Why Bandwidth is Still Undervalued in Crypto | Logan Jastremski

I think the Ethereum community is actually extremely brilliant. I think the hard part that they're going against is they're trying to fight physics in my mind, which I think is a losing battle where if you're prioritizing these stakers that only have a certain amount of compute capacity or have a certain amount of bandwidth capacity and the entire system is bottlenecked because you're designing the constraints about around these relatively limited actors within the system. I just don't think that is the correct thing to optimize for.

Let's face it, building on crowded L1s or expensive L2s is just not going to unlock the next generation of applications. Supercontainers offers dedicated customizable app spaces running on Supra's 500kTPS Layer 1 blockchain, giving you complete control over gas tokens and fees. Get notified when Supercontainers are ready for builders super.com/containers. This episode is brought to you by Anoma, the universal intent machine, introducing a new era of applications where you define the outcomes you want. To stay up to date with Anoma, you can follow them at x.com/anoma or sign up for the newsletter at anoma.net. Special thanks to Anoma for sponsoring today's episode. All right everyone, welcome back to Empire. We've got, uh, Logan. Logan, is this Is it with a J or a Y?

Jastrzynski.

Jastrzynski.

The American version.

The American version. Yeah. Logan's been on the pod before. I think people have probably heard the story, but I just have a ridiculous amount of respect for Logan. We met when he was running around with a 30-pound backpack on his back.

Still do that.

Yeah, still do that.

Yeah.

Trying to get jobs and recording podcasts and flying all over the place recording podcasts and didn't end up getting a job. And so he did what any sane person does is he launched a fund and bet on himself, and it ended up working out. And the reason it ended up working out is because he is incredibly intelligent and had a really unique thesis at the time, which I think can be categorized by, you know, bandwidth and throughput increases over time and compute increases over time. And you should bet on the blockchains that are optimizing for this. So I don't know, I always like chatting with you, Logan. And yeah, welcome back to Empire.

Uh, thank you so much, Jano, and really appreciate the kind introduction. It's, it has been a long journey. I think, uh, I'm just appreciative to be here. And I think more so than anything, uh, maybe we touch upon this throughout the podcast is there's been a lot of things that often change within the industry. And I think you can actually make a lot of money about front-running a lot of these narratives. I think for Frictionless and what I've tried to think about are the things that are not going to change. Because those are a little bit more easier to reason about. I think that Jeff Bezos quote is saying like, everybody asked me what, what is going to change? And I tell them what's not going to change is people want faster shipping and cheaper products. And so I've kind of taken a little bit of that approach just to get a baseline of reality in the industry that is so crazy.

Maybe we can actually start there. I originally wanted to talk about breakpoint stuff, but actually the best place to kick this off would be with that thesis. So maybe just tell me, you know, in an industry where everyone's talking about narratives and things are changing all the time, like, maybe just expand on what you think won't change. And like, what is the core thesis that you have when it comes to the industry today?

