Delaying the Inevitable: Muir Glacier and the Ethereum Difficulty Bomb | Hacker News Reader

Delaying the Inevitable: Muir Glacier and the Ethereum Difficulty Bomb

blog.cotten.io

90 pointscottenio70 comments

> Why? Because Ethereum isn’t just a store-of-value. The Ethereum network also contains smart contracts that execute native code on the blockchain.

It's funny when things are presented this way. Ethereum was never intended to be a store of value. It's a distributed computer that needs its own token economy in order to charge "hosting costs" to the computational agents running on it, and to allow those agents to trade work done for other agents for transfer of "hosting costs."

Ethereum would still be doing what it's designed to do, even if the price of ETH tanked. (In fact, dapp developers would probably prefer ETH to tank, since that'd mean transactions would be cheaper and less crowded by speculative traders and selfish miners.)

The other funny thing is the common misconception that Bitcoin does not have smart contracts. Bitcoin transactions are actually pretty programmable. You can have a coin be unspendable until a certain date, or have it require n cosigners to be spent, or have it become spendable only once a secret is revealed which matches a hash.

Bitcoin most definitely has smart contracts.

This is true in the same sense that grep is extremely programmable, but comparing grep to C would be strange indeed. Ethereum is a Turing complete VM, allowing you to build extremely complex systems (we, for example, are trying to build a more transparent and distributed directory for a Tor-like service), while Bitcoin lets you sort of control transactions a bit with what amount to something closer to "rules" than "programs" (due to the explicit lack of looping control structure).

> A vibrant ecosystem of banking functions, asset management, trading exchanges, and more have sprung from this, requiring Ethereum to quickly respond to external events and triggers.

That's a bit of a stretch isn't it? Other than speculation and money laundering, I don't think any dapps have gained any traction at all, let alone a vibrant ecosystem.

Decentralised Finance (Defi) apps are starting to gain traction on Ethereum. Theres about half a billion in funds currently making use of them. https://defipulse.com

Even those are mostly exchanges (speculation), derivatives (extra speculation) and stable coins (to enable speculation on exchanges which want to avoid AML and KYC in the facilitation of money laundering).

Whether you agree with the use cases or traction they're gaining doesn't negate that there's a lot of others that do find them useful.

That's totally fair and I only call it out because in my original post I said "other than speculation and money laundering" (which I freely admit is a big use case) and you provided me a list dominated by speculation and money laundering.

Maker and Compound both support interest and borrowing which are what you'd find in a traditional bank (at much worse rates). The money laundering argument could be applied to anything that can't be controlled by governments. Unfortunately there are many countries where monetary policy is oppressive so Defi finance is a legitimate alternative. Just because technology can be abused doesn't mean that it shouldn't be used. See the internet etc.

You're completely missing the point raised though.

They said there are no real uses of this technology outside of speculation and laundering and your response is that tools for speculation and laundering are useful. That's not really what the argument is here.

The argument is that this technology is being touted as diverse and widely popular when it in fact is not and mostly only used in circular use cases involving itself.

No I just said you can use it for saving and borrowing which have nothing to do with speculation or money laundering. Also it’s impossible to tell what percentage of transactions are related to money laundering in crypto and you can’t outright dismiss legit use cases because of this.

The opposite of anti-money laundering (AML) isn’t actual money laundering

AML implementation is a poor and misapplied user experience that is usually done as an excuse to prevent money laundering. Many services apply it arbitrarily with no scrutiny and use it to steal users funds. Unclear regulations per country dont help and unscrupulous actors take advantage.

Onchain and noncustodial services are either exempt or function autonomously, removing this particular duty of the state. They also typically have unlimited amounts that you can move, compared to arbitrary monetary amounts that centralized systems impose. “DeFi” efforts are being made to make the experience of using onchain noncustodial services as fast or faster than offchain custodial services and a lot of progress has been made towards that.

Regarding the actual money laundering everyone’s afraid of, even the strictest regulations never actually prevented that while taxing all business and financial institutions. The Patriot Act wouldn’t have flagged any of the 9/11 transactions, HSBC still laundered billions for the actual cartel, minorities go to jail for accidentally using over $10,000 in cash, and even Al Capone could have passed the KYC part of AML regulations, so who is this for? Any way its going to be moot and already is for a lot of people.

We use the stablecoin Dai for althea.net. There are several networks running providing internet to people with wireless mesh nodes that pay each other with Dai.

Sounds like pied piper.

