📺 London Hardfork Livestream | EIP 1559 Goes Live!

yeah which is more accurate though so this one says 28 minutes and we got 34 over here either yeah that one keeps being pushed back constantly there's another one watch the burn says and ethernets 37 minutes yeah i think it's going to be that one right there yeah you think it's going to be watch the burn i mean they either scan and watch the burn agree i don't know what algorithm they're going to consider

welcome justin hey justin yeah hey um [Laughter] we'll see we'll see the feedback is greater than uh than the issuance if it

is then yeah i think you know the analogy is kind of you know ethereum is uh you know standing on its own feet it's financially independent kind of the the income that it has is greater than the expenses um that kind of analogy but we will see um i mean it's looking very hard right everyone has to pay their taxes [Laughter] so yeah my expectation is that the the

base fee is going to match roughly the the slow price on gas gas now the org so maybe something like 35 grey in which case you know we'd be at mach 3 so we'd be burning three times more than the proof of uh issuance of work okay not on not on proof-of-stake on proof-of-stake transaction spam just lined up so we can get that that basically over 110. at least we could just start burning each just for a little bit

the eighth pump is on as well oh is it yeah but i don't trust these price movements uh right now i mean keep in mind anthony looks at the one minute chart so yeah 26 30. i'm selling the news right now literally staring at the one-minute chat oh my god it's pumping permanent damage to those receptors this is it guys next stop 10k hey we're always going to 10k it's

always in the in ethereum's always going to 10k it's just a matter of the detours between now and then mm-hmm yeah and hey hey dave i was gonna make a public service announcement i assume we've like formally kicked off are we doing it yeah am i online yeah we are currently live streaming we are live we're live david hey everyone yeah david you wanna kick us out yeah kick us off david do something formal we're live i think it's funny but not london london's not live london is not live

25 minutes yeah say what you want dc yeah i was gonna say um because i just uh tweeted at metamask and i was just asking if they've already updated their wallet to reflect the base fee functionality and so it sounds like uh my crypto actually responded and said that um the the functionality is is is available if you want to build it yourself if you wanna do a custom build locally but otherwise once once there's a little bit of run time with one five five nine they'll roll it out more broadly so just stay

tuned for that in the meantime it's important to remember that eip1559 is backwards compatible with all all existing wallet software so users don't necessarily have to worry about that kind of compatibility there may just be some situations where in the interim you're paying a little bit more for gas until you get a wallet that supports the base fee how about metamask dc are so is metamask going to roll this out pretty quickly that's right that's what i was talking about it was metamask uh yeah my crypto just responded to my tweet

and i assumed that they know so that's that's the situation on metamask have you guys seen any uh early looks of what the metamask interface is going to look like post eip1559 yeah i saw it yeah i saw two it's pretty similar there's like just three boxes still but it's a little different i forget where mask has a has a website for it but uh i'll get it one sec um i can put it in the chat if ryan wants to bring it off yeah i'll bring it

up and it shows like a few screenshots there i think or there was a video of it there we go wow [Laughter] the video is really good um it show it shows like exactly what it looks like oh nice wow that's coming on this mask

uh-huh sorry it's good that they've got dedicated section to it for users yeah this is great you can probably skip ahead to the um the device if you just yeah it's definitely gonna pop up now that's yeah there you go kind of like looks like that oh man so much better hi ads i feel like this is actually s main gas intuitive now total yeah if you keep going it will show like

the advanced settings okay so this is just the basic setting right there oh god that's so nice look at that yeah and then it's been dropped down here i've been dreaming of that literally i'm gonna miss having gas now open on one tab then entering it in the transaction doesn't confirm so i speed it up again i'm gonna miss that this whole entire like two year process of waiting for eip one five five nine just allows me to uninstall one browser extension finally

