Ethereum Cat Herders PEEPanEIP, 7 Jul 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: PEEPanEIP #39: Difficulty bomb & EIP-3554 with James Hancock

Description

Topics covered
-----------------------------
1:04 - Introduction
4:10 - Why was the Difficulty bomb introduced?
6:33 - Block time mechanism
8:44 - Why block time recedes (ref to chart)?
10:05 - Why does hash rate matter?
10:41 - Difficulty bomb analogy
11:58 - Summarize the working of difficulty bomb
13:50 - Difficulty equation
14:25 - Takeaway from 2016-2017 difficulty bomb appearance
15:30 - EIP-3554
16:10 - How to calculate the difficulty bomb using a fake block number?
17:45 - Python script for difficulty calculation
19:00 - Simulation of what will happen with half the difficulty & double the difficulty
22:27 - History-timeline of difficulty bombs
23:17 - How is the pushing of the blocks decided? Is there an algorithm? Why is there a big difference in pushing the blocks from Muir Glacier & London?
27:40 - Why ice age exist?
32:20 - Do you see a future for difficulty bombs with PoS & Merge on the horizon?
34:30 - Will EIP-3554 be the final EIP for difficulty bombs or we can expect more?
36:07 - Does market price affect the appearance of a difficult bomb?
38:00 - Is there any research work done on the economic effect of the ice age?
42:55 - What is a PoA chain? Why is the difficult bomb proposal not activated on the PoA chain for testing purposes?

Resources:
-------------------
EIP - https://eips.ethereum.org/EIPS/eip-3554

Discussion - https://ethereum-magicians.org/t/eip-3554-ice-age-delay-targeting-december-2021/6188/8

Slides - https://docs.google.com/presentation/d/1jDIA0TwbzB87WW_3Y0hD7lONUG__nK5cfvGajErv7Wo/edit#slide=id.p

Ethereum Muir Glacier Upgrade (EIP 2384: Muir Glacier Difficulty Bomb Delay) - https://medium.com/ethereum-cat-herders/ethereum-muir-glacier-upgrade-89b8cea5a210

London upgrade proposals - https://www.youtube.com/playlist?list=PL4cwHXAawZxppsQYazgJ3EWWWjY2vNxVp

PEEPanEIP playlist - https://www.youtube.com/playlist?list=PL4cwHXAawZxqu0PKKyMzG_3BJV_xZTi1F

Survey:
---------------
Ethereum blockchain users and developers research - https://docs.google.com/forms/d/e/1FAIpQLSeugRHv93fizew0zeqXFOOnQIbjbWnVHoThLJR3f3g6gVvXhQ/viewform

Follow at Twitter
-----------------------------------------------
James Hancock @JHancock | William Schwab @William94029369 | Pooja Ranjan @poojaranjan19

Contact Ethereum Cat Herders
---------------------------------------------------
Discord: https://discord.gg/tzYmDmF
Twitter: https://twitter.com/EthCatHerders
Medium: https://medium.com/ethereum-cat-herders
GitHub: https://github.com/ethereum-cat-herde...
Email: support@ethereumcatherders.com
Website: https://www.ethereumcatherders.com/

A

Hello and welcome to another episode of a video series where we discuss upcoming features and improvement proposals for editing, blockchain, I'm putin, and today we are going to talk about difficulty bomb or ic genomia, along with the latest improvement proposals to deal with the situation, eib 3554 to push the bomb to december 2021 to talk about the difficulty bomb. We have god the author of the improvement proposal and also a contributor to penny series. James hancock welcome james.

A

It's always fun to have you on bbene, but until today you were on the host side, so this is the first time you are on the hot seat as a guest. I wouldn't miss the opportunity to know you more uh so before we get into the proposal discussion. Maybe please talk a little bit about you, how you got into tdm and the journey so far.

B

Yeah I'd be happy to it's, it's funny being on this side, I feel like I don't know what to do, but normally just kind of go I'll, go along any any questions or interruptions at any time. Please feel free to do that, um so I started into ethereum in 2018.

