Unchained - The Chopping Block: How Jito Helped Put Solana on the Map Again - Ep. 588

Episode Date: December 29, 2023

Welcome to The Chopping Block – where crypto insiders Haseeb Qureshi, Tom Schmidt, and Tarun Chitra chop it up about the latest news. This week, they are joined by Lucas Bruder, co-founder and CEO of Jito Labs, an infrastructure provider that mitigates the impact of maximum extractable value (MEV) on Solana — specifically spam and wasted block space. He and the gang discuss how the Jito airdrop this month helped rejuvenate Solana; how Jito differs from Flashbots, its counterpart on Ethereum; Solana’s potential scalability; and how Solana believers have been vindicated by the developments of the last few months.  Listen to the episode on Apple Podcasts, Spotify, Overcast, Podcast Addict, Pocket Casts, Pandora, Castbox, Google Podcasts, TuneIn, Amazon Music, or on your favorite podcast platform. Show highlights:  Why Jito is a crucial development in the Solana ecosystem and its potential impact on network performance What sets Jito apart from its counterpart on Ethereum, Flashbots, in terms of functionality and benefits Whether Solana's speed uniquely influences Jito's effectiveness compared to Flashbots Lucas' explanation of maximum extractable value (MEV) and how it operates within the Solana network, providing insight into its complexities and advantages Whether Solana's infrastructure could sustain block production if the AWS integration encounters disruptions How the Jito airdrop may have rejuvenated the Solana network, potentially marking an end to its "ice age" What recent surges in on-chain activity in Solana indicate about the network's health and user engagement Where MEV fees in Solana are directed and who benefits from this value accrual within the ecosystem Whether Solana dapps are evolving their architectures to mitigate MEV opportunities and enhance network security How Tarun envisions the design of Solana's fee market for optimal efficiency and fairness Whether Solana has the inherent capacity for infinite scalability and what potential bottlenecks might impede this growth Hosts Haseeb Qureshi, managing partner at Dragonfly  Tom Schmidt, general partner at Dragonfly  Tarun Chitra, managing partner at Robot Ventures Guest: Lucas Bruder, CEO of Jito Labs Links MEV: Previous coverage of Unchained on MEV:  Unchained: Why Is Ethereum Trying to Maximize Value From Users? Two Sides Debate The Chopping Block: Why the Once-Taboo MEV Is Now a Core Part of Ethereum CoinDesk: What Is MEV, aka Maximal Extractable Value? Jito:  Unchained: Jito, Solana-Based Liquid Staking Protocol, Airdrops Governance Token Solana Validator 101: Transaction Processing Jito Block Engine Expands Access to All Solana MEV Traders Solana’s recent activity and bull run: Unchained:  Anatoly Yakovenko on Solana’s Astounding Recovery and Its Future Plans Solana Takes the Lead in Daily Stablecoin Transfer Volume Solana Surges to $84, Flips Ethereum in Weekly DEX Volume Paxos Bringing Stablecoin Issuance on Solana, Diversifying Beyond Ethereum Solana’s model: Solana Proof of History: How Solana brings time to crypto Learn more about your ad choices. Visit megaphone.fm/adchoices

Yeah, there's been a ton of new activity on Solana. There's a lot of new capital. So if you look at the bridging patterns and things like that, there's a ton of new activity coming into the space. There's a ton of super high quality projects that have been grinding basically since like the middle of last year. And, you know, I think a lot, what happened with all the FTX stuff that really suck for everyone, but there's a lot of teams that kind of just kept grinding and building.

Not a dividend. It's a tale of two pawn. Now, your losses are on someone else's balance. Generally speaking, air drops are kind of pointless anyways. Unnamed trading firms who are very involved. D5.Eat is the ultimate pump. DFI protocols are the antidote to this problem. Hello, everybody. Welcome with the chopping block.

Every couple weeks, the four of us get together and give the industry insider's perspective on the crypto topics of the day. So quick intros for us who've got Tom, the DFI Maven, and Master of Mims. GM. Next, we've got to Rooters. the gig of brain and grand puba at gauntlet. Aloha, and it does look like I'm almost in Hawaii today, although I'm not in Hawaii. Great.

And we've got Lucas, a special guest, the CEO and Solana Savant at Gito Lab. Nice. I like that. Thanks for having me on. I'm excited to be here. And I'm Maseeb, the head hype man at Dragonfly. We are early stage investors in crypto, but I want to caveat that nothing we say here is investment advice, legal advice, or even life advice.

Please see Chopping Block. at XYZ for more disclosures. So we've been off for a couple weeks. We're coming back in the swing of things now at the end of the year. We just had our end of year review, but it turns out it was a little bit premature because the end of the year was not quite the end of the year. There was a lot more year to go. And it turned out that in the last couple of weeks, one of the most notable things that

happened is there's been a crazy bull run in Alt-L-1s and most notably in Solana. Salana, I think when we were talking, when we were doing the end of year review, Solana was at $60,70. It is now over $100. The entire network is over $50 billion in FDV. I think it's like 60 billion FTV. Market Cap is like in the 50-ish billion. And it's basically, it's near, I think it's now the number four asset on coin market cap,

cementing itself now as being the predominant non-Etherium smart contract platform. So as the as the guest for the show, we thought we'd bring on Lucas. So Lucas is known on Twitter as Bufilu. Is that he you pronounce it? Bufilu. Buffalo. Buffalo. So Lucas is known as Buffalo.

Wait, wait, wait. Can we, can we get the etymology? Like what, what's the story behind the name? Oh, probably not appropriate for this podcast. Maybe if, if someone takes message drinks or something, someone can get the full story. Okay. I have a lot of trouble imagining now how that story goes, but okay. Good to know.

So one of the reasons why we brought on Lucas is that in his Twitter alter ego, Bufelieu, he's one of these guys who I'd call like the Solana reply guys, that anytime anybody is talking about Solana, him or Mert or Anatoli will show up and be like, no, no, no, no, you have no idea what the hell you're talking about. Here's the lowdown of how Solana really works. unlike some of the other characters out there, Lucas really knows his shit. So he's the founder of Gito. And so Lucas, why don't you explain for our audience really quickly, what the hell is Gito and why does it matter in the context of Solana? Yeah, Gito is a Solana M. M.E.V. infrastructure provider and basically trying to mitigate the negative impact of M.EV on Solana, specifically spam and wasted block space.

and so, you know, I think a lot of people or a lot of chains have this issue that are high throughput, low fee chains where you just have a lot of arbitrage. You know, if there's like little microarbitrages, then it's going to take up a lot of block space. So Gita tries to build infrastructure to efficiently extract them, even, and try to kind of save some of the block space for users and some of the good bots on the network that are trying to make markets. and things like that. So a while back we had on Phil Dyan, who's one of the co-founders of Flashbots, which is the Ethereum version of what you guys are building. How would you compare what you guys are building on Solana to Flashbots, which is the Ethereum counterpart?