You quoted "a vibrant ecosystem of banking functions, asset management, trading exchanges, and more" and called it a stretch. Now you're complaining it's just stuff for speculation like exchanges and derivatives. That's literally what the quote said was there.

A popular asset management system is this: https://www.tokensets.com/

Maker and Compound allow borrowing backed by collateral.

Decentralized exchanges include the various DEXs and Uniswap.

All of these get substantial usage.

Would you say the same thing about cash? Because Dai is the closest thing out there to true digital cash.

Cash is great for money laundering. Much better than crypto.

Sure, but they're not speculating on the value of crypto (which is what people are actually concerned about when they talk about speculation.) These apps are just generic infrastructure, like a stock market is generic infrastructure. And, as such, they have value as infrastructure, independent of any of the assets people might list on them; just as a stock exchange has value independent of the ridiculous penny stocks people might list on it.

Origin Protocol Dapp is a good use case. Non speculative, uses DAI for payments.

Not a lot of traction yet, but definitely vibrant.

"Money laundering" is a dysphemism. The proper therm for it is "electronic privacy", or even just privacy.

Unless you're an ancap, this argument doesn't actually say anything; you can protect the proceeds of any crime from investigation and impoundment under the aegis of "electronic privacy". When people say "money laundering is a feature of dapps", it's pretty clear that they're referring to organized crime.

If you are an ancap, more power to you, but it's useful to have arguments that are persuasive to the huge majority of people who are not.

"Money laundering" is also when you conceal the source of funds that haven't been tainted by crime

Everyone should have the right to privacy, electronic transactions are no different.

Of course, you can use the good old "if you have nothing to hide then you have nothing to fear" argument, which has already won over...

That's true, but when people say "Ethereum is used for money laundering", that's not what they mean, so it's not a useful argument. If it helps, just substitute "organized crime" for "money laundering". That's what people mean.

Ethereum doesn't offer any serious financial privacy (yet) so neither does your statement make any sense.

Privacy is important though, doesn't matter if it's financial or not.

In the future, should more serious privacy options come to Ethereum, then it's important that everyone should be entitled to use it, just like we are allowed to use end-to-end encryption today.

> under the aegis of "electronic privacy".

Regardless, that still doesn't mean the electronic privacy is the same as crime.

Or, would you be willing to hand over every bit of electronic information about you over to the government?

Oh, you don't want that? Well, what do you have to hide then?

As an outsider to the blockchain world, I would be more interested to know why proof-of-stake is not used by Ethereum yet.

Found one article on the subject: https://medium.com/ibbc-io/the-beautiful-complexity-of-pos-3...

Because it's hard. Also technical debt moving from Eth 1 to 2 makes it harder.

My guess is another blockchain will execute PoS better because it won't have tech debt.

The original plan would have been hobbled by technical debt. They were going to implement the core algorithm as a smart contract on the old chain. It would have supported a much smaller number of stakers and shards.

In mid-2018 they threw out that plan, choosing instead to implement PoS as an entirely separate chain, integrated with sharding. This entailed a delay of over a year, but it meant a minimum stake of 32 ETH instead of 1500, and an overall much more efficient and scalable system. There are three main phases, the first being the core proof-of-stake "beacon chain," which is currently running a multi-client testnet.

There have been various proposals for migrating legacy Ethereum into all this; here's the latest: https://ethresear.ch/t/alternative-proposal-for-early-eth1-e...

Cardano is PoS and released its testnet last month.

Tezos is another.

Cosmos is PoS and released its mainnet in March.

A Proof of Stake consensus for a public blockchain like Ethereum has different requirements and is substantially harder to execute than for a permissioned network.

Definitely agree, although it’s important to clarify that Cosmos is not permissioned. Anyone can start a validator and join the validating set.

Proof-of-stake is cryptocurrency's fusion power: perpetually "less than a year away"; it's the technology of the future and always will be.

What are the problems with current PoS systems? Not being glib, seriously interested as there are at least a few that implement some kind of Pos (for example dPoS with Ark)

Permissioned networks are a very much easier setting to design a consensus protocol for than public ones.

Describing a consensus algorithm as "Proof of Stake" simply means it uses some kind of stake to secure its decisions, this doesn't say anything about the assumptions it starts from.

Existing "PoS" blockchains don't have the same decentralization goals as Ethereum.

> Existing "PoS" blockchains don't have the same decentralization goals as Ethereum.