B

I was in washington. I was in washington dc and I drove a car to toronto and I just was committed to that summer. Finding cryptocurrency people to work with. So I I got an apartment. I met some some friends. I was there for like three months: nothing really materialized. Until the last week I snuck into edcon toronto.

B

um I I had a I went to yoga earlier and I had my yoga mat with me and because the yoga mat and my yoga clothes were so distracting. No one checked that I had a tag or not.

B

So that was there I met, and then I met some ethereum people there I got my first thing I got into was eve signals with griff green, and that was awesome. I flew then, to berlin and so far it had been kind of ancillary to the ethereum ecosystem.

B

But actually my introduction to or ethereum and working on and what ultimately led to me working at the ethereum foundation um is this proposal that I did at the status hackathon. That was right after uh devcon that was in prague. So I think that was 2018 right.

B

Yeah 2018 fall yep, so I I take the train down. We go and tj russia, and I was there and lane reading was there and lane started talking about this difficulty bomb and how it was in ethereum, and I was just kind of like what in the world does that mean um they were trying to figure out when it would go off and what does what does all that mean? So I spent my the hackathon basically researching. How does the difficulty bomb work and then figuring out?

B

How can we figure out what it's going to do when.

A

B

We can make plans for it appropriately, and so I made this presentation the same one that I that I gave at the status hackathon um and then that was kind of my introduction to people in ef and then from there. I started kind of volunteering all stuff anyway, that was probably longer than it needed to be, but that's the story into ethereum.

A

That's all good, it's always fun to know people we are working with. Thank you so much for your introduction. So without further ado, let's peep in.

B

Yes, so I this is the presentation and like boom: when will the time bomb fit happen to you, and so, when the when ethereum first started, they wanted a kind of there's a whole bunch of reasons that people like the difficulty bomb or think that it existed or different things.

B

But as far as I understand, the original reason was a forcing function that when we moved proof of stake, the miners can't just keep the chain going. So if you wait long enough, this, the ethereum blockchain will just stop making blocks in a reasonable time to actually make like the block time just start increasing.

B

um So there's these comp there's these formulas in the in the yellow paper that take a while to kind of like took us a while to kind of figure out and get into, um but you can see as the difficulty bomb goes off. This is the average block time you can see that they start getting longer and longer and longer and in like these chunks.

C

um Did you want to talk about like the different parts of the equation over there, or even like give a brief overview about exactly like what makes the difficulty bomb kick in, or maybe why the ridges seem to have retractions every time they start going up.

B

C

B

Start yeah yeah, we'll start with this and then I'll go deeper into it later on. So the the as we start first with ethereum, has a block time targeting mechanism, so the hash rate and ethereum is always going up or down, but we want to have consistent walk times so there's a mechanism inside of it. That is that targets uh block time and the way ethereum's works is if it's below 10 seconds.

B

It increases the difficulty. If it's above.

A

B

Seconds so the block, so it finds a new block, it decrea. It decreases the difficulty and if it's like over 30 seconds, it doesn't more and if it's 40 seconds it doesn't work so they're sort of uh on one end it can correct further than on the other rim. So there's the: where was it the blockchain?

B

The block time mechanism ends up targeting about 13 at 13 14 seconds, and we don't actually know why that's value because it doesn't it's.

A

B

13 seconds individually, like specifically.

A

It's literally just saying.

B

If it's less than 10 seconds do something more than 20 seconds, do something and then the complex interaction ends up having it have an equilibrium at the 13.5, and I think it has to do something with uh so I someone sent me that the exponential distribution on distribution of blocked times end up being the geometric mean it's about like 13 out of fronts, from 10 to 20..

B

That's not super critical! So when they put in the difficulty bomb what it does, how how it works. Is it it messes with that mechanism that ethereum uses to target- um and it does so in like a kind of interesting step function, so you can see in here on these like little green dots and then it goes up flat and then the green flat for a while and it jumps up and then it's flat for a while and it jumps up and it's flat for a while.