Yeah. So Salana, the main difference is speed on Solana. I think that's pretty obvious. Ethereum, you know, there's 12-second block time, 12 or 13 seconds, and the throughput's way lower. Solana, it's 400 millisecond block times. The leader is bouncing across the world all over the place. So the systems that we run are very high performance, very latency sensitive.

You know, multiple locations around the world. I think I don't know if people on PBS do that. So that's kind of the main difference from like a super like macro level. When you like start to dig into the weeds a little bit, the architecture looks a little different. So on PBS, they're building entire blocks on Gito block engine. We're building pieces of blocks, essentially bundles. So the unit that we're operating on is a little different than PBS. And yeah, there's like the transaction flows a little different, which we can dive into.

So I want to be mindful. A lot of our audience is non-technical. And a lot of our audience probably, you know, they've heard these words before, maybe one place or another, but maybe a little bit hazy in their minds from the last time they heard it's kind of alphabet soup of different MEV related things. So just very briefly, I would love if you could quickly explain, just for our audience, one more time, what is MEV? How does it figure into Solana?

And why does the fact that Solana is fast matter for making this problem different or harder or easier? Yeah. So MEV stands for maximum extractable value. and essentially it's the value that block producers can extract or like, you know, make by determining the ordering of transactions in their block. So, you know, I guess for maybe for people in the audience, you know, there's some networks with validators, some with minors. You can basically just think of them as the person who's collecting all these transactions, figuring out the ordering of them, proposing them to the rest of the network, the rest of the network. the rest of the network kind of takes these blocks, looks at them, it replays all the transactions,

they come to consensus on the state. And when you're a block producer, you are the one who's determining the ordering of transactions. So MED is basically, you can think of it like a building more efficient blocks and more profitable blocks. So some networks, like Ethereum before FlashBots, It was a, I think it was a priority gas and then time auction.

So you have this Mampool, you have all these transactions. At the very top of the block, you have the transactions that pay the highest gas and then kind of lower as it goes. Salana, it's a little different. There's a, Solana's a continuous block stream. So the leaders are processing these transactions live and sending them out to the rest of the network. everything else is replaying it live. And also, there's no, there's no MMP pool built into the Solana protocol like there is Ethereum.

So transactions are going straight to the leader. And the leader is kind of prioritizing them. So with the introduction of MEV, you're essentially the validators and miners and things like that and the network that are producing these blocks are somewhat outsourcing the construction of either the entire block or pieces of it to do. different parties that can run more optimized algorithms to make more money for validators and stakers. So just to make sure I understand the details there.

So in Ethereum, nothing actually gets propagated to everybody. Nothing actually gets confirmed as state until it's packaged up in a complete block. And those blocks happen every 12 seconds, more or less on the dot now with proof of stake. In Solana, that's not how it works. In Solana, it's almost like streaming block production where it's sort of, like we call them blocks for the purposes of consensus, but really, you know, transactions as they're getting sent to the leader, and the leader is the leader for like, you know, some number of epics or whether. I don't know how long an epic is. I don't know, maybe let's say some number of seconds,

a lot of blocks when the block time is like 400 milliseconds. Pieces of blocks, or I guess what are called shreds in Solana, they get confirmed and streamed out, even though the entire block is not done yet. It's like, okay, well, the first five people in this block are these guys, and then the next five are these guys, and the next five are these guys. And so in a system like that, where it's basically streaming as opposed to these all-in-one block containers, doesn't that make M-EV just totally different in a system like that?

Yeah, it's super hard. It's much different than Ethereum. So every leader on Solana, every slot, which is basically like a window that you can produce the block in, it's 400 milliseconds. Leaders have, multiples of four consecutive slots.

So a leader will have four, 400 milliseconds slots or 1.6 seconds. And then it all, you know, that leader is in Germany. The next second, next 1.6 seconds will be in Tokyo. And you could be in California. It's kind of just like bouncing all across the world. And then also you have the streaming part, which you're mentioning. It's kind of like the video stream that we're talking on right now. Where Ethereum, it's kind of like choppy and, you know, it's,

you're sending the entire block at once, the Solana leaders are actually streaming the blocks out. So it's basically like a fire hose of shreds that are going out, which are essentially reconstructed on the fly by every other validator and RPC on the network. They're replaying it and so on. And so the MIV game is definitely way different. I guess one of the main differences is that latency matters a lot more, where the blocks are streaming out so you can actually,

you're replaying stuff live and you can you can actually the leader can process a transaction send out the shreds your node can replay them and you can fire off a transaction and land later within the same slot so it's um you know compared to ethereum where you kind of have this like 12 second pause and you are basically trying you know you'll kind of say this on like pbs where the bids kind of ramp up closer to the end where people are kind of waiting until the last second to send their trades. Maybe I'm getting too far in the weeds here, but it's basically the latency matters a lot more on Solana. I'm curious actually about, so I think there's a lot obviously about the consensus and block building process that's definitely between different between Ethereum and

Solana. I'm curious in particular about the difference is in architecture that you've chosen between Gito and FlashBots as sort of a brief recap and summary. FlashBots makes a software that boost. People run relays, which are sort of this matching auction exchange that match people who want to buy Blockspace with people who are producing Blockbase or validators. And there's much of these different relays, Flashpots runs one, blocks right, ones run a couple different SaaS companies run them. But they sort of make this software and people use it. You think it's a very sort of general unopinionated piece of software. Geo, I would say, is much more sort of appally in that you guys have your own liquid staking token.

You, I think, went open source somewhat more recently. I think you guys have your own relay. And I'm not saying that's like most of the volumes going to your own relay. And it feels like more of a clean, vertically integrated sort of polished experience. How do you sort of think about those different design choices between with flashbots and and GEDA. Yeah. So I think Solana, MEEVE is kind of in the first ending.