I think we did probably too much homework, in my opinion, on just the underlying infrastructure. And I think this was mostly just because my time at Tesla and Elon's kind of first principles approach to the industry. I think the thing that we ultimately realized, probably much quicker than others, is to your point, Jason, is the industry is largely going to be, at least from a scaling perspective on blockchains, limited by two main factors, which was overall throughput, and then kind of on the computational side, how to increase compute over time. And I think the industry, at least initially, was relatively conservative about what those requirements should really be. Should they be on Raspberry Pis? Should you be able to do— should you have larger bandwidth requirements in terms of going from like kilobytes to ultimately like low megabytes? And I think for us, the thing that we just kind of uniquely bet on is these requirements over time are just going to steadily increase. The first one that kind of pushed the boundaries on this was Solana and still gets a lot of flak today. There is a tweet as we were filming this with Snowden ultimately kind of poo-pooing on Solana's decentralization, which I would disagree with, but ultimately just betting on the fact that over time for this industry to really matter, you're going to need a certain level of scale and you should bet on the underlying architecture that allow you to increase bandwidth and throughput, bandwidth, which is throughput over time, and then also compute over time. And then I think that has allowed us kind of to move up the stack to applications is kind of where I think the industry is going and could also lead us into a lot of the conversations and takeaways from Breakpoint and our recent, recent Asia trip was about products. The thing that I've been saying more recently is better products, not better crypto products. Just like kind of like tongue-in-cheek, but ultimately saying like, for the industry to actually matter, you actually have to just build better things that people want to use. Like, sure, it's cool that it's uniquely on crypto rails. Like, I love this industry, I'm sure as you guys do too. But at the end of the day, like, these are still products and they need to actually deliver a better product experience, I think, for mass adoption. So within high-throughput blockchains, the things that we've kind of been narrowing down on our strike zone is overall in the next 5 to 10 years, we think trading volumes are going to like 100x, if not 1,000x over time. And then Deepen, because I think Deepen and the decentralized physical infrastructure space actually makes better products. So long story short, we've kind of been focusing a lot on the infrastructure side, which is just figuring out that, hey, you're going to need to increase bandwidth and compute. You can kind of move those pieces around, whether you're using EigenDA or CelestiaDA, ETHDA, or, uh, Movement, uh, Eclipse, whatever, Optimism, name your L2, or you can integrate those in kind of a Solana monad suite type architecture, but that physics has to happen. And then what are the things that are going to be able to uniquely built on once you actually scale the underlying kind of core constraints?

Interesting. I want to go into the products and talk about Breakpoint. But before we do that, you guys have built a brand as being very convicted on Solana, I would say, and being very early and very right and making a good amount of money on Solana. Is it— do you guys make bets on other ecosystems too? Are you investing in other ecosystems? Or is the idea that, look, we, we picked the right ecosystem and with Solana, and now we're going up the product stack, we're investing in products on Solana, or are you looking at other ecosystems that fit your thesis?

I think at the time when we started, it was really only Solana in our point of view that was high throughput. Um, since then it has brought in— you have the two spinouts from Meta, which are Sui and Aptos. Uh, you have Keoni building out Monad, uh, and then you have say that ultimately kind of spun out of Robinhood and all the saga that happened there. And so I think for us, our big underlying thesis is just build on infrastructure that actually scales. And I, I think we're opinionated in the point of view that I think the integrated approaches is probably the much cleaner approach, um, than kind of trying to modularize the stack. Uh, and we can probably get into that a little bit later. But really focusing on just the infrastructure that engineers, application engineers can focus on applications and not infra. I think the underlying thing that really frustrated me when I first joined the industry and like seriously started taking a look in 2020 was talking with engineers and they would remove lines of code because gas costs were too expensive. And so they're trying to optimize code just so that their application would actually work at scale. So within the high-throughput blockchains, I do think it's probably going to be a parallel or a power law in terms of like who gets majority of market share. I think most people like to hold hands and say it's going to be a multi-chain world and that like everybody's going to win. It kind of reminds me of Wagme a little bit. So obviously Solana today being the first, having the most robust ecosystem, is probably in my mind, still going to be one, if not the dominant player in the high throughput future. But we're very open-minded to Sui. We're very open-minded to Aptos, to Sei and Monad, and even more in the future. I think I view these all as engineering systems and they're selling kind of their ecosystem to developers. So we track very closely how the developer ecosystem is coming. So we're open. If there's anything that I've learned in this industry is don't be a maxi.

Hmm. Can you say more about integrated versus modular? Is this a— do you think that this is the right approach from a product perspective, or do you think that this is the right approach from— is it more about, you know, value won't accrue in the right areas? So it's not as— the modular stack is not as interesting from an investor's point of view.