B

So this is every 100 000 blocks the difficulty bomb increments one um and that has like it increments to the next step.

B

So it's sort of like you, get this kind of interesting step function and you see that translated into block time so whatever so. If you look at this, every single one of these sawtooths is a hundred thousand blocks. The difficulty bomb kicks it up a notch. A hundred thousand blocks typically bomb kicks it up.

B

Does that help? Should I go further into that william? All right am I losing everybody already.

C

No, that sounds great, um I mean maybe as a bit of a follow-up in the slide before this it looks like any time. There's one of those hundred thousand step like a hundred thousand block steps.

C

um It looks like then, the block time kind of recedes a bit is: does that also have to do with the the equation, or is that just some also kind of inscrutable property of the the hash power that gets put into ethereum.

B

It does it, it is this that's well spotted and the what so the the blockchain, even though it's over 20 seconds here, 25 25 seconds here, like the difficulty mechanism, is still trying to get below point, but.

C

Okay, I got it.

A

B

So like it first increments, and then you see the targeting mechanism doing its best to get it back down below 10 or below 20, but the difficulty bomb increasing is much so it it it like. It can't completely overtake, um got it yeah, that's actually a good segue into so this is way previously when hash rate was less the first time the difficulty bomb. The hash rate looked like this and then the next time the hash rate was almost double, so why? Why does that matter is because this is all I.

A

Actually brought.

B

A like a prop to kind of help explain so, if you think about in in this equation here difficulty equals parent every the difficulty of the next block involves the value of the last difficulty, so there's sort of like a feedback loop, that's happening because the you look at the last one and then you use that to tell you how much you're going to push up or down so the example I like to give is that the difficulty bomb is like this balloon and it's always there in ethereum.

B

A

B

Like the targeting mechanism and the strength of my arm is based on the difficult the current difficulty of the system and by by difficulty, I mean this like the like this, like the blockchain difficulty version, where this is how hard it is to find a block probabilistic.

B

So if I, if I have a really high hash rate, if I have a really high difficulty, my arm is really really strong. If I have a low difficulty, my arm is weaker, but most of the time the difficulty bomb is so is so weak that it doesn't matter. I can easily I can easily go up and down, but at some point so every hundred thousand blocks it.

A

Adds another balloon.

B

And then it adds another balloon and it adds another balloon and it keeps being easy for a while. But at some point it turns into the house and up- and my arm is no longer strong enough to get back to each to equal.

C

Can I try saying that back to you just to make sure I got it yep? um Basically, what you're saying is that the difficulty bomb increase incrementally adds some difficulty, but that difficulty is a bit constant. So if there's a really really high hash rate, so each little increase in the difficulty isn't going to make so much of a difference to the hash power. That, anyway, is mining on the network, but over time, as that goes up and up and up and up eventually does get to the point where it does significantly affect block time.

C

B

So the the amount of difficulty that it's adding is a constant, but the amount that the network can respond with. The targeting mechanism is relative to the difficulty.

C

B

That makes sense, so it's it's like it's kind of.

C

Weird and complicated.

B

How it works- and this is part of why it's hard.

A

B

It goes off because, in order.

A

B

When it goes off, you need to know what the difficulty is at that time.

A

And no one knows what the.

B

Difficulty is going to be in four months in any year or even less, and today we know in a week pretty good idea. The farther you go.

C

B

You know and then yeah.

C

Yeah, like china, going on right now. I think we probably actually have a hard time knowing what the hash rate's going to be in a week right now,.

B

Yeah the cash rate's certainly going down, and we don't I no one would have predicted that at the time we're trying to do before suddenly hash rate would go down, even though price levels, I think it sounds like you got it is everybody else, any other questions that can help kind of get us into the into this yeah. So here is the e we go back to this. The difficulty is the parent difficulty, so the last block, plus e e, has its own function.