What about Cardano and Tezos? aren't they decentralized PoS blockchains?

And Cosmos

How is Ark less decentralized?

“Existing ‘PoS’ blockchains don't have the same decentralization goals as Ethereum.”

What makes you think this?

From my understanding, the issue with PoS is that if a chain forks you can mine both sides without penalty. A consequence of this is that if someone wants to retroactively create a fork that starting from a block a few years ago, then people could do a few years worth of mining of this fork in an instant. And it wouldn't be possible to tell these two chains apart cryptographically. With PoW it would be computationally expensive to pull such a stunt.

Not exactly, it's here and technically working in multiple production blockchains, it's just less secure than Proof-of-Work. So it depends what your definition of "works" is.

Proof of Stake has a separate set of problems to worry about than Proof of Work. It's a solution to some of the scaling problems, but at the expense of some security.

As with all engineering, it's a set of tradeoffs.

> [With PoS, Ethereum would lose] the baggage of wasting the electricity of a small city to keep itself running.

"Small city" sounds suspiciously small. Last time I checked, Bitcoin had the same electricity consumption as the entire country of Austria. I can believe that Ethereum has less miners, but by that over 2 magnitudes?

> but by that over 2 magnitudes?

Why do you question the claim?

The power consumption doesn't scale between the two by the quanta of miners, it scales differently because the algorithms are different and thus the power consumed is different.

I have no idea how the algorithms differ nor how the difficulty differs among the two, so I don't have an intuition about how to guess whether or not they should scale relative to each other. I suspect that ethereum's market cap would be lower and probably have lower mining rewards therefore less capital invested in its global mining pool. I suspect that ethereum's algorithm requires more energy per block than bitcoin's, but I have a hard time guessing how much. Especially since it's a little less mature, maybe fewer ASIC miners so even less efficient in practice.

There’s also the fact that Ethereum has an almost completely different architecture including block times which affect the frequency of mining, different reward structures, etc. As far as I know its not necessarily the difference between Bitcoin’s SHA-256 vs Ethereum’s Ethash as much as it is about the architecture itself.

There's so many inputs that it's difficult to casually come up with an intuition-guided estimate. So it makes some sense to have faith in the empirically guided value. We hope that they took the time to validate their algorithm for extrapolating the energy consumption.

The trinsicoin algorithm is really just a minimum bound based on the most efficient rigs known for a given algorithm and the cheapest industrial rates for electricity. The Litecoin Foundation actually helped out with some of the math and data resources.

The total power consumption of any given crypto currency scales according to the ongoing value of new block rewards per fixed unit of time. You can ignore the rest of the factors (like algo, block time) for the most part. Miners run if they can make money for a given cost of electricity, and don't if they can't.

Indeed; see for example this screener showing daily issuance in dollars:

https://messari.io/screener/play-around-074BC5B4

from which we can infer that Bitcoin uses about 10x more electricity than Ethererum.

I can't comment specifically on the details, but one consideration is that Ethereum's mining hash is designed to be ASIC resistant.

Bitcoin is entirely mined by dedicated hardware that's incapable of doing anything other than mining Bitcoin. Ethereum is mostly mined by GPUs that can obviously be repurposed for other applications if need be.

If the electricity cost for Ethereum mining gets too out of control, the miners can just point their GPUs at some other task. Whereas Bitcoin miners are stuck with sunk cost of rapidly depreciating capital, and pretty much will never turn even during electricity price spikes.

But the ASICs are used because they are more efficient. Forcing miners to use less efficient hardware is going to drive up the average energy per hash calculated.

The system is meant to balance itself out one way or the other. You throw more efficient resources at it, and it becomes less efficient to maintain its 10-20 second block times. It's actively anti-efficient.

The idea behind ASIC resistance is not efficiency, but rather to limit centralization. If you can use generic hardware that anyone can access it is intended to reduce capex required to become a cost-effective miner. Now of course, I don't think that makes sense, but that's the idea.

Ah, you're right. After having enough hashes to secure against attacks, they don't really do anything useful. People will just mine as many as they can with a given budget, which buys the same number of kWh no matter how efficient the hash calculations are.

You're completely ignoring the difficulty mechanism in Bitcoin. When everyone has a 1TH/s miner that consumes 1kWh and suddenly a 10TH/s miner that consumes 1kWh comes out everyone is going to switch to the new miner. Because Bitcoin guarantees that a block is mined every 10 minutes it has to make the calculation 10x harder than before. The end result? The energy consumption didn't change.