The one thing that, like, even if you remove the tech stack, one thing that I've really appreciated about blockchains more and more is just the value of, like, the shared state within them. Um, and that generally the more that you fracture that state, it's less of even like a technical challenge, even though I love to nerd out on the little, like, technical architectures and different debates there. It's, it's more so on the social aspects. And I think we've started to see this over and over again, like which L2 do you build on? And there's another L2 and you kind of get the infrastructure fatigue. I kind of view the high throughput and the integrated high throughput chains as like when you decide to build on a SUI or a Solana of the world, you're opting into some shared set of standards and that's plugged into exchanges, the wallets, like everything just kind of naturally works. And that's because people, by using that underlying chain or the integrated chain, are saying, I opt into this standard that I want to use. And so even outside of the technology perspective, the social aspect I think is very underappreciated, and it is only continuing to be highlighted as more and more kind of friction enters these different landscapes, because it is hard to either bridge money from point A to point B, Or you just create new VC investments on the infrastructure side to actually fix those, um, fix those properties to begin with. But I think one thing that I really learned at Tesla is the best part is no part. If you can delete something, you should delete something. And I think blockchains more broadly like to invent new infrastructure things, rightfully and wrongfully, but if you can delete things, you should. And so To me, it's just that integrated approaches are cleaner. I think it allows application engineers to focus on applications. And so, that is the thing that gets us excited, actually getting adoption.

So, what ends up happening? Obviously, you like Sui and Aptos, and you mentioned Monad and Sei and Solana and stuff. What ends up happening with this modular ecosystem, in your view?

I don't think it dies. I mean, I'm all for experimentation. I am personally of the opinion that the integrated chains become more expressive over time. Things like token extensions on Solana with the Move programming language being more secure. These kind of advancements kind of continue to push forward. And so the main reasons that you ultimately hear for modularity is like, hey, we need to capture more value or we ultimately want to customize the stack more. I think those, and this has kind of trended true over time, is that the fragmentation that that potentially introduces outweighs the slightly less customization that you may get with kind of an integrated chain. But you also get all the shared state that kind of co-wraps the shared state around that. And so I don't know. I'm all for experimentation. I don't think there's necessarily— we should stop doing that. I think people should push the boundaries on what is possible. I just don't think there's enough differentiation between most of these things. And I think this is why you kind of see some of the fatigue in both the private and public markets.

Logan, what do you think of— there's a whole crop of companies that have popped up trying to bring Solana to Ethereum or bridge Solana with Ethereum or bring the SVM to to Ethereum, you know, a lot of different ways to describe it. What do you think of this crop of companies?

I think it's a cool experiment. I— we have not done any of those type of investments.

All right.

Let me rephrase the question then. Why don't you like it? What is— what is not— yeah, what don't you like about it?

So the thing that we ultimately realized was that, look, you're going to have— if you want to be fully decentralized, The reason why throughput is generally a bottleneck is because blocks are a certain amount of bandwidth. If you can propagate larger blocks to the rest of the nodes in the quorum, that's great, but you have to physically have faster internet connections if you want to be decentralized. And so what we— if you only kind of have these like SVMs, L2s and you're ultimately using, uh, Ethereum for DA, you're still going to be limited by Ethereum DA, which is generally, uh, I think since the blobs, I can't remember, I haven't been following it super closely anymore. It was like 0.375 megabytes, uh, which in the grand scheme of things is like less than like broadband, like very slow broadband type internet. Um, And so you have to have both high compute and high bandwidth. And I generally think what you, you see is like whether Eclipse or Movement, if they want to use Ethereum as settlement, that's fine, but they're actually not using it as like the full Ethereum DA because the Ethereum DA does not have enough throughput. So they use something like a Celestia or they use something like an Eigen layer to offload the data throughput. And this has kind of always been our point, is that you can't obfuscate the physics of these things. You, you have to do high compute somewhere. You can move that from the L1 to the L2, or you can move the bandwidth from like an Ethereum to a Celestia or an Eigenlayer, but they still have to happen. And so that's why we've just been focused on what are the underlying primitives that are going to continue to need to scale. And I think FireDancer is kind of demonstrating, look, you can go from kilobytes to megabytes to ultimately gigabytes, uh, and possibly beyond. Um, and so just being long the things that I think are going to continue to compound— most people, I think, vastly underappreciate the high-throughput blockchain still. I think the timeline may be consensus that Solana is going to win, but I don't necessarily think that's reflected in people's personal bags. Um, retail is generally ahead of the curve. I would say fund managers have still underallocated to Solana and the other high-throughput ecosystems. Um, even though that it may be consensus on the timeline, it's definitely not consensus, I would say, with people's portfolio construction. Um, and so I think we just want to continue to be long the things that are going to compound. Um, people underappreciate how things, when they actually start working, how hard they can actually compound.