B

I don't know if I put it in here.

B

But it's a constant function. I thought I had it in here. It's it's! It just will add this number so like, as I was saying, the amount of blooms that get added is the same, but how strong? I am isn't the same.

B

uh So last the 2016 2017 when the difficulty bomb showed up it was way lower value than when it showed so kind of the takeaways from that is the higher the hash rate is the longer my arm can react, and so the longer it stays hidden and then because of the nature of every step, is an exponentially larger step.

B

um It will explode faster when we see it because this increase gets bigger and bigger and bigger every time, and so then it's hidden it chills up and then boom. We have have a problem of yeah, so then tj ed and lane- and I kind of did this for the status hackathon and it was really fun.

B

um So, let's get so that was kind of the way the bomb works and how we how we figured out. However, now we get into today and what's going to happen next, so we have this eib3554, which is an adjustment to the difficulty bomb and the function we use.

B

So if so, if you, if so we, if we look back at this, the block, it's a it's a constant, what the difficulty bomb will be based on the current law. So if we're always increasing in blocks, how do we make that less and the answer? Is we fake it? We fake it by adding in a fake block number to the difficulty bomb only part, so the normal ethereum is going on. Its block number is going up, but as part of the difficulty calculation we minus nine.

B

So for the proposal is a minus nine million seven hundred thousand blocks so we're setting the ice age or the difficulty bomb back in time, nine million nine thousand seven hundred thousand, um which that will happen at the berlin fork when that happens.

B

So so at that time, we'll we'll kind of rewind the clock a bit for that 100 000.

B

This is the function that I use to come up with when and how much to adjust. So something that I noticed something that I I noticed. I went back in time and looked at all the times the difficulty bomb had went off. You can.

A

Go you can look at the block.

B

Time and you'll see like three times this kind of pattern has emerged and the ratio when that shows up is when the rate, the the difficulty adjustment ratio compared to the current the difficulty bomb compared to the current difficulty ratio. When it goes over 0.1, then you start seeing effects.

B

When the sawtooth.

C

Pattern starts showing up. Yes,.

B

So this is a program.

A

B

It's just a python script. Anyone can run just pop this into your whatever and put in your block numbers the current difficulty, how much you want to adjust and how many months in advance you want to.

A

Look and it'll it'll give you this.

B

Ratio of the difficulty compared to the difficulty bomb I'm using.

C

The word difficulty.

B

B

So the amount that the balloons are pushing up versus how strong my arm is the ratio that gets over 0.1. We start seeing when it happens.

B

So if so, I we modeled it out this script using the current difficulty- and we said- let's just use that for now and see what happens and the time that we, based on this adjustment of nine million. Seven hundred thousand, we calculated we'd, probably see in the first week of november the difficulty box and I then ran a simulation of what would happen if we had half the difficulty or if we had double the difficulty. So if ethereum.

A

B

Rate jumped up into a million or like a lot, a lot, a lot more then. uh What when might it show up and if ethereum hash rate have to difficulty have, and but it's still pretty close.

B

So as we get closer, we're going to want to be able we're going to want to run that script again. So just check, probably you're, going to want to run it for right now to see how the current difficulty bomb is doing because of the hatchery decreased due to china happening.

C

Right, I was about to ask: do you know roughly what the hash rate was when you ran that, like you're, probably in the upper 600s, I'm guessing.

C

um I took a quick look while, uh while you were talking we're holding it around 500 right now, a little bit.

B

Yes, so I the difficulty is it's hard to tell, because this is such a big.

C

Number but I'm gonna say that's 739, because I don't think it's dipped further than that. It's a 739 terahertz just to be more specific and we're at about 510 hertz right now, so I mean the truth is no.

B

This is, this is the difficult one. This is it. This isn't the this. Isn't the hashtag? Oh, the hashtag, yeah you're right. I conflate the difficulty in the hash right, yeah, good point yeah, so we want to actually look at the.