And I think our infrastructures may be more similar to MvGath. So before proposer builder separation, all this stuff, flashbots created a fork of the Go-Etherium client called MevGath. And they were kind of the only auction year. Basically, I don't think there was, they were kind of the block builder and the relayer. So basically they were aggregating order flow and doing the simulation and they might have been sending bundles at first and then move to full blocks as kind of a stepping stone to BBS. I think Gito is kind of in the same state right now.

So Gito created the Gito Salon of Valadair client and there's the Gito block engine and it's it's kind of similar to like the early days of MEV on Ethereum. So Gito runs the auctions. We have an in-house simulator that our engineering team is built that kind of does a bunch of crazy simulation to filter through all the bundles and whatnot. And then the unit of operation for us is bundles. So we're kind of sending bundles to the Gita Salana leaders when they're up similar to like the early days. I think that will evolve over time. There's a lot of protocol changes that are coming to Salana. you know there's like fire dancers coming

there's a lot of talk on like bankless leaders where basically the leaders aren't actually executing the transactions they're kind of just like packing blocks there's like asynchronous execution there's all this crazy stuff that's coming and you know it's like do we there's a lot of questions in my mind like what what are we building for and like a multi-year time horizon

I don't know if it'll look exactly like PBS and like swab and all that that stuff. I don't think that architecture makes sense on Solana, but it's definitely not the end state in the current form. So right now, what is the percentage of validators or leaders who are running the GTO client? Yeah, so there's currently 47% of the Salana stake running the Gito client. And this is like large institutions like Coinbase figment P to P to in like staking facilities to like, you know, kids and like small businesses that are running it, super easy to run. Right.

47% of stake, 300 validators, or almost 400 validators. So a natural question, you know, if you look at the trajectory of flashbots, flashbots started kind of running the only auction by themselves and standing in the middle of Ethereum infrastructure. And they were criticized pretty roundly as being like, hey, why do we have one company that's kind of sitting as an intermediary to this, this part of. call. And eventually they moved away and kind of decentralized the relays such that FlashBusters only running one of many relays. Now, I can imagine that perhaps someday in the

future you guys proceed the same thing, maybe there are going to be multiple different relays or auctioneers sitting besides yourselves. That said, given the latency differences between Ethereum and Solana, one of the common critiques that Solana gets is because of the relentless latency sensitivity of Solana because of the fact that not only are there these 400 milliseconds slots, but you have these streaming of blocks and the fact that, you know, the leaders are moving from jurisdiction and jurisdiction really quickly. There's so much of a centralizing force in whoever is the fastest, whoever has the best infra, whoever has the best software to run this auction is going to get overwhelmingly

an advantage over other people who aren't running at the same scale with the same level of professionalism. So do you see that as a concern that even if you do try to nominally decentralize and say, okay, well, anybody can run an auction. Here's like the format, RPC format, whatever, to do it, that you sort of end up in the situation where one party, maybe yourselves, just de facto are the auctioneers for all Salana block space. Potentially, I think it'll be interesting to see how the Salana protocol kind of evolves and like all the things I was mentioning earlier. I think that the way that our auction is currently designed,

you don't really have those forces yet where we are, the Gito Block Engine is kind of running. It's kind of like somewhat discretizing these auctions into what today is currently 200 millisecond auction windows. So that's like enough for basically someone in like, someone in like New York could potentially be competitive in an auction in like Tokyo or something.

And then we have services like shred stream, which basically provides low latency shreds to anyone that signs up from all these leaders that are running this code and kind of like democratizing access to the shreds and low latency feeds. So in the current sense, it's not necessarily that much of a problem. I think it'll be interesting to see how the protocol evolves. And yeah, I mean, I think it is worth something keeping eye out. I think some of the sexes and like or centralized exchanges and like Tokyo and other places. It can certainly, they certainly have the potential to be like large forces of

gravity where there's like a lot of sex-dex-arb and, you know, those people are making a lot of money and, you know, state kind of starts to congregate around there. I think, you know, G-Dos four block engines hoping to spin up six, seven, eight, and like Q1 of next year, kind of separate stuff out, move it across the world and decentralize it a little bit. So, do Just to build the intuition here, if Gito, let's say your guy's servers go down, you forget to pay the AWS bill or whatever, what happens to Solana block building? So that's actually a pretty cool thing where the network will continue to operate as normal. There's basically these things that we call relays, not to be confused with flashabout relays,

they're kind of like transaction forwarders. And the validators, when they start Gito Salana, they connect to these transaction forwarders. Anyone can run forwarders. Gito Labs runs a few. Some of the validers are operating their own. And basically, it's kind of like a transaction forwarder that will kind of send transactions to the block engine. And the block engine is really only sending bundles to the validators. So if our block engine goes down, unfortunately it happens sometimes. I got paged at like one in the morning yesterday because they went down. I had to debug some stuff. But like the network keeps going as normal. the functionality that you lose is basically the ability to send bundles.

So we're not like, you know, we're pretty important to the network and reducing spam and trying to atomically capture MEEV versus it kind of going to like the spam game. But we're not so critical in the sense that if like we forget to pay a bill or there's a bug or something, then the like network stops working. There's also like a lot of like heartbeats between. all these pieces. So there's like 500 millisecond heartbeats and, you know, health checks and all this stuff.

And I've thought about this a lot to where, you know, if something bad happens, then the validator will kind of disconnect from everything and like revert to normal settings and the normal transaction processing. And the way that our bundles work, it currently runs in parallel to the normal transaction processing code. So, yeah. Got it. Okay, so let's zoom out a bit from the micro details of GTO and let's talk about your

irdrop, because your air drop, I think, is probably one of the legendary airdrops in crypto history at this point. It's analogous in my mind to what the compound air drop was, or not the air drop, but rather, but the liquidity mining on compound, where it was the, it was sort of the, it was the moment that kicked off DeFi Summer. And in my mind, the Gitoirdrop is kind of been the thing that brought Solana out of the Ice Age and reanimated a lot of the animal spirits and the excitement around Solana. So why don't you walk us through what happened with the Gito AirDrop? What was your guys' experience of it? And what do you think went well or went poorly in what is now being called the Gito stimulus check? Yeah, I mean, it's cool to see how many

people are excited about participating in GEDA governance. I think there's a lot of excitement around Solana right now. There's been a ton of increased activity and interest level in GEDA, but also other Salana protocols that are kind of like pre-governance token. And so it's really cool to kind of somewhat like participate in that. I think the like, you know, the foundation did that the GEDA Foundation did the, the AirDrop, I think, like, my personal opinion is that I think it went pretty well. I think if you want to, there's definitely like a relatively small number compared to some of