Yeah. Do you have thoughts on the Solana L2s? Like, I think Zeta launched an L2, Ellipsis Labs, which built Phoenix and then recently launched Atlas, is another L2. I'm curious what you think about these because it kind of— it's Solana ecosystem and it's, you know, the whole thesis of Solana, but it also is this kind of fragmentation and modularization of Solana. So I'm curious how these two things come together in your head.

I would say, so I just did a podcast with Anatoly in Singapore and it was great. And he was like, look, we're nowhere near this end state of Solana architecture, but in the end state, I don't think L2s are necessarily going to be a reality. And if you go to engineers and say, hey, we'll promise you this thing in the future, just don't do what you're doing yet, or don't build an L2 today, but in the future, you won't need it. It's not a very compelling pitch. And so I understand why people want to experiment with Layer 2s today. I think the unique differentiation, or at least the Layer 2s that are very intellectually interesting to me, is the ones that keep the state on the Layer 1. So all, there's a bunch of different names for these. The industry always kind of confuses me on these. Today, I think they're either like coprocessors or ephemeral rollups. I actually just talked with a really interesting team this morning called MagicBlock, and they're building an ephemeral rollup for gaming, but that can be generalized over time. And the reason why I found them fascinating is because you can spin up these extra compute, if you will, where you can have lower latency and generally more transactions just because it's on a single computer. Uh, and, but that state does not actually live in that rollup. It's settled, um, or that state actually lives on the layer 1. So instead of actually being fragmented, like you have in generally the Ethereum ecosystem where Base has its own state, Arbitrum has its own state, Optimism has its own state, all that state in this type of blockchain or an ephemeral rollup would live on the layer 1. but you can do kind of really extreme amounts of compute or very fast latency if you wanted to. But that state, and the key difference here is it lives on the layer 1. So you still have the full decentralization of the layer 1, but if you need extra compute or you want to do lower latency, you can. And I, I think these small differentiations, right? Again, like going back from the technical standpoint, how they ultimately fit together really matters. Um, so I, I think short term, I, I definitely get why people are experimenting with them. I think if you do experiment them, these ephemeral type rollups where that state actually lives on the layer 1, but you can do more interesting things, either more compute or lower latency, that to me is interesting because again, it goes back to you're actually building a better product and you can still tap into that global shared liquidity that Solana has now started to build, you can still be composable with all the other applications and you don't have to pick, hey, should I build on Zeta's L2 or not to call them out or any other L2 that Solana could potentially build. It's all that logic and state is actually on the layer 1.

There's this idea, and I've heard a couple of people, the likes of Kyle Samani and others say there may be a network high throughput blockchain that may be a 2x improvement to Solana, but it's not going to be a 10x. And because of that, Solana has established itself and has a, you know, first mover advantage in high throughput chains and has that moat. Would you agree with that? And how much of that is actually defensible?