C

Historical hashtag yeah for the difficult.

B

Because the hash rate is what ends up affecting the difficulty, the difficulty is what is actually used in calculating the adjustment to the.

B

Mechanism where I just must be there.

B

So when I did it, it.

B

B

It was about here yeah, seven, so seven trillion the difficulty. I guess that's a terra.

A

B

To comparison, so it's gone down to six, so it hasn't half. So it's not going to affect the december calculation very much, but it probably.

A

B

The calculation that's happening right now for what will happen before really.

C

Interesting yeah also going to depend if, like we see hash power, get plugged back in somewhere else and all that yeah very, very interesting.

B

So we went through the fake number. We have the script on how this we predict it.

B

Looking for the ratio when it when the correct ratio happens um and that's kind of you can go so every you can actually go back. I thought I had this on this this thread, but out of it oh yeah.

C

B

So the first bomb went off may 1st in 2017 the coefficient reached one and the.

B

Adjustment was zero in the beginning, then the adjustment was three million. Then the adjustment was five million, so you can uh see how we push this.

C

That's all the times.

B

So I think that's it as far as that, what about questions, or I can um presentation now, there's like it's it's interesting, because it's not there's not like a lot to it, but it's weirdly complicated. At the same time,.

A

Right, I then first of all thank you for explaining this like going into numbers, and all these charts are really helpful. uh I have a few questions so uh so my first question is about the delay calculation.

A

My understanding is that the proposal eip 3554 that is getting into london is the fourth time like the fourth eip, which is there for pushing the difficulty bomb um earlier. We had eight 69 in june 2017 and eight one two three four in july 2018 and vip 2384. That was your glacier in december 2019. You just have shown in the fellowship of magicians right, so this calculation of these, like uh the new glacier.

A

If I take, for example, that was the last one that was like four million blocks for 611 days, and this time it is 500 000 extra blocks uh roughly five months. I believe. So. What is the basis of this delay? Calculation? It is just the time period that helium community wants to push it, or is there any any algorithm for doing this?.

B

You mean on how do we decide the day we target or how do we target the day? We decide.

A

Yeah, so whenever we decide like okay, this is the proposal and it will go for pushing this many blocks. Is it just because we want next upgrade to happen at that time, or is there any calculation going in the background that suggests that this should be done, because I see a big difference between what was there in your glitch here and this time so just trying to understand. Why is this different.

B

Yep so uh every time it's been a little bit different, um the first.

A

Time when we, I guess the first time I.

B

B

Or the pre-istanbul- I don't remember why so the reason why there's a big gap this time last time with your glacier and then there's a smaller one. This time is because uh what istanbul was delayed. It took a long time to get through.

A

B

Think the difficulty bomb was going to go off what it did and it started going off. So there was a uh eric connor came into the awkwardest chat room and said: hey block times are getting longer like oh yeah, you're right. So this was uh no end late november and istanbul was already ready to go.

B

So we made it. We we didn't want to.

A

Change this ten bowls, we made a second fork that.

B

Went out that the the mere.

A

Four glacier.

B

Was to address the difficulty bomb, and I was the only one- and we just said it- we literally set it back as far as possible, so it we. We went back in time as many blocks until you got as you could or as we could at the time, um and that was just because we wanted to not have to worry. We didn't want to show up on a thing again. We didn't want to show how to show up unannounced, and we thought we were going to do forks a little bit more.

B

This kind of gets into hardcore nation stuff and also just when and how we plan things.

B

So then we were berlin.

A

B

Up being delayed quite a bit, yeah berlin.

A

Ended up being.

B

Late quite a bit, and so that- and that was because of another eib with the bls free compile, ended up taking a long time to figure out if we power, if how to make it safe, there's a whole. I would if you want to know about that, go look at the els file and others that went into this.