the other protocols that you're seeing. And, you know, I think the game has definitely changed on Solana in the past, like, what is it, three weeks or something to where I think, like, you probably can't do what the GEDA Foundation did. But I think that it was a lot. It was a lot of went pretty well. And if you want to get a lot of people... You mean because of the rise of sibbles and inorganic behavior. Yeah. Yeah. Yeah. Yeah. I think that if you want to, I think you have to be careful because there's a lot of people that are not necessarily long term aligned with protocols. And so you have to be even more careful now that there's a lot more attention on Solana. Right. So just high level numbers for the audience. So the, the, the, the, the, the,

was in early December. It was at present prices over $200 million worth of Gito to over 11,000 addresses on Solana. You know, many people got tens of thousands of dollars in anirdrop checks. And it is created a huge amount of airdrop farming and the kind of anticipation of what's the next air drop going to be. There's excitement around Jupiter, around some of the other products on Salon. Yeah, margin pie drifts. Camino. Right, right. A lot of anticipation now for, okay, what's the next

irdrop going to be? And is the scale going to match what people saw with Gito? So I think one thing that might be worth pointing out to listeners is how Solana air drops in this cycle sort of differ quite a bit from Ethereum air drops historically. I think the notion of point systems where people have to interact with protocols and you don't exactly know what the distribution will be. And there's more of this almost game-like aspect to how AirDrops work now. I give the Solana ecosystem probably personally, like most of the credit for really pushing that forward.

I think in Ethereum, people still stick a little closer to passive liquidity-based incentives. So it would be great to hear about how you thought through this point system, how you kind of view it evolving and sort of how you view the importance of these points systems within the ecosystem. Yeah, I think points offer, there's a lot of protocols that are doing points now in Solana.

So I think Tensor might have been the first one or maybe margin phi. There were some Gito points for a while. We're seeing, I think, Camino has points now where they will assume. parcel as points. I think it's a good opportunity to kind of experiment with mechanism design and kind of see how to like, how do you kind of quantify people's contributions to a protocol?

So like, I think for a lot of protocols, it might be like TVL times. time. Maybe there's certain things that you want to incentivize more because your protocol makes more money on it. Maybe it's like volume or something like that. So I think it's kind of a interesting opportunity to kind of like play around with things and kind of see what people do. And I think we'll see more of it. We're already kind of seeing this on some Ethereum protocols. Like I know, I think doesn't Eigen layer, do they have points right now? they have like a billion dollars of TVL with points saying like I think I saw something with rainbow wallet the other day

rainbow wallets making a ton of money with points so I think it's you know you can kind of it's kind of like a playground that you can experiment in and see what people do with different kind of like with the leaderboard basically yeah I guess a question I was like you know why why do you think that point systems took off on Solana versus elsewhere.

Because I do think there is some level of, hey, it's actually quite a bit more sophisticated. So we have to have like a more complex set of contracts. Maybe that's the reason or you know, like there's clearly something that I think the last six months is the reason that the gestation of this kind of idea worked much better in Solana. Yeah. And of course it gets copied. But I'm curious like what do you think the things were that like led to that? Who is the first protocol that created points? Was it blur?

It was blur, yeah. I mean, I think it's easy to see why people are doing it. If you look at protocols that implemented points, kind of before and after, you can kind of see what's going on. I think it's easy to look at other protocols and, like, kind of look at it before and after points and you see the growth. Like I know, I guess, TensorFlow and margin phi were kind of the early ones in Solana, and you look at the growth in MarginFi.

and then you kind of look to see what happened to the GDSL TVL, and then if you're another protocol in Salon, and you kind of see this, it's like, oh, we should probably do this. It's kind of like an interesting marketing mechanism. Yeah, so you think it's more of a protocol founders looked at it and kind of implemented the playbook more aggressively versus like, hey, we're doing kind of more technically sophisticated point system. Because I feel like if you look at the contracts and the code,

the code is much more, you have to think through a lot more about reasoning about it than you do with your classic liquidity mining contracts. How many of these point systems are even on chain? I thought most of them were off chain. Most of them are off chain, but they keep a lot of logic

for like keeping track of actions that are done or like aggregating. Or just like logs and whatnot. Because like in some sense, yeah. And I feel like it's actually like going to be a much richer, you know, I think in 2024 we're likely to see this system evolve to be much richer than what we see in like the current

liquidity mining universe. And so that's sort of why I think, at least to me, I feel like there's something very interesting about the fact that it really took off on Solana. Like I feel like BlurPoints sort of worked, but they attracted like three whales, you know, like they attracted the few small, large NFT players. whereas the Solana point system seem to have been much better

at getting distribution and some of it feels technical, some of it feels cultural, so that's sort of where I was like, where do you, you know, where do you attribute that? Because I can kind of see both arguments for it.

Yeah. I mean, I think I'm definitely in like the Salana bubble and I don't necessarily follow the Ethel 1 and all the L2 protocol super closely. I think there's, There's a lot of super legit teams building on Solana, and there's a lot. I think that are relatively new within the past year or two.

And I think that I don't know if that's necessarily the case for Ethereum, some of the L2s. Maybe I could definitely be wrong here. And so I feel like there's a lot that are kind of in the same path that decide to launch points programs. And yeah, I mean, it's been super successful for a lot of protocols I've done it. So I think it's going to keep going for a while. You're starting to see some of the Ethereum protocols that I mentioned earlier start to do it. Okay.