it always jumps up right after you send too because metamask is defaulting to that price and then everybody goes one above it so you send it and you're like oh man then you gotta resend guaranteed every time and that's like the bidding dynamic that you know like hasn't been adequately modeled and made we may see some nice improvements just in the turn in way in terms of how gas prices move because you don't have that human psychology kicking in where it's like oh i need to bid higher than is being

suggested so i can get my transaction so i think it'll be pretty interesting to watch oh metamask is telling all the plebs to use 35 i'll use 36.2 yep exactly yeah i think this is i think it's really good that they put together a video like this and a whole section on it as well yeah good job nice yeah but i mean the whole point is that mostly users won't have to touch the advanced options anymore like they have they currently do like yeah so

i'm not if i'm not mistaken watchtheburn.com plus ultrasound.money plus etherscan block countdown are all pointing to the exact same minute okay cool now finally have consensus it's going to be at 20 20 minutes 20ish 21 minutes yeah brian try um refreshing ultra sound up menu hey hey oh yeah there you go 120 bucks yep we have consensus

question do you know who's behind ultrasound.money does anyone i don't know i don't know well i i i was on the banker's podcast at the very end saying that i bought the domain ultrasound.money and i was looking for people to to go help build the website so what happened is that about 10 people from the community came forward you know many of which were just volunteering their time and now we have kind of this decentralized team of people just working on this on this website

uh which is super super super cool only in crypto can get full time attention for free yes and yeah the team is also helping with the twitter accounts so the ultrasound money twitter account there's a there's a team of of people behind that as well i love the website it's clean as hell the website is super clean every few days i get like dms by the ultrasound money uh twitter like which is like bat

signal [Laughter] or just like who's in my dms oh it's ultrasound money account cool that's all they said [Laughter] oh man it's a good me like i love it and it triggers people just the right amount too which makes it a very good meme i love how just non-ethereum people think it's the stupidest meme ever because they will let you know how stupid they think it is and i'm just like ooty ooty tweeted out yesterday the ultrasound money meme is the cringiest

meme out of the eath head camp i've ever seen and i commented right below it yet here we are he keeps on talking about it you guys might are we all there the ones that are live on the call the ultrasound on the bottom ryan oh down the bottom oh yes so i i think anybody missing italics missing i'm definitely missing [Laughter]

so what does it take to get on this list you have to wear the bad signal right you have to wear the bat signal and have a lot of fun your profile yeah it's ranked by follow account so the way it's ranked i believe is that the the first few are ranked by follow account and then the last few are kind of chosen at random from the community so every once in a while if you're just a community member you should see your proof out there uh but that's a cool way to do it

oh hey vitalik you're here for one second okay so eric eric you said is it too early to drink mariano i think you concurred it's not what do you got i got a beer you actually actually have a beer i'm with my i'm with my mata right now it's 7 15 a.m but uh i'll i'll join with a beer

all right yeah it's 7 15 and i have a beer but i would have a beer but i've been roped into doing a live stream podcast with bank list later this afternoon i want to keep a clear head well i'm going to have a beer though you know while while we're hanging out i would be interested to get the perspective of the eip1559 authors and how they how they feel on this day one first metallic

how do i how do i feel on this day um let's see yeah again definitely happy that this thing that's been worked on for like literally almost four years now as uh finally turning into reality yeah yeah hmm okay sure let's see when when did that or if original eip 1559 paper cut man let me to check resource i have a long time ago i think it was like 2019 2018

wow let me check draft position paper on resource pricing um this was august 18 okay yeah so it's uh and okay so it was not something we've been working hard on for four years it's something that we've been working on for three years 2018 was that e3 search post on first price auctions yeah yeah it was and then i was trying to remember when that um economics and computation conference that carl and i went to that ended up inspiring the whole thing

um it was at that event i think it was like you know as you see i guess 2018 that uh we were talking with some of the local economists and they mentioned how uh first price auctions are evil and yeah um and then some of them um suggested second price auction as an alternative or was that good if maybe it was both um but then second price auctions of course has to have the problem that they're not collusion resistance and in uh we shang