I would. I mean, you look at the difference between Ethereum and Solana and it's probably still FUDed today because it was actually the first blockchain that did not just copy Impasta, the VM. It structurally made different changes. It decided to, instead of running on Raspberry Pis, we may run on a $3,000 or $5,000 server. Instead of doing kind of low megabytes, we're going to hook this up to gigabytes and really push the throughput limits. And that resulted from the product experience in a very meaningful differentiation. I would argue it was much greater than a 10x, probably more than a 100x difference in just general performance. And I think generally that will kind of expand over time. And the new chains, as much as I find them very intellectually stimulating, the goal in my mind with all the high throughput chains is that you write the software. So when you go from 1 megabyte to say 10 megabytes, and when you up the core counts, you naturally have linear scaling so that you can just, you can continue to increase those kind of raw requirements over time. And once you do that, it's not that you're done, but there's like very like minor things that I think we talked about in the last podcast where it's like, all right, how do you do parallel processing? Is it deterministic? Is this optimistic? The new thing I think now is like, how do you do the different consensus algorithms? What is the final finality? Do you have one leader? Do you have multi-leaders? And I think these things are very intellectually interesting and can do an entire podcast about the differences between those. But it's not as clear, at least on the surface level, that these are even 2x better than Solana. And so, I mean, I talk with the different high-throughput teams quite a bit. And I think the goal and one of their hardships has been how do you uniquely differentiate yourself from a Solana of the world if it's harder to demonstrate that it's not a 2x better, or if it is, like, how do you clearly lay out that out so everybody understands that? From Ethereum to Solana is very evident.

Yeah, I really like the episode you recorded with the MegaETH founder because Most of the time I hear builders, particularly of L1s or L2s, just go all out and saying, try to king make and say, we are the best, categorically the best. And everyone claims to have the highest TPS or whatever. And so it's almost impossible to believe by definition. You have to make a trade-off. We can talk about what those trade-offs are, but he was the first person to just very intellectually honest to say, we've made a very explicit trade-off. And when I first heard that pitch, they said, we're going to run a Centralized Sequencer. This is how we think we can get the most out of Ethereum. And you don't hear many founders in the space, like, talk about that. I guess Toly is one where he's always stuck to his vision of Nasdaq on the blockchain. And you hear him time and time again say that. And I think that is perhaps a way of differentiating and explicitly saying what perhaps the use case is. But I'm curious, like, when you hear these founders, like, Not to go too much into detail of the different trade-offs of these chains, but like when you, maybe double-clicking on, if it's not a greater than 2x, how much? And might there be applications that are perhaps best suited in Solana? You're seeing deep-end payments really take shape there. But do you think there's an opportunity for chains to specialize? And if so, what might those use cases be because of a particular trade-off that they've made.

I, I think that's kind of why I've biased, at least in my podcast, talking with a little bit more of the engineers, because I found that the engineers think very hard about the limitations of their systems and what design decisions that they're trying to make, because they have to explicitly write that in code where the, the non-engineers, the more BD folk, they're great to talk to, they'll hype you up, but they'll, to your point, Santi, tell you that you have infinite TPS, which I've found, uh, just not true. But the MEGA team is doing something interesting, and I appreciate their opinionated point of view. I think on the other high-throughput chains, I don't know, it's hard. Um, I, I, what I have kind of come to appreciate is that application engineers want their blockchains and their applications to be used, uh, and that is why everybody was on Ethereum Layer 1. It's where all the wealth effects were. It was where all people were trading. It was where all the magic was kind of happening. And as you kind of push all that traffic to Layer 2s, it's not bad per se. It's just that social coordination. Are we going on Optimism, Arbitrum, Eclipse or whatever, like, there's just more overhead to determining, hey, where am I going to go? And so I think as more and more applications kind of coalesce on the Solana ecosystem, you see now new applications like Dawn or Adrena, Helium, like Hive Mapper, these things I think will start to compound as well, where If more people kind of build here and more users start to co-locate here, that, that actually to me is the thing that's going to be hard to overcome. Um, I don't know, I think it's not going to be a winner-take-all. It will be a winner-take-most where, uh, you'll probably have a couple high-throughput chains. Um, I think what Keoni and the Monad team, also the Sei team, are doing is fairly interesting and trying to kind of modernize the Ethereum virtual machine, make that high throughput. We'll see what happens there. But again, it is a little bit of a harder sell because how do you uniquely differentiate yourself?