Let's take a step back again from Gito entirely. And let's talk about Solana. So you've been one of the great kind of outspoken emblems of the Salana community. And now that Solana is in the spotlight and it's gotten this incredible surge of activity and attention and excitement, I want to talk about, you know, one of the things that's going on in the public conversation, which is this back and forth between Ethereum shills, Solana shills, and like kind of the critiques and the strengths and the weaknesses of Salana and the Salana roadmap. So one of the things that a lot of people talk about when they're firing pot shots at

Salana and saying like, oh, what the, you know, why is Solana pumping so much? This is crazy. Is that there's a claim that not that much has really changed on chain relative to where Salana was, call it three months ago, four months ago, when Salana was a fraction of what is trading at today. From your vantage point where you're sitting in Gito, what do you see in terms of what's changed in terms of on-chain activity from now to, let's say, four or five months ago? Yeah. There's been a ton of new activity on Solana. There's a lot of new capital. So if you look at the bridging patterns and things like that, there's a ton of new activity coming into the space.

there's a ton of super high quality projects that have been grinding basically since like the middle of last year. And, you know, I think a lot, what happened with all the FTX stuff that really sucked for everyone, but there's a lot of teams that kind of just kept grinding and building feels like those teams are finally like getting the recognition and adoption they deserve. I think there's a lot of people that were kind of, I think, skeptical to use Salana or that maybe they just weren't interested in it. it. You know, I got to like bridge. I got to go through some sex or go through a bridge and get money on here and there's not really that much to do. I think there's certainly a lot of group chats on them in where people are, people have been talking about Salonamore. They've been trading more and, you know, using the applications a lot more. There's just like a bunch of crazy stats. It's like there's like multiple like five or six thousand tokens being minted every day. There's a ton of new volume on chain. We're seeing, there's a lot more people trading on chain. There's a lot more MEV happening on chain.

JETA is basically 50% of all the MF tips have been captured in the past two and a half weeks. Or 50% of all Gito-Salana related MEV tips have been captured in the last two and a half weeks, 75% in the last four weeks. So there's like a ton of volume happening on chain. Wow. And so a lot of this, a lot of this resurgence in trading has been these new Salana meme coins, if I understand it clearly. Obviously there's Bonk, which has been around since the beginning of the year, but it's kind of gone vertical over the last month. And there's been this just panoply.

And there's also inscriptions now on Salana, I understand. And so things are kind of just going crazy. Can you talk us through like what you're seeing in terms of where the MEV is coming from? Is it mostly from meme coins and inscriptions and this kind of thing? Yeah, there's definitely a lot around meme coins. And even like just like normal trading like Solana, USTC and things like that. There's a ton of arbitrage. There's people that are getting to like some of the longer tail MEPI,

like liquidity sniping and things like that. And yeah, I mean, you know, on Ethereum and other chains, I think MIV is pretty closely correlated to volume. That's probably the thing you can the most closely correlate it with. And so you have a ton of new tokens popping up. There's a ton of price discrepancies between all these tokens. The fees on Salon are a hundredth or a thousandth of a penny. So basically any trade that happens is going to produce an arbitrage.

So you're seeing a lot of that right now. There's been a lot of like a lot of DAPS that in like protocol developers that have been working pretty hard and they're starting to see more adoption. I think there's been a ton of changes to the Solana core protocol as well that I think if this were happening like two years ago or three years ago, like the chain would be struggling super hard. So like introduction of quick, there's a much better scheduler in place. There's priority fees. there's still a lot of work to do on all these things, but they're kind of like the base building blocks

are getting much better to where the chain can actually handle it without kind of dying. Yeah, yeah. So actually, maybe a question or like something that's, you know, I think there's a lot of chatter on Twitter about is like, you know,

should the dominant form of value that gets generated in the salon network come from MEV because transaction fees are low versus, you know, like, there's kind of this philosophical divide of like, MEV is actually parasitic to a security of a network or MEV is actually a very good source of revenue for the network. You know, where do you, where do you fall in that kind of dichotomy? And I know Haseeb probably also has some like, well, so, but MEV on Solana through

GDR, MV doesn't get burned, right? MV just gets distributed to stakers. Is that correct? Yeah, so there's a few different forms of MEPE on Salon happening. One is like people are still sending using priority fees. I guess I don't know if you want to count priority fees as like MIV or not. You know, depends on who you are, but 50% of priority fees get burned. There's a ton of bots and users interacting with the chain. So those priority fees are getting burned. The system that the GDISLano Valdeir client and that all that hooks up to,

it doesn't burn the MIV. So it basically collects MEP and distributes it to validators and stakers. So there's not the main source of like any potential MEV burn would be from priority fees on Salana. I think one of the main themes in sort of the EVM ecosystem over the past, let's say a year and a half in the realm of MUV has not just been about sort of redistribution and, you know, Mev Boost and your relays and your different block builders. is it's also been about MEV minimization and redesigning protocols to reduce opportunities for

any of the things like, you know, intents or moves to RFQ systems for Dexas. Do you see a similar kind of trend in the Solana ecosystem or like, you know, how are people sort of re-architecting their DAPs in Salada to reduce MEV given sort of the different constraints of building on Solana? Yeah. I think we're definitely seeing that. I think it's catching up to Ethereum a lot. I think if you, like the most, one could argue the most valuable thing about uniswap is the front end.

You know, they started charging the front end fee. I think, I don't know the numbers off the top of my head. I think if you were, I'm guessing, maybe you guys know and correct me if I'm wrong here, but I would say there's still a lot of people on Ethereum that probably aren't using aggregators. Maybe I'm wrong here. I think on Solana, there's Jupiter. I would say majority of people use Jupiter. It's kind of like the one inch equivalent.

And so Jupiter, I think, it'll split your swap into, if you're swapping from like USC to Seoul, I think it has a limit of four or five swaps within that transaction. So it'll actually pull, and gas is so cheap. So it'll basically like pull from all these different deckses. And so I think on that sense, like I think, you know, Jupiter is an amazing product. It's built anyone that adds some,

type of swap interface is going to put that in their app, whether it's like any wallet or, you know, Phantom backpack, other wallets. So I think on that sense, there's a lot being done. I think we're definitely still noticing some like slipups from DAPs and trying to reach out to them directly. There was a, I'm not going to name the DAP, but there was a DAP that was allowing people to adjust their slippage, but they weren't showing users price impact. hacked. So people were selling tokens in the DAP for an eating like a five or 10 percent slippage, which was causing these like massive arbitrages. And we saw it in the graph and reach out to them. It's like, hey, you guys need to do this. And they're like, oh, shoot,

we need to push out like an emergency update. I think we're like starting to see it go towards the aggregator stuff. I think the nice thing about Salana is that maybe RFQs, because it's so cheap. I think you can probably have more liquidity on chain and you don't necessarily need RFQs. But they do give you better pricing. Maybe they're like supplementary to on chain liquidity or something. The interesting thing going back to what Turin was mentioning is that right now, despite the explosion activity on Solana, fees on Solana are pretty limited, right? It seems like much more value is getting captured by MBV, which actually gets a share of the economic value that's being exchanged on Solana

compared to the fees which are more a function of how full are blocks and because of Salonnas enormous capacity blocks are... My understanding would be that blocks are not full

there's still plenty of capacity on Solana. Is that more or less correct? And if so, how do you think about the economics of Seoul as a token given that

so much of the value that's being captured by Salon as an ecosystem is not actually getting burned or affecting the supply of Seoul, but rather just being paid out to validators or to Stake's soul. Yeah.