wu's econ language he could use the word credible auction but like in that case credible credible mechanisms they're there there's low income but credible in this case just means that like immune is a collusion between mechanism participants and the mechanism operator which of course we want um second price auction simply means like a normal auction except not the top price gets selected but the second highest price gets selected correct so the the one who bid the top price wins but but they only pay the price of

the second so vitalik what was the ethereum community's reaction to this research post well as you can see it's 18 hearts but i think that the problem is that there's a big difference between um being fascinated by something theoretically and being willing to throw in the energy to actually get it out there in reality yeah we kind of then we we brought it to

the community like i think it was like march of 2019 so it's even been two and a half years since i mean as in like try to put it in a medium post and explain the benefits a little more and then it seemed like from there it started to pick up some steam yeah there's a lot of the usual suspects there uh commenting i saw philippe yeah mika dan robinson well it sat for a while before um you know eric wrote up the eip right eric

yeah honestly it was just i was frustrated with the user experience of every day and like realizing that there's no way we're going to hit mass adoption really with the way that fees were being handled and i i don't know i would always just browse youth research and somehow i ran into i mean it wasn't even this discussion it was the paper that you were referencing vitalik and i kind of reached out to you and said you know hey why isn't this picking up some steam and then yeah we just kind of i mean i believe at the time

just vitalik and i co-wrote the original eip and then after that it really picked up steam that's when a lot of different people jumped in and i can't really take much credit at all when you ask yourself why isn't this picking up steam that's a sign for action yeah totally yeah totally i was like this seems so obvious let's do something about it and then yeah all the other everyone else that got involved jumped in and kind of took it from there which was awesome so it's funny it's oh sorry go ahead this is the original eip right here that

we're looking at that's the final but yeah the original draft was way more than that it's funny though right like there's all these great ideas out there and they start off very technical and usually like on each research and it really needs to go through this like interesting process of translating that to the community and like why it's beneficial um like the whole i mean it makes sense why this took three years or three and a half years or whatever it did if you really think about it

yeah i mean there's work sorry i was just gonna say the bulk of the work got done like the last year as well like since that bitcoin um funding that happened in like june of 2020 that was like a strong signal to like everyone hey we want this we want this now and it still took a year from there so yeah this is a really big and complex change i mean the amount of work that's gone into it is just incredible yeah that's what the multi-sig that kind of got behind it in the community pushing for it was pretty badass yep and everyone i'm making a whole

bunch of get going donations and like directing half the matching [ __ ] at one time yeah i mean i was just reflecting on how much effort has gone into this all the way from you know um when it was proposed to when it was first being analyzed by rick dudley and his team through to kind of you know even the analysis done by third parties like tim rufgarden was also important in the advancement of this and i think more recently the work that tim baco and you know james hancock and others did to

kind of advance this was was significant and even when hudson was involved with that process as well so i think it's been an incredible effort and and a long time coming so i'm excited there was a while there where i was worried this wasn't going to happen i was worried that no one was really going to pick it up and coordinate it and run with it but this is an example of you know sort of the chaotic decentralized nature of the ethereum community actually like um you're pulling this out and making it work making it happen so it uh you know

pretty exciting ryan you got to remember we were focused on our like most favorite eip prog pal so we had other priorities oh god let's shut down the stream after that got brought up spend yeah surprise everyone we're talking prague we're actually overdue for a frog pound debate guys sorry do you always start talking about prague after one beer [Laughter]

so when we switch to starks and it turns out that for a stark you actually need to do a little bit of proof of work to grind some merkle root if you want to like reduce the number of challenges to cut the the data size of us dark down just to troll everyone can we use proc power as the proof of work algorithm no i'm done vitalik can you give us the intuition on traditionally why a second price auction is better than a first price auction um sure maybe so the economic arguments

for why second price auctions are optimal is uh basically that like imagine that you have some person and what are we optimizing um we're optimizing for basically a naive algorithm uh being the op being the optimal algorithm for each participant um and so i'll go i'll explain the property that second price auctions have that's really nice um essentially yeah the let's say you value like something