B

um So that was ended up being delayed quite a bit. Then, uh when berlin happened, we decided we wanted to be a little more smart about when we're setting it. So we said we want to have something done either. The merge will happen by the note by december or we'll have some uh feature fork. That can happen within that time frame that will allow us to um address the ic and it kind of goes into there. There's a couple reasons why the ice age can exist and you can subscribe to whichever one of these.

B

So the first one is: it keeps core devs accountable, because there are people that own there are.

A

B

The keys to the github that that has the actual repository like people have commit rights to them, and if one day they said we're done we're not going to make any more changes. There's not really.

A

B

People to do about it, but in this case someone literally has to make a change to the code base or the or the change slows down into nothing. um So so there isn't like this tyranny. Of course,.

B

The other one is it's just nice to have a timing. Function like trying to coordinate hard forks in upgrades and changes to ethereum is super complicated and difficult from all the different aspects of the community that are touched by it, and so it's easy to kind of kick something down the can so having a. We need to do something by this date means we have a very easy way to say any eip.

B

That's ready by this time can make it in and anything that's not are just going to have to wait, because we have to do something at this time, no matter what, and that just makes it a lot easier to coalesce on.

B

What's what is like feasible for the time and effort to do right now as for right score developers, um then the third one is that for some reason, if there's a four like someone wanted to do a hostile fork, takeover or or whatever- um and we can go back and see how easy, because the ethereum classic had the difficulty bomb in it when the fork happened and they had to go in and disable it. So it's so doing a fork of ethereum is actually you have to anyone who wants to do it.

B

You can't run an old version of death and just say great.

B

uh You actually have to have a enough of a dev team and a way to get the information out like a way to you need to be able to make a release, and you need to be able to distribute the release in like a trusted way on your own, to get people to install your code to get people to install your thing.

B

So like it's not like it's people, um sometimes you hear the rebuttal that oh, it's like the difficulty bomb is such an easy thing to handle or to change like if we go into literally. The one line of fake block number equals x is the only thing that changes when we're when we're changing for the r. So if someone goes into the code base and changes that one number and do then you'd think that would be kind of easy, but the hard part isn't making the code change.

B

The hard part is getting legitimacy around your code around your releasing the code change, because we, if you want to download geth, you go to the gift repository and that has a long uh like history of legitimacy among things for security among people uh like who is it that manages it and all those things and- and that's.

A

B

Every client, so they would basically sing another mine same with aragon like there are there's a. I don't know. Provenance is the right word, but it's there's the legitimacy. That's part of that repository that you can't fork. You have to make that on your own right. So someone who wants to release they need to figure out a way to get the change.

B

Then everyone to believe that the code is actually what they're saying it does and all the stuff when you're writing you're downloading a random exe from a random file from somewhere that you don't know what it does and no one has trusted before.

B

C

B

C

A very difficult.

B

Thing to do to coordinate a release around so that's kind of like four different reasons that the difficulty bomb.

B

Affecting things and reasons why people like it or people don't like it, let's see if I can summarize so I.

C

I guess yeah questions is right, no, it's because I'm looking at that so and I think I've seen the question actually just start to surface now I mean, I think, one of the kind of original reasons out of what you mentioned- and I mentioned earlier also was um to kind of push towards transitioning to proof of stake, and now that that actually does seem to be on the horizon somewhere.

C

Do you see a future for the difficulty bomb once we're on proof-of-stake, or do you think that it's just do you think it should be disabled will be disabled shouldn't, be disabled or anything else.

B

The so this would be totally my my own opinion, not the uh official.

A

Further quartet.

B

I've officially left the appearance foundation, so it's even further than it used to be. um I there's a there's a thread somewhere actually on the theory, magicians or e3, about like in theory, because that you're right. That was that's like the other reason that when proof of stake launch happens, the proof of work chain will slow down until nothing, and that will just happen on its own because of the difficulty.

B

So as far as should we do it or not, if I was them in my from my experience of trying to coordinate changes, I would leave it in or I would.

A

Put in a different way.