And does that matter? There's a lot of good questions and answers in there. Yeah, right. So right now, the validers and stakers that are running Gita Salon are making roughly two and a half to three times more per block than non-Judo validers. Two and a half to three times more. Yeah. Holy shit.

So, yeah, so like on... To be clear, that's like net of inflation. That's not just two to three times more fees. It's two to three times more including the block reward. Yeah, so that's... I'm just talking about the block reward here. And then it's kind of up to validators on how they want to distribute them. Do they want to keep it all?

Do they want to share it with their stakers? It's all kind of encoded on chain and all the tip programs and stuff like that handle it. So I think the average valet are not running Gita is making around like I think it's 0.01 sole per block. The ones that are running GD are making 0.03 and they kind of distribute it to stakers as they want. And they can choose 0% to 100%. I think that there's a lot of there are a lot of full blocks right now. There's no dynamic base fee on Solana or like there's not really.

Like EIP 1559, it's good. No matter what you think about it, you know, some people love it, some people hate it. It's some type of economic back pressure on the system. And basically your transaction has zero percent chance of getting scheduled if your base fee is lower than the 1559 base fee. It's like, you know, you're going to, if you try to build a block with that, it's going to break consensus and, you know, it gets rejected or whatever. Salana, there's static base fees, and there's no, there's like priority fees, which work decently well, but there's not really like a really good back pressure for a lot of transaction spam.

Lucas, I thought on Solana there was like a global fee floor that like rose and fall based on utilization. Is that not correct? No. So it's right now it's 5,000 land ports. per signature, and then there are priority fees. And the scheduler will attempt to prioritize transactions based on the priority fee. I think there's definitely a lot of room for improvement here, but essentially there's no

dynamic base fee here or no floor. And I think this is a very heated topic in Salon right now. Wait, so hold on. So, okay, because we actually about a year and a half ago, I think we had an episode about Salana where we were arguing pretty vociferously about Salana's fee market being broken. And this is before, I think, what's called state-specific fees where you have fee markets based on particular pieces of state, local fee markets. I think, yeah, that's what they're called. Where, you know, if you touch one contractor, if you touch an address and that address is particularly hot, meaning it's getting a lot of contention, a lot of people trying to touch it, those, the fees on those particular applications. applications or addresses float and they may be higher than it would be touching a cold piece of

state that not many people are contending over. There is that local fee market such that if I'm touching Jupiter, for example, Jupiter's contracts are going to be hotter and that's going to be more expensive than if I'm touching some lesser used application, right? That does exist. It exists in the context of the scheduler, but it's not actually enforced in consensus. So if you are, basically, if you are a leader, you can, if you wanted to, this wouldn't really make any sense, but you could basically pack a block full of low priority

transactions. And it's kind of an implementation detail that's up to the validator. And it's per state, per piece of state. But the default validator is going to do the logical thing and like maximize the total priority fees over the block, correct? Yeah. Yeah. Right. So, So in practice, we do have these local fees that get enforced by the validators such that somebody touching a more expensive piece of state is going to pay more than somebody paying something cheaper. So there is some. Yeah.

No. Okay, give me the details. Let me understand. It gets a little tricky when you start looking at the current scheduler. And the fact that you're streaming blocks as you're producing a block throughout the entire slot. So you can imagine a world where like the validator, it had a super hot piece of state on the previous block. Does it get reset every block?

Yeah. Like there's no, there's no minimum. There's no dynamic floor that's enforced. There's no persistence across blocks of this was hot the last block. So we're going to keep it hot in this block. Yeah. So there's no EIP 1559 equivalent or however you want to call it. There's a lot of, it's becoming more of a hot topic recently.

Depending on who you talk to, there's a few different proposals to fix this. One is the program, what is it? Program rebatable account. Whatever the name is. It's pror. I don't know what it is. But there's one where it's like, oh, maybe the applications should control what the fees are per account. and then there's also, I think, people are finally getting over.

Let me ask you this. How do you think fee markets should work on Solano? I think that ideally there's some EIP or like controller-based account fee model. So like the little EIP 1559 per account. And each of them has their own little curve that's moving up and down. Yeah, so I think, yeah, like you can do like, I think Anatolyeta post or something that he wrote recently for like exponential moving averages where you kind of look at the past end blocks and it's above views as many, this much of the block space arises and if it falls off, then it kind of goes down. I know Teroon's been like massively nerd snipped into looking at this. So I'm kind of curious what he has to say about it.

Yeah, actually, it's funny. I'm writing a blog post about this right now. So I have it fresh in context. So an interesting thing about, yeah, Saloness fee market is, A, the block rewards actually are distributed for voting. It's not like the proposer gets the block rewards. And B, the main incentive to the proposer is the fees. So 50% are burned. But then there is kind of this, yeah, there is this issue of persistence.

But one question you might have is, I have a bunch of. different local fee markets, right? So like maybe tensor is, you know, is hot and Jupiter is hotter and Camino is less hot than Jupiter. How do I decide whether these things should be increased in a correlated way or not? So I think a lot of the local fee market initial design was sort of, you could think of it as having been done looking at NFT mints that caused issues on the network, where there's a single piece of state that 99% of demand is going to, but the rest of state doesn't matter. And I think the thing where these base fee adjustments,

these kind of floor adjustments help you with, is when you have many parts of state that have increased demand and their demand is correlated to each other. So you need to kind of give a floor so that those pieces that are correlated don't just take up all the block space immediately. And so. It sounds like you're imagining as kind of global fee level as well as these micro local fee markets and they're superimposed on each other.

And so, you know, there have been many proposals for this. People like me and others have written some academic research on this. But there's sort of this idea of like you have global control plus local control and you have some way of synchronizing them over every block. So there's sort of some global floor. But then there's also some local sort of like velocity. like how fast you increase the fee as a function of the amount of demands seen in that block and the previous block.