Yeah. At this point, people should go definitely listen to your podcast. I think you are the person that can bridge these two worlds technically.

Logan, what's the podcast called? You wanna just plug it?

It's just my name, Logan Jastrzymski. I should probably rebrand it 'cause Jastrzymski is hard spell, but perhaps we can post it in the show notes.

That's how you know it's a good podcast when you haven't spent any time on marketing.

We've been slowly grinding over the bear market.

Yeah, I call you the compiler of the space because you can translate what is very technical audience with, you know, just making it more digestible and easy to understand without necessarily compromising the essence. So at this point, you've talked numerous times to a lot of these builders. Again, you spend a lot of time in Asia talking to them. You're also at Breakpoint. I'd like to transition maybe the conversation over to that. Just what were the highlights of those discussions? Anything relevant that came out? And then maybe talking a little bit about Firedancer, because I think that's, you know, all eyes were on Firedancer, and I'm just kind of curious to get your take on that specifically. As well.

Yeah, I mean, to me, Breakpoint is always a very fun conference. This— I've actually been to every one now, um, so it's kind of been interesting, the high highs and the low lows. Uh, definitely got extremely spoiled in the first one, as I think it was not representative of reality, kind of marking the pico top in 2021 where Solana went from 20 cents to $260 or whatever. I think FTX had some jet skis in the pool at that time. It was a little crazy. But the progression of the different breakpoint events has been really interesting because obviously in 2022, shortly thereafter, you had the FTX collapse. 2023, it was a little bit somber. Everybody, it was just like core builders really there. If you stuck around post-FTX and were one of the so-called 75 devs, you really believed in the ecosystem. And to me, the mark of this last breakpoint was really kind of a celebration of the trials and tribulations. It didn't feel overly enthusiastic, and it didn't feel like this was a deep bear market. It was just kind of people are there and they want to build products. And to me, that was probably the most refreshing and best one yet, just because the industry Solana specifically has gone through these really hard periods in its life. And now it's kind of come out stronger, where people— the underlying software is nowhere near perfect. There's still a lot of bugs and kinks, as Mert and probably Tolya would tell you. But it's gotten to a point now where application engineers can focus on applications. And so, I think the reason why there were so many talks on different apps is because that underlying kind of thesis of frictionless of, hey, once these things actually, you, you get sufficient throughput, you get sufficient compute, you can move up the stack. And I think that's where a lot of like the smart VCs are now going is it's not as much focused on infrastructure anymore. It's what applications can you uniquely build that are just strictly better products? Um, so to me that was just like a real treat of the industry. Um, just having everybody talk about their products and being able to, uh, kind of showcase at that breakpoint. It was, it was a rather cool trip.

Yeah. What about Firedancer for people that might have missed it? Of course, all the videos are online. People can go watch them at whatever 2x or 0.75 or 0.5 speed. But what is your impression of Firedancer and perhaps, uh, yeah, just some takeaways from that.