B

Of doing it, first of all, I would not do the ice age itself in the way that did it, but I would have something that then, is a a good accountability. Metric forcing function for devs to do stuff until redeemed at like the protocol, is ready for ossification, and it's like we're like yeah. This is solid, all the all the big things that we want to do with ethereum.

A

B

um And we don't need to keep doing these hard forks to keep them going, then I would say so I I would leave it in for that reason, just because it it gives a really good scheduling.

A

Oh yeah, I think I I followed some of the discussion around random and like how to manage the things, but that would be a separate topic altogether. So just uh to sum up, uh will this uh the the upgrade going on in middle london? Be the final eip?

A

Are you for like a upgrade, or we can expect one more.

B

uh It depends it depends on if we have a in december, if it's the merge, it's done if, if in december or it's a feature for work like what we have seen normally with ethereum, then we will probably do one more otherwise in december. Otherwise, in january, we'll have like 25 second block time right. That would be. That would be yeah, and I wanted to say one more thing, so they keeping it.

B

There are notable members of the core devs that are one and gone and have wanted it gone forever and other ones who like it and don't like it. So it's a it's. Definitely do your own, your own, your own research and figure out like there are.

A

Smart people on.

B

Both sides that uh have really good reasons, so it's kind of a matter of them figuring it out amongst themselves, and then the community also like anyone watching this. If you feel like there's a reason to have it or a reason not to have it the feedback on why it's really important.

A

So my next question is around the signal of ice age being noticeable. I remember you mentioned about like how people can um actually see like the higher the hashtag longer it can stay. Hidden, so was wondering. Is there any effect of market prices on the difficulty bomb or is just the hash rate that is uh imposing any changes or.

B

Making so it's it's a it's a long chain of things that affect other things, so you can, you can make the tie. There isn't like a this. The price affects hash rate, thus like a scientific. This is what will happen, but you can do some economic generalizations that make sense like if the price is going up, then the amount that miners or block producers are making is going up. So more people are going to want to mine.

B

So then the difficulty, if more people start mining, there's more competition. The difficulty goes up so usually you'll see hash rate following price, where.

A

If there's a price increase like.

B

B

Actually see you can see when the 2017.

B

And stuff- I guess not exactly, um but when, like miners turn off when it's not when it's not economical for them to mine miners turn on when it is, and that has to be uh so price then leads to hash rate changes. Password changes leads to difficulty changes. Difficulty changes directly affect the calculation.

C

I mean, I guess you could also, I seem to remember seeing people theorizing around the point that if the ice age does kick in also that it should have some kind of general economic effect, because the ether's emission rate will then begin to slow. As in, if we're expecting an inflation rate of whatever ether per day, then, if the ice age kicks in, that emission goes down, because we have less blocks per day that are generating ether and that theoretically, that should drive a bit of scarcity.

C

I I don't know if there's been any modeling done from when the I like those times that the ice age kicked in, it always struck me as something that probably should never have gotten to the point where it was economically significant. But do you know if there's been any work done? Also, just like the the effect of the ice age on the kind of like monetary policy? The theory of this it does.

B

So there's some there's actually some.

C

Interesting history.

B

About this, uh too, I'm glad you brought this up so first, the effect, I think, if you think of it as proportional to maybe I should just.

B

Make sure I find the.

B

Right, um so if you look at.

B

I need to go to the.

B

Time, how are we.

C

Doing on time,.

B

Or what this is, this is fun for me, uh so you can do it yeah, so we got this so so, if you, if so, if you think about it as a proportion to normal, this relative to normal is maybe like five percent and believing that, but you can think of so you're you're making not that much here, that's about twice, so the block time is twice as long so you're getting about half of the block reward. You would expect during that time.

B

um So I in theory, hash rate, probably would turn off, but in practice this function explodes so quickly, there's no way uh you can keep up and if hash rate starts turning off, it makes it worse. So uh if miners start turning off because they're not profitable the the strength of the arm goes down so then the difficulty bomb pushes are harder, and so you actually get longer and longer.