And so you try to adjust the two of those such that you maximize some outcome, like high throughput or, you know, minimum number of missed transactions. But another thing that I think is actually important in, yeah, disclosure, Gito investor. So this is one of the things that made me most interested in Gito in the beginning, was that Gito does sort of these state-based auctions. So unlike M-EV and Ethereum, where you're collecting all the M-EV across all applications, putting it into single block and then submitting the single block,

here you sort of can think of the auctions in GEO is happening on a per program or per state basis. Like if there's different uniswop-style pools that are, getting different amounts of demand, you could almost think of their MEV auctions run separately before they're merged. And that allows you to have more efficient priority fees from MEV

than this kind of fully aggregated version. And I think Solana actually is a little further, actually a lot further in having that running live. And Ethereum is kind of getting to that point. With all the L2s giving you like many different blocks, people are starting to realize they basically have to have some notion of like, you know, state-dependent MEV. And I think Solana kind of pushed the barrier in that a lot further due to necessity. And, you know, obviously Lucas knows this better than anyone else.

Well, it seems like it's not just necessity. It's also that the design of Solana makes that a lot easier because of the fact that every transaction declares the state that it's touching in advance. In Ethereum, you don't do that, right? Because everything is. kind of at runtime, you can have a transaction that says, oh, I'm not touching uniswap. I'm actually an AVE transaction. Please schedule me at the top of the AVE auction. You win the AVE auction. But actually secretly, you're also doing some uniswap arbitrage at the end.

And it's hard to statically detect that. Is EIP 2930, which is the optional access lists? There's a lot of debate, almost as much debate in the Slana community over EIP 1559's like these kind of global controllers. In the Ethereum world, there's a lot of debate over these kind of like, can I add there's almost a soft fork and access control list of like, hey, if you tell me which particular state you're touching, like you're a roll-up, I'm only going to touch like these particular accounts.

I can schedule you more efficiently. And so we're actually starting to see some of that happen because of roll-ups, because they're kind of inefficient, you know, storage usage sometimes or the costs are sort of like inefficiently being allocated. And I think I suspect my prediction is that the Ethereum model with many L2s will actually start to look more like the Solana model over time, where you actually do have to be auctions. In the MEV auctions, yes.

I mean, stateless clients have been a story in Ethereum since forever. I mean, they're so far from ever actually getting shipped. But in a stateless client world, that would be true. for Ethereum as well, that a transaction needs to specify what state it's touching, and then you can kind of say, great, we're going to, we're going to cord an off people touching this state versus people touching that state. The difference, of course, in Ethereum is that there is no parallelism in Ethereum, meaning that even if you're not touching the same contracts, ultimately, Ethereum and the EVM require a strict ordering. So it always is to be transaction one, then transaction two,

then transaction three, even if the transactions don't interact with each other, because there are just these bottlenecks in Ethereum, like these resources that every single transaction touches. One question that I have that I've seen many people kind of getting into Twitter arguments and, you know, telegram chats and all this stuff, a lot of the Ethereum, or I should say, a lot of the Salana people will say, well, look, Salana right now, you know, we've got like the four compute unit cores or whatever they're called, but we're going to scale it up over time to like 16, then 32, then 64, and then, you know, whatever, it's just going to go to infinity. And it's going to be arbitrary levels of parallelism in Solaniland, such that there's no limit

to scaling Solana performance. And my natural thought is always, surely there is some bottleneck somewhere in Solana that is like some kind of global state somewhere that's getting touched when people are doing things on Solana. My question is, is that true? And if so, where is it? Like, what is that global bottleneck that stops Salana right now from scaling to 64 cores or 128 cores or whatever it is that people are imagining Salonnas going to scale to? Yeah.

I think that I don't know if I've seen people saying that it's going to scale it infinitely. I think there's definitely people that are saying it can scale a lot more. and I think there's a ton of there's a ton of scheduler changes that I think will make it better I think there's a lot of room for improvement in the actual replay as well so like a few months ago I wrote a better replay algorithm it sped it up like 50% or 100% I was kind of scared to ship it to prod but it definitely

you know use less compute ran faster I think there's just like a lot of these tweets As far as like the actual limit, I think this is probably where the thing that Turin was talking about earlier, where you have this, you have these like per account limits. And then you kind of have this like global limit floor that kind of rises. And I think you can take into in that global limit, you can kind of take a lot of different things into account. So maybe the global limit includes like what is the what's like the max throughput that we're trying to, to talk. target for packets on the network. Basically, how big are the transactions in bytes and how much, how much throughput and like, how much bandwidth is it required to spit that, those transactions out and process them to like how much,

how much account state was loaded and stored in that block. So I guess, like, I don't know what the current limit is. It's probably like memory or memory speech. or something. But I think that's something that you can kind of, when there's a better fee model on Solana, I think it's something that's adjustable. And you can kind of find where that global limit is and use that for that global limit floor in addition to these separate kind of bucketed fee markets.

Okay, let me ask the question a different way. What are the hard lines for Solana? In Ethereum, the hard line is that you have to be able to validate Ethereum. on a laptop, right? That's like the ideological center that Ethereum uses as a constraint when thinking about throughput. If you look at Binance Smart Chain, their hard line, as far as I understand, is that you have to be able to sink the chain.

And there was a time when Binet Smart Chain was running at like, I think, $25 million gas per second. And that was just so much that literally, even like Enterprise-type validators who had, you know, huge boxes and we're running the stuff as fast as they could, they just could not sink to the state of the chain. And so they lowered the total gas throughput to about a third of that so that new people who are joining the chain could actually catch up to the tip. But that's about it. Basically, as long as people can sink, I don't really care. As much as I can get out of like this kind of, you know, souped up throttle to 11 EVM chain, that's what I'm going for with Binance Smart

chain. Yeah. how does salana think about that is there a hard line on salana of what we will and will not do in order to get a faster higher performance chain what is the what is the binding constraint beyond just what can we get out of raw hardware yeah i don't want to speak for like salana labs here i guess my or like the foundation or anyone i think um currently the the hard limit is like 48 million compute units. It's kind of a proxy for a lot of different things here. And on Ethereum, I guess, you know, there's a 21 million gas per block.

I don't really know how all the things influence, how all the different things that transactions are doing and block validation and whatnot fits into that. But, you know, I guess that maybe that's something you could view or people if they, it's like gas per second or something where it's like 21 million gas and then you have 12 second block times and there's a bunch of signature verification, whatnot happening. On Solana, it's 48 million compute units. I feel like on Solana it's probably basically like what is the what's the lowest barrier to where 66% 66.66% of the change. can keep up.