that's being auctioned like let's say we're auctioning and i know um yeah um which you pay for with metamask um so you uh you value this at ft at ten dollars and uh the question is like what value do you wants to bid right um so if you bid less than ten dollars then there exists the possibility that let's say you bid x for x is less than 10 then

someone else um who values who values it somewhere some somewhere between x and 10 there's the possibility that they would bid some amount in between x and 10 and so like in that particular case or basically there's uh the possibility that you and uh you end up not actually getting the item uh basically because of this other person ended up getting in with a lower bid now in the case where there is someone who or if you bid greater than 10 then there is the possibility that

you win and you end up having to pay more than you um or a larger amount than what you actually want right and if you bid 10 um then like basically all right so here's the thing that happens the thing that happens is that your bid and like the exact value of your bid um it does not influence how much you pay it only influences whether or not you win right that that's actually that's a really important criterion and when do you wants to win well he wants to win when the amounts

that you're going to pay is actually yeah is less than the amount at which you value this nft of the nft of a doge um so if you bid 10 which is equal to your valuation then if the the highest bid among everyone else is less than 10 then you win which and you have to pay an amount less than ten uh but then if uh someone else is uh willing to pay an email greater than ten then you don't win and in that particular case he wants to not win

because had he won you would have had to pay an amount greater than 10. right so basically yeah like the fact that your the amount that you pay does not depend on your bid the only thing that your bid determines is like basically whether you get the item or you don't get the item and you only get the item in those worlds where you have to pay an amount less than you value anyway um like basically it means that the simple algorithm of like just bid the amount at which you value the item is just going to give you the best results right so it's a very simple mechanism in order to

choose your correct bids you don't have to um know anything about anyone else's preferences you don't have to go second-guessing other people there isn't like some risk that you get that you guessed wrong and there's some inefficient outcome um so and it doesn't help if you have if you have other participants like if you are 10 people and you're trying like you can't actually yeah you can a little bit i'm so this is the the problem like basically

if you control the top two uh the participants with the top two evaluations um and then your optimal strategy is not to bid honestly your optimal strategy is to bid what the third valuation plus one cents this is assuming you have like assuming information oh yeah yeah like but if you have like i think even the slightest amount of information about the probability distribution of the third then like i think that ends up like distorting and giving you at least a little bit of extra profit so if i'm sell but if i'm selling the nft i prefer

first-person auction correct no you'd i'm actually no so there's these theorems that are called revenue equivalence theorems that basically show how um for a wide class of possible auctions your expected revenue is uh going to be exactly the same i know this is really weird and spooky and i actually haven't really yet examined the proofs of this but it's uh true um so yeah now if you want to survive uh collusion then like well

basically well one thing you could do is instead of a second price election you know you have a case price auction and case price auctions are resistant up to k minus one colluding participants um but in general like yeah that you hit you hit upon something really important which is that auctions are not uh like infinitely collusion resistance and that's actually why eip1559 is not a second price auction it's a totally different kind of mechanism where that's just a fixed fee

people compare it to a second place auction though right um they i guess they do though compare those comparisons are slightly misleading like i think the the correct statements to make would be that it avoids the inefficiencies of first price auctions the same way that second price auctions avoid the inefficiency of first price auctions so like the key property right in the second price election is like the amount that you bid does not determine what you pay it just determines if you pay and in eip 1559 that's true as well right the

max fee only determines whether you get in it doesn't determine what you pay um dc when vitalik was talking about if you want to buy this nft for ten dollars he was talking about your crypto punk hoodie yeah no one's going to be buying that one anytime soon so good luck although i do want to you know i think there's a common misconception around some of the nft stuff and people are like oh ethereum's too expensive so

