Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Mainnet Merge Announcement

blog.ethereum.org

Post reply on HN
Print view View original

511–520 of 609 posts

Re: Mainnet Merge Announcement

#511

Post by NavinF »

AlexandrB wrote: post #298
exo762 wrote: post #239

Earlier quoted context omitted.

How can you detect that people are NOT running slasher nodes? This model assumes that 1/n is honest. This is almost as safe as it gets.

Depends how big "n" is. It's certainly not the number of ETH users. Running a slasher currently requires 1TB of SSD among other things[1]. I assume these requirements will grow with the popularity of ETH? If so, in a world where 100TB of SSD space is required the number of individuals who can run slashers drops rapidly. In the extreme case, aren't we looking at a small(er) number of well-funded organizations doing al…

A 1TB SSD is like $50. I put one in a small USB-C M.2 case and use it as a flash drive.

I’m pretty sure hardware costs will always be a rounding error compared to the $50k that “home stakers” need if they want to run their own validator.

Re: Mainnet Merge Announcement

#512

Post by lupire »

tboyd47 wrote: post #279
proto-n wrote: post #266

Earlier quoted context omitted.

Let me make sure I get this right: if you have two chains, one is mined by 49% hash power, the other is mined by 51% hash power, you think the 51% one won't be the longer one?

As I've explained in nearly every comment, one is mined by 100% hashpower, the other is mined by 51% hashpower. The 49% are still using the other blocks, too.

It's a blockchain, not a blockbag.

Re: Mainnet Merge Announcement

#513

Post by koheripbal »

drog wrote: post #461
usrusrusrusr wrote: post #417

Earlier quoted context omitted.

>There is no designated class of slashers. Not true. In your defence, I also only learned this today. https://docs.prylabs.network/docs/prysm-usage/slasher https://lighthouse-book.sigmaprime.io/slasher.html

I know about that but thanks for your comment - I’ll clarify what I mean. It’s just the specifics of how this works - slashing is part of the protocol in the sense that I described: when you are chosen to create a new block and you have proof that someone violated the rules then you include this proof in your proposed block and update balance. Any validator can do it including the smallest of home stakers. More detai…

Slashing is only detrimental to the bad actor if the fork fails. If the fork succeeds, then there is no penalty for a malicious fork.

This is why they are limiting the number of validators.

Re: Mainnet Merge Announcement

#514

Post by IncRnd »

dsimmons wrote: post #479
IncRnd wrote: post #474

If anyone wonders why many "regular people" are uncertain about doing anything with cryptocurrencies, look at the top of this article for the reason. It's not that people are incapable of understanding the subject, but that the purveyors of the subject constantly coin new words and meanings. * Ethereum is moving to proof-of-stake! The transition, known as The Merge, must first be activated on the Beacon Chain with th…

To be fair, I think this is mainly aimed at developers (and ETH ecosystem developers specifically). While I don't disagree that there's a "vocabulary" problem (e.g. "dank-sharding"), in a perfect world, a normie would largely just interact with an application without needing to know any of this.

You say that from the standpoint of readers at this HN blog page, but this blog post is the top blog post and first one seen on https://blog.ethereum.org/. Many people who want to understand what they are looking at prior to purchase would look at this page and quickly close it, never to invest using etherium.

Do you download binary utilities for execution, when every page on the website (or app store) is filled with made-up jargon? That's the issue that I see here.

Re: Mainnet Merge Announcement

#515

Post by DSMan195276 »

tboyd47 wrote: post #243
remram wrote: post #184

Earlier quoted context omitted.

If the 49% mine a block, the 51% can ignore it and will eventually surpass the 49%'s chain's length long term, no matter how lucky they are short-term.

The 51% can ignore the 49%'s blocks, but there is no way the 51% can force the 49% to ignore THEIR blocks. The only way 51% of miners can fork the blockchain and actually overtake the other 49% with more restrictive rules is if they have a hidden nuclear reactor dedicated to mining Bitcoin that produces hashpower equal or more than 49% of the network and they flip it on at the moment they fork. Again, it doesn't matt…

> The 51% can ignore the 49%'s blocks, but there is no way the 51% can force the 49% to ignore THEIR blocks.

If the 51% can ignore the blocks from the 49%, then why does it matter what the 49% is doing?