So how fast can we go such that 66% of stake can keep up? Ideally, 100%. Some people, they call them potatoes and Solana. There's a lot of validators that are running potatoes. I guess for context, our servers are, we're running, we're probably running like 20 or 30% compute usage right now. So I think there's a lot of space and room to speed that up. I mean, there's also a lot of extra bandwidth that we can use.

as well. Actually, I just want to add one kind of thing to your question, Haseeb, which is like, I kind of view the, you know, the Ethereum model and the Sondon model when you're talking about scaling compute, like, hey, we have a lot more parallelism in synchrony. As just different programming models for parallelism, like in some sense, right, you know their limits to how fast particular algorithms can run when in parallel versus serial and you can't improve them beyond that amount. And there's a set, however, how you represent your parallel computation impacts how easy it is to actually achieve whatever scale you want, right? Like if you want the algorithm to run 10x faster, well, one form of parallelism might work better than another form. And what I mean by that is like, you know, something like Nvidia's kuda works well for linear algebra of parallelism because that's just like, what it's designed for, but it works bad for, say, sorting a list. It's sort of inefficient at doing certain things like that or things that have recursion. And I think in general, you should

view the Ethereum form of parallelism of we have these multiple units that are running separate computation, but infrequently synchronized with each other versus the Solana version where actually they have a shared memory map or they have some shared context that they have to keep in locally at all times as two different programming models. And some algorithms will be much easier for one and some algorithms be much easier for the other. And there'll be some that don't work on both. And the history of parallel computing has had this happen, whether it's things like MPI in the 80s, things like Kuda now. And this general idea that like the way you represent, how you parallelize your computation,

dictates what applications work well on it, I think is going to be just the truth for blockchains for the next few years. But I think that's maybe one of the reasons that's also hard to compare these things apples to apples. Yeah, I think that certainly makes sense. All right, I know we're running up on time. So I wanted to just kind of close out, Lucas,

with, I guess, getting a little bit more of the kind of felt experience of what it's been like for you personally. And just what it's been like for people in the Salana community. So Salana, you guys have really been through the Valley of Darkness with the collapse of FTX, Solana going all the way down to $8. And now you've had this kind of euphoric

coming back into the light. Salana now surging to be number four on coin market cap, you know, over $100. Everybody's super in the money. Jito has been super successful. How does it feel being part of that cohort where you guys, you know, you guys are really toiling in this environment where everybody kind of left Salana for dead.

And what does it feel like now being on the other side of that? Do you have time to really rejoice? Do you feel like, yo, guys, we made it. Like we were right. We believe when no one else believed. Or is it just like everything's happening so fast you're just trying to keep up with it? What does it feel like from the Solana OGs right now? I would say all of the above, for sure.

Sure. I mean, I think last year at Breakpoint, super exciting. The Judo Stake Pull was launched, the Gito Salon of Outer Client of Outer client. I think it was out the Judo Salon of Outer client was at like 3% of stake and to like FTX happening. And like a lot of people were leaving. It was like a couple weeks after FTX was breakpoint, right? Or was it after FTCS? No, it was on the plane. It was on the plane. It was on the plane. It was on the plane. ride home that the rumor of Binance buying FTX was like popping up on people's phones and Twitter. And I remember the internet was super spotty and everyone just like keeps trying to refresh Twitter to like get their, get the newest news. So yeah, it was pretty stressful.

There's definitely a period there where it's like, you know, you're doubling down, but also like questioning yourself. Like, this is like worth it. But, um, I think, like, a lot of teams just kept building. And I think you, Salana is just, it's so much bigger than FTX. Like, I think people associate the two so closely. But if you actually like, like, do people actually know, like, the Salonacore engineers? Like, there's a lot of really smart people there. There's a lot of really smart teams.

And I think everyone kind of bonded together. And I think the last, like, certainly the last few months. And then the last three weeks is, it's like, some sense of like excitement, pride, maybe like a little vindication for people that like made fun of Solana the entire like basically the last year. I think there's also like a lot of, there's a lot of excitement and hunger.

I think there's still a lot to build and a lot to do. And I think a lot of people are kind of soaking it in and like making sure that they keep their servers running and things are going smoothly and all that, But also, like, you know, the things we're just getting started here, and there's still a lot to prove. There's a lot to build a lot of, like, users and marketing and things like that to keep people onboarding.

So I think, you know, definitely nice to, like, soak it in and kind of reminisce on, like, how crazy last year has been and how stressful. But I think there's also a lot of opportunity and a lot of excitement. And I think that a lot of people feel like that. on Salonra right now. That's great. Well, it's awesome to see that you guys getting to see the fruits of your labor and getting vindicated for having believed in something at a time when few other people did.

And I think there's a lesson for a lot of entrepreneurs out there. Any advice that you would give to somebody who's thinking about building in this environment that you've learned from your experience going through the peak, the trough, and then back to the peak again. Yeah. Oh, man. I think, yeah, I think you have to, you have to be confident in the choices you're making and make sure that you have, that you have a right team to support you and you have supporters there. I think anyone that's an entrepreneur is like kind of like messed up in the head a little bit, especially in this industry. And I think having a good team on your on your side definitely helps a lot i think also making sure you're um i think there's a lot

there's a lot of value and staying focused in the right thing there's a lot of um especially when prices start going up i think it's really easy to get distracted or try to catch the end of trends and kind of like miss miss the trends and you're kind of always chasing and i think it's really important to kind of figure out how you are going to kind of attack something and grow and stay super focused on that. And I think, yeah, those things are definitely super important as a entrepreneur. Well, the Solano ecosystem is very lucky to have you. And congrats on all the success and hope to see more of what Gito becomes as you guys keep evolving alongside Solana. Thanks. Yeah, I appreciate it. Thanks for having me. All right. And congrats to you as well,

Turin for having invested in G2. I guess you two are the, you two, you two both deserve all the success you've had. It's really Lucas who deserves all the success. Okay, great. All right. Well, we got to wrap. Lucas, thanks for coming on. And until then, I guess it's the last episode for the year.

We'll be back next week. And hopefully more craziness will come of our way by then. I think the thing that's needed is Haseeb Whiff hat. That's the meme point that you really know. Is that right? All right. Good to know. Good to know.

We can sponsor that for next year. And then we also got to, at the end of year, we've got to go back for a PayPal bet. So that's going to, next episode, we'll adjudicate how the PayPal bet played out. Anyway, we got a wrap. Thanks, everybody.