I, I think if MegaETH is the logical center, uh, kind of like end state of L2s, maybe Firedancer is the logical end state of like fully decentralized high-throughput blockchains. Um, we again, maybe like wrapping it back to like the starting point, you had Bitcoin, which did extremely low throughput but didn't have the ability to do smart contracts. Ethereum kind of innovated on that with the ability to do smart contracts and Turing completeness, but it still took a lot of the kind of like ideologies of Bitcoin, especially around kind of low throughput and low hardware requirements. And so when you go from ultimately kilobytes of throughput per second and you go from megabytes, there's 1,000 kilobytes in 1 megabyte, and there's 1,000 megabytes in 1 gigabyte, there's 1,000 gigabytes in 1 terabyte. So anytime you make one of these jumps, it's a dramatic increase in performance. It's, it's not small. Solana was the first probably to go from the kilobytes to megabytes. And Firedancer, to me, at least on the surface, looks like it's going to be the first client that will actually allow you to go from megabytes to gigabytes. And that's just a dramatically— it's a massive increase in kind of performance. And so It's been cool to kind of see what will the end state of these blockchains look like, and I kind of think it will resemble more of like almost high-frequency trading, um, because what teams like Jump ultimately do, and why I think they have a lot of unique, uh, expertise in kind of building these high-throughput blockchains, is because when you're doing high-frequency trading, you're ingesting tens to hundreds of gigabytes of throughput per second, trying to make a decision on all that data, and then trading on that data. And I kind of think that's where what blockchains are going to look like in the end state, which is why I think they're so well uniquely suited to build something like Firedancer that will ultimately take us from megabytes to gigabytes. And so they kind of did like a test teaser, where they said, Frankendancer, which was like a certain amount of like the FireDancer modules were live, actually in production today. And then I think the next day they had another talk saying that actually the full FireDancer client is also live on testnet. And so they're making good progress. They caveated it saying like, hey, we still need to do more testing, but we can actually demonstrate that this thing can be and was live in production. And so I'm excited about that. I think at the end of the day, what it will result in is similar to what happened when you go from bandwidth to kind of gigabytes or fiber optic internet, where, look, you could build YouTube-type applications once you went from dial-up to broadband. If you do kind of another 1000x, that allows you to do 4K gaming on someone else's server that you can stream to yourself in ultra low latency. And so I think with any of this, it always takes time for engineers to experiment what they can do in this new sandbox. But I think just now people are starting to experiment even with like Solana's more high throughputness. Obviously, every time you experiment to kind of push up to the limits, sometimes breaks and then you have to fix that. And I honestly, going back to how can the other high throughput chains kind of differentiate themselves, to me, in my mind, it's Solana being— it almost has the success that Ethereum has where you have to fix things as the plane is flying. There's a good amount of TVL, a good amount of applications. If these chains can see, hey, Solana potentially run into this networking issue, or we should learn from certain aspects there and apply those. That could be a unique advantage. But I'm just excited to see what application engineers ultimately build because I really do think as the underlying infrastructure continues to compound, you'll just see more and more interesting applications start to be built and applications and products at scale. And I think that's the big thing that the industry really needs to take us to the next step.

Yeah. Putting your investor hat on, How do you think about, and you alluded to it earlier, of your excitement in applications? What, you talk about Deepin, perhaps others, but how quickly does, I appreciate it's untested, but how quickly might we, might builders appreciate and then users appreciate the benefits of the improvements with FireDancer? And perhaps give us a tease of the applications that you think are just kind of now uniquely enabled with this huge upgrade.

I think it will probably take some time. I was a little surprised by how long it took for people to experiment with more high-throughputness of like a Solana of the world. What you saw initially is kind of like all the EVM clones that were just rebuilt in the SVM, partly because anytime you have a new ecosystem, you kind of need the core building blocks, the borrow-lends, the AMMs, the NFT marketplaces for like people to start speculating and just get, get the ecosystem off the ground. So I wouldn't say like overnight we'll see like explicitly new applications unless some engineer comes in and just like full really starts pushing things. But I think I'm uniquely just excited about what can you actually do when like you expand trading volumes from, I would say, like a billion dollars today to $100 billion a day, or $100 trillion— or not $100 trillion, like $1 trillion. Um, I, I generally just think people, because they're so focused on the narratives and moving from like hot ball to hot ball, underappreciate some of the things that just will generally continue to work and like actually potentially work better long term. And so we're very interested in the idea of how can you actually make these decentralized finance competitive or even better than the centralized counterpoint. Um, the thing that originally got me excited about the industry with Ethereum was kind of banking the unbanked, uh, the people that didn't have access to bank accounts, uh, in emerging markets. And unfortunately