Occasionally the 49% will be able to mine a block and temporarily create the longest chain, but the 51% can always just ignore that block and continue mining off of the one before it - eventually their chain will be the longest and the block(s) from the 49% will be lost.

Really this isn't theoretical, it happens all the time by accident when two blocks get mined off of the same previous block. When that happens one of the two blocks gets lost, and the only difference with this scenario is that the 51% have enough hash power to ensure that their blocks are the ones that always (eventually) win and the 49% blocks are always lost.

Re: Mainnet Merge Announcement

#516

Post by jbaczuk »

stale2002 wrote: post #377
jbaczuk wrote: post #242

Earlier quoted context omitted.

> Even wind/solar/nuclear has a carbon cost to it. Yeah, so does life.

But the point here is that if there are easy ways to reduce energy usage, that have few drawbacks, then that is a good thing to get that free lunch.

I understand. My point is there is no reason to reduce energy usage. There is reason to reduce "dirty" energy usage. There is an abundant supply of "clean" energy.

Re: Mainnet Merge Announcement

#517

Post by jbaczuk »

rcxdude wrote: post #283
jbaczuk wrote: post #209

Earlier quoted context omitted.

If your answer to clean energy to just use less energy, that seems inane. Also the increase in energy costs because of the war in Ukraine is directly related to dependence on foreign fuel which is not clean. So the problem isn't energy consumption, the problem is using fossil fuels.

Energy reduction is basically the best possible option. It's why insulating houses (in both hot and cold climates) is possibly one of the most effective green initiatives, since so much energy goes into heating. Even with a pure renewables grid where you have huge excess of energy at certain times of day and year you will have bottlenecks due to inflexible demands when supply is at the lowest and that will determine…

> Even with a pure renewables grid where you have huge excess of energy at certain times of day and year you will have bottlenecks due to inflexible demands when supply is at the lowest

Not with good energy storage solutions.

Re: Mainnet Merge Announcement

#518

Post by jbaczuk »

falcolas wrote: post #316
jbaczuk wrote: post #88

Quoted post unavailable.

> here is no a shortage of available energy we have a nuclear fusion reactor in the sky that provides more energy than we could ever use. We can only capture a small fraction of that energy; limited by land space and efficiency of the capture devices. In fact, we capture so little that we primarily rely upon the sun's energy that fell on past-earth to power our world today. To take this idea to the next step, even if…

We are not limited by land space. What do you mean "fell on past-earth"?

And it's not wasteful if those government controls are harmful.

Re: Mainnet Merge Announcement

#519

Post by unixbane »

IncRnd wrote: post #474

If anyone wonders why many "regular people" are uncertain about doing anything with cryptocurrencies, look at the top of this article for the reason. It's not that people are incapable of understanding the subject, but that the purveyors of the subject constantly coin new words and meanings. * Ethereum is moving to proof-of-stake! The transition, known as The Merge, must first be activated on the Beacon Chain with th…

This. The wikipedia page for years seemed (not looking at it again, lol) to be written by people of the same ilk, who introduce a bunch of terminology without explaining what any of it means. When I read the wiki in 2020 it was literally impossible to understand, knowing already at the time how Bitcoin and a few other alt coins worked, as well as someone being into smart contracts well before even Bitcoin was created. And Ethereum is literally just a Bitcoin + smart contracts. There is no other core, fundamental concept to understand in it (until now: PoS). Also yes, as someone who has used hundreds of cryptocurrency products, this applies to basically anything related to cryptocurrency like you said.

One absolutely obnoxious practice is how they all make their own fancy names for millicoins microcoins, etc.

Re: Mainnet Merge Announcement

#520

Post by fleddr »

camjw wrote: post #47
davidy123 wrote: post #32

Oh funny, I just, inspired by Stable Diffusion, ordered a 3090 ti at what I thought was a good price. Well, I have 30 days to see what happens… There could be an interesting ripple effect if GPUs become more widely available in general. «Simple division yields an estimate of 13.982 million GPUs mining Ethereum on April 16 2021.» (from https://linustechtips.com/topic/1327701-honest-question-how-... )

Off-topic but have you got any recommendations on stuff to read to set up a GPU at home to run stable diffusion?

https://www.youtube.com/watch?v=z99WBrs1D3g
Post reply on HN