What makes Bitcoin more energy intensive then? The price of Bitcoin. When the price of Bitcoin doubles that also means the mining profit doubles. More miners (= more energy consumption) join the blockchain until the profit margin is back to the previous level. The price of a Bitcoin is 70x higher than the price of Ethereum so that suggests that Bitcoin should use 70x more energy.

I don't think you can directly compare the prices of the coins, but you need to look at the reward that's at stake.

12.5 bitcoins every 10 minutes at 7277 USD/bitcoin = 9096 USD/minute

3 eth every 14 seconds at 132 USD/eth = 1697 USD/minute

then you'd expect the bitcoin miners to use 5.3x the energy of eth

EDIT: oh, someone already did the same calculations earlier a couple of comments below :)

Basically correct, except that it isn't the price of Bitcoin or Ethereum that matters, per se, but rather the value of the reward for each block relative to the amount of time required to mine it. For Bitcoin, ignoring transaction fees, the rate is 12.5 bitcoins per 10 minutes; for Ethereum it's 3 ETH per 15 seconds. If the price of 1 BTC is 70x the price of 1 ETH then you would break even expending about 7.3x over a given interval to mine 12.5 BTC vs. 1x to mine 120 ETH. If you spent 70x, however, you'd soon be bankrupt despite the higher BTC price, because you don't earn 70x the value.

So energy usage should drop quite a bit after some difficulty increases?

The difficulty increases or decreases to maintain the target block interval. The only way that impacts mining revenue is if the actual block interval deviates significantly from the target—for example, if the hash rate suddenly increased such that blocks were being solved in five minutes instead of ten, that would double the revenue rate (and thus the economical power consumption) temporarily until a difficulty increase brought the interval back to ten minutes per block.

What does cause significant changes in revenue is the periodic halving of the block reward. Absent an opposing increase in price and/or transaction fees, when the reward drops from 12.5 BTC/block to 6.25 BTC/block in May of this year the energy budget for profitable mining should be cut roughly in half.

Sorry yeah I was mixing those up. But is it going to use not-obscene amounts of electricity eventually?

Etheriums PoW algorithm isn't ASIC resistant, just largely unimplemented. Indeed a large reason NIST made Keccak SHA3 is that ASIC/FPGA acceleration is on par or even faster than SHA256 acceleration.

It may take a bit more ram and need a decent bus as a result which is why people mine on GPUs that have both. Rather than cheap ASICS that are cheap because they have neither. "Resistant" ain't the word to use.

The article actually has a good link to https://www.trinsicoin.com/?focus=ETH where if you scroll down it breaks down the math.

(You have to hover over Blockchain, Mining, and Energy to see the parameters.)

According to https://ethernodes.org/ there are ~7.5k nodes & according to https://www.etherchain.org/charts/distinctMiners ~60 miners running ATM. Small city sounds a bit big in that context.

This link to the list of Ethereum mining pools might help make more sense of that: https://www.poolwatch.io/coin/ethereum

As you can see there are ~570k workers contributing to ~79k mining addresses. A lot of the data you see about "distinct miners" might be pools rather than your loner GPU miner.

Ethereum mines 2 ETH (exactly) per 10 seconds (approximately).

Bitcoin mines 12.5 BTC (exactly) per 10 minutes (approximately). Normalizing the ETH rate against the BTC, we get 2 ETH * 6 * 10 = 120 ETH.

So at current market rates, 120 ETH * $132 vs. 12.5 BTC * $7,341, $15,840 ETH for $91,792.5 BTC.

This suggests that the BTC network should use roughly 5.8 times as much energy as ETH, to derive an equivalent profit per unit energy.

A number of blockchains already have proof of stake:

https://tezos.com/ for one is pretty cool

So Muir Glacier is not mentioned anywhere in the article text.

That is hilarious and I fixed that :P

Given that they keep postponing the transition to PoS, wouldn't it be smarter to remove the artificial difficulty increase and make the network more efficient?

The reasoning is that somewhere down the line introducing an ice age may lead to a contentious hard fork. Better to include it upfront so that expectations are set that PoW on Ethereum is not a long term thing.

Fwiw the transition doesn't look that far off now. There's a working multi-client testnet for the new (separate) proof-of-stake chain, and a recent proposal by Vitalik to transition the old chain into it more quickly than previously expected: https://ethresear.ch/t/alternative-proposal-for-early-eth1-e...

Reply on news.ycombinator.com