that's why these mfts cost a lot but realistically what has happened with some of these nft mechanisms is they're using a gas auction as the primary distribution method which is sub-optimal in a lot of ways and i think you're going to see more and more projects move away from that but basically rather than selecting the buyers through other some other mechanism like say a dutch auction or something else they were basically letting people kind of bid up gas and whoever gets the transaction in first wins um so i you know i i've never liked that distribution model because it

favors people who who know how to use the network a little bit better or might be able to set up bots to get transactions in faster and things like that so i think that's going to that model will slowly kind of phase out as we move forward because you don't have so let's look at time windows that are only kind of there's a frenzy going on does this change did the mechanisms change at all really or does it revert to the public first price um it reverts to a public first price when there is a frenzy going on well um it reverts to a public first price for

the subset of users that really care about being included quickly right if you're if you're okay with waiting two minutes in the extreme cases then you can just continue to kind of live in the normal world and you just have to wait a bit more does it reduce the chance of frenzies because there's going to be less of like the the psychological model of i need to click the button regardless like just for end users yeah that that's definitely one reason why it reduces frenzies another important reason why it reduces frenzies is that

transactions that get included quickly are less likely to fail and so you're less likely to need to resend um a third reason why it reduces frenzies is just the fact that if there exists a burst of uh many transactions like like end times block size transactions instead of it needing and blocks to clear at the inside and over two blocks to clear so if it is frenzy resistant some in some cases would that actually lead to slightly less gas uh less fees for the end users

interesting i i guess i mean unless of course like induced demands takes over and like we just get even more users we are we are uh less than five minutes out i think depending on which clock you are monitoring so on etherscan we are close to four minutes just under four minutes on ultrasound.money we are at about five minutes and 20 seconds so this is fast approaching um hudson since you've lived through many of these hard forks sir how do we know if this is successful like

what like what happens when the hard work happens that's a funny question first um first hi trent let's all say hi to trent hi trent hey hey i'm i'm good to just sit back and listen to all you guys talk about nonsense so keep going look at those flowing locks over there wow so basically right now at least so i used to run the core developer meetings every other week and during hard forks we would

um before there was like the proliferation of all these live streams some of the time there would be just a private zoom call with some of the core devs um or if something went wrong there would be a zoom call with the core devs that still happens um so we'll know if something went wrong because there'll be like a few people who know where to go for the information like me trent danny and tim and just be leaking stuff to twitter slowly over time that's generally how it goes down so if some of you guys suddenly disappear from this call then that's probably bad news

is that what you're saying hudson yes pretty much so yeah so but for real though like it's pretty organic and usually during times of emergencies that we just kind of the cordov just kind of come together and then just troubleshoot and fix it and it's pretty straightforward there's no one usually panicking or anything like that it's really professional how do the core devs typically like feel before a fork is is it just what a question like look man i i'm just looking at this clock uh this countdown and i'm you know

get a little nervous to be honest you know things could you never know uh but i'm not a core dev how do they feel about these things danny vitalic what do you all think or trent or anyone else i don't know because i'm not a coordinative but i think they're pretty nervous medium they usually go well everyone's definitely watching because that might require action uh but you know i don't think anyone's well all right guys we have a six blocks left so i'm going to go ahead and cut to

sponsors real quick uh here we go i think ultrasound is a bit slow there's like four blocks or something now yeah we're almost there oh man this is geez geez it's happening it's happening guys four blocks remaining anthony how does the 30 second chart look is are we pumping we have since anthony last commented the eath pump is on we have dumped

yes actually as as for a convention three points bitcoin it's ruining our party that bloody penalty two blocks two blocks remaining i wonder who's going to get the first burn transaction all right all right here we go one block remaining guys it's happening eip 1559 history 1559 excuse me one five five nine five nine one fifty five nine what is one thousand five hundred and fifty nine

there we go it's done as expected so i'm expecting it to be 100 gas use for the next roughly five minutes and the base feed what's that the base gas price well but we might see the gas go down if miners didn't set uh reset their target so that'll be interesting to watch so we started a one way