to describe the big picture.
I'll use peer and worker interchangeably from now on in this file.
Bitcoin, the original cryptocurrency, dealt with the lack of trust by creating a race between the peers. The first worker to produce a solution would broadcast it to the other nodes for them to check. Each of those nodes checks the solution. If they accept it, it (probably) becomes part of the block chain. Otherwise, they reject it & the search continues. Such a system is called proof-of-work & the nodes that search for new blocks & then broadcast them are called miners.
An alternative to mining is proof of stake. With proof-of-stake, each peer is associated with collateral. The network of peers take turns packaging transactions into blocks. As with mining, they all check a proposed solution. If it's acceptable, the group decides whose turn is next, & that peer begins working on the solution to the next problem. In this system, the workers as called validators (at least they are in Ethereum).
For this, you'll need to know something about Cryptographic hash functions, but here's a cheat sheet of important details...
For proof-of-work cryptocurrencies, this “hard problem” everyone mentions is: Given the data for a new block in our blockchain, with some spaces where you can adjust values, adjust those values until the hash value of the whole block is less than Limit.
Example to make it concrete...
Let's say that I have a hash function that gives an 8-bit result (which would be useless other than for explanations like this one). Its inputs are a 2-digit number. So if I give a number to our hash function, it gives me a number on the interval 0..255 (because that's what you can represent with 8 bits, unsigned, starting with 0).
Let's say that you tell me part of the input is 5. I can find another digit, D. You want me to find D such that hash(5 + D) < 128.
Notice that the “less than 128” restriction is equivalent to saying that the high bit of the hash value is zero. Let's say that I pick 8 for D. I apply our hash function. There's a 50:50 chance that the high bit is 0. If it is, I've succeeded & I report to you that D = 8. Otherwise, I try another value for D. Again, there's a 50:50 chance that the high bit is 0. If it is, I report that value to you. Otherwise, I try again. I keep doing that until I find a value of D for which the hash value is less than 128. You can check my work to see that I'm honest.
You could make it more difficult by telling me to find a value of D so that the hash value is less than, say, 64; now it'll take me twice as long (on average) to find a suitable value of D. Or you could specify that it's less than 32. And on, & on.
Okay, big deal. We have this 8-bit hash function & a 2-digit input. Big whoop.
But in practice, the hash functions give 256-bit values & the input is mebibytes long with multiple bytes (not just a single digit) for me to adjust. If the restriction you place on the output is still that the high bit is zero, I still have a 50:50 chance of finding a suitable value with each attempt, which is still easy. But if you specified that the top 30 bits must be zero, now I have about 1:billion chance of finding a suitable value each time I adjust the value.
Glancing at a Bitcoin blockchain explorer just now, it looks like the top roughly 40 bits must be zero. 2^40 is (very roughly) 1 trillion. So each attempt at a solution has a 1:trillion chance of succeeding.
As more peers join the network, as custom hardware increases the computation rate of each peer, the network raises the difficulty by specifying that more & more of the high bits must be zero. And they are all racing to find the next solution first. That's why proof-of-work consumes enough electricity that people are concerned. Also, since finding a solution is unpredictable, sometimes it shows up faster than expected, sometimes longer.
(I think the Bitcoin protocol actually specifies a target value such that the hash value must be less than that target. I discussed high bits above, not target values, because I feel it's easier to understand how the difficulty changes as the number of high bits increases... which is the same as the target value decreasing.)
The fiddly bits of the actual peer-to-peer protocol figure that one out. (I'm declaring this legitimate technicality below the Hand-Wavy Conceptual Level Of Detail.)
Because the workers are not racing towards a solution in a proof-of-stake system, they needn't use as much electricity as miners do in a proof-of-work system. In theory, they needn't be specialized hardware. However, where the proof-of-work network effectively excludes participants that haven't purchased enough custom hardware, the peer-to-peer network of a proof-of-stake system excludes participants that haven't staked enough collateral.
And here's some background knowledge that could help if my explanation made too little sense...
Cryptographic hash functions & hash functions functions in general are fundamental to cryptocurrencies.
Though they are currently mostly treated as a “store of value”, cryptocurrencies were intended as implementations of electronic cash. Before Bitcoin, there were other implementations of electronic cash. Read about the work of David Chaum if you are interested in that.
Note that digital cash differs from electronic funds transfers & credit card transactions.