With PoW you had capital expenditure, but also operational expenditure (e.g. electricity costs). Your capital also depreciates over time.
With PoS, you don't have any operational expenditure, and the sticker price of your capital expenditure stays the same, and you can get it back when you unstake.
That’s not true. There is operational expenditure, like electricity cost, internet data, hardware that holds the client softwares, etc.
It’s just that the depreciation is slower, and less expensive over time than GPUs or ASICs.
Maybe this isn’t the case with some of the other DPoS chains out there, where people can delegate their stake to validators, making it unnecessary to run any hardware.
Interestingly, unstaking is not actually supported by the network yet so the staking only goes in one direction. The price and demand impact when that does go live will be interesting to watch.
Not OP, and not about opex specifically, but around PoS and centralization...
Before we had ~3 groups (miners, holders, users) that all kind of needed each other. Now there's no more miners. Given that crypto loves zero-trust and all that, I think it'll be an interesting experiment to see how the randomness of staking allocations plays out; if randomness streaks towards major capital, it'll look like they're favoring themselves and their stakes will accumulate %-wise increase at a higher rate (centralization!). The other "random-streak" outcomes aren't as bad (imo) and there's some game theory around this topic that I'm only topically versed in. Complicated by the part that miners did have some of their own problematic incentives and externalities.
In summary I view it as moving away from an unstable 3-body problem down to a more stable/centralized 2 bodies, one of which has greater influence. Hopefully good stewards and all that, but instead of forced cooperation among the 3 we now mostly trust 1 group.
Basically yes but in fairness there is a bit of operational expenditure. You need to pay a bit for energy (there is still a computer that needs to run), internet costs and your HW may need to be renewed every 5 years or so (perhaps longer). Typical costs of a normal consumer grade computer plugged to the internet, almost insignificant but not strictly zero.
PoW was worse because it had an economy of scale. For example, it is far cheaper to add mining power if you already own a big mining centre or buy ASICs in bulk.
PoS does not have this: your rewards are always linear to the amount of ETH you stake.
This means that, while PoS is still controlled by those with the most money, it does not trend to centralisation as harshly as PoW.
Pos is controlled by those with the most money, and they continuously gain more money through staking rewards (i.e. the rich control the system and automatically get richer).
With PoW, you have to sell/spend some of the coins you earn in order to pay for operation expenditures.
I don't see how it is beneficial to anyone (other than ASIC developers and energy companies) to build a tax paid to ASIC developers and energy companies into the protocol.
The energy expenditure anchors the money into the real world, making it a hard money like Gold, and it also facilitates decentralization.
Energy is naturally decentralized all over the world.
A "tax to energy companies" is a subsidy from the energy company's point of view. If you want more of something, subsidize it. A world with more energy is better than a world with less, as we're all in the process of relearning.
There is no evidence that the exchange value of any PoW coin has anything to do with the energy used to mint it. In fact we have counter evidence, for example that there are PoS coins that have value
That's the same in both cases. Both times you are turning money into more money. Literally the only difference is that in PoW you have external costs. But since these costs are physical and out of the scope of the chain, they do not scale linearly.
In theory, yes. But in practice, most miners have joined a mining pool. Mining pools allow miners to share in the rewards, which means they have a vastly more predictable income per block. A solo miner would probably not have a statistically significant chance of finding a single block for the next 100 years, while pooled miners earn bitcoin through finding blocks roughly equivalent to the mining pools relative size compared with the total hash rate. Since mining is a cutthroat, bleeding edge, hypercompetitive system, that means dependability and stability are very important.
It's not "lottery as a banking system", because there are rules for validators for failing over and others replacing them. This is what complex consensus mechanisms are about - how to have 100% uptime instead of few nines. The system has built in incentives for the operators to keep it running smoothly, but still not being unable to change or reject the transaction payloads, like PayPal or banking system could.
This is categorically incorrect. You can have any amount of ETH (up to 32 because there's no value in going higher) and be a validator participant if you partake in a pool.
Sure, but that doesn't make the difference between getting a lottery ticket and not getting one. If you have 1 ETH in the pool when the validator hits, you get the corresponding fraction of the award, minus the pool fee.
I mean......isn't that any lottery? People who buy 100k lotto tickets have an undeniably higher chance of winning than people who bought a single ticket. Here it's the same - the higher your stake the higher the chances. But you don't need to stake all 32 eth to participate.