B

It's not an effective it, so it so in the in the the the argument that it would somehow get like an equilibrium or something or like it could go down enough, that it would matter that doesn't really hold water, but that it affects the economic policy and how much ethereum is minted definitely happens, and so the first two times when they did the difficulty bomb, uh like the four two addresses they paired it with the reductions to supply. So it's either.

B

It started out as five ether block ethereum did and then they went to three and then they went to two and when they, when they went from five to three, was the same time. They did the diff when they went from three to two at the same time, and istanbul was the first time I think or the one after that, uh when I started getting involved was the first time we had a only, I guess, maybe his mirror. Glacier was the first time we only had. We had a difficulty bomb.

B

Only eip- and there was like a discussion about- is that a problem because um there's some like uh the longer that ethereum has worked, the more issuance happens.

B

And so there's some I does that is there some kind of uh initial promise or economics something that was made in the beginning of the white paper or yellow paper or whatever that needs to be kept. um I don't think I I think it changed. It started as kind of that attitude and then changed to a. This is just how we manage release cycles kind of stuff.

B

A

B

The controversy, as far as whenever you're touching either issues in any way, um yeah that does that answer.

C

Your question: um I believe that it does.

A

So my last question is about the uh test net platform so uh like we are expecting london upgrades soon and uh we are expecting. uh The proposal is going on godly chestnut on june 30th.

A

I understand this difficulty bomb is never activated on uh girly network, uh the test network or any kind of poa chains for people who are learning it for the first time. Could you briefly talk about what a poa chain is and why the difficulty bomb proposal or of the ice age is not activated on on a poi chain for testing purposes?.

B

Yeah, so the poa stands for proof of proof of authority and I'm not the I'm, not an authority on proof of authority. So as far as my understanding goes that in both um I will start with robston, which is a test net that has proof of work. It's the closest thing that what happens on main net they're miners who are mining in roston and finding blocks the difficulty adjustment uh algorithm like the time targeting block time.

B

Targeting mechanism exists on robs uh when you go to any of the other ones: gorly kovan, rinkaby, they're, all poa networks, and they don't have miners. So instead they have a group of authorities who are assigners. Who say these are the next blocks and they they don't have a timing mechanism at all. They just work together and they say they they pick literally every 13 seconds or something like that.

B

They make they they it's much closer to how in proof of stake the the block, the staking block producers have a block time that they stick to and their their job is to coordinate and make and keep that time um versus in proof of work world.

B

There is miners and hash rate is kind of free flowing, and you need to adapt the system to target time. So, though, so the the the the difficulty bomb never goes live on, coven or gorilla. It doesn't even matter because there's no block targeting mech timing mechanism to even target a difficulty bomb form.

A

That is good, so we are very close to the our uh would like to wrap it up uh on behalf of ethereum characteristics. Thank you so much james uh for taking out time and uh talking about one of the most popular terms that we have been hearing for for a very long time. Yet very few people understand how it actually works, and you have explained with the script at the program.

A

I hope that is going to be very helpful for new people who have just started learning about the ethereum and the proof of work.

A

I can say with the pride that I am one of those lucky ones who got to work with you closely, especially when you were working on istanbul, berlin and a bit of london too, and I had a wonderful experience. Thank you so much for your contribution and, of course, for this proposal.

B

Yeah, it was awesome.

A

On this note, uh for all our youtube viewers, thank you for watching this episode. uh If you like, uh hit like share it with community members, you may find guests twitter handle and the resources that were discussed in this meeting in the description box below, as per the analytics of ethereum cathedral's youtube channel over 85 of the viewers who are watching. This series are not a subscriber yet so please, please, please do subscribe to ethereumchaturance youtube for meetings and all educational resources. Those are coming up.

A

Next week we are going to have a special episode on a brief history of ethereum's future with a special guest, benjamin addington, so stay tuned. Keep watching keep sharing your love with the dmcat leaders. See you in the next episode have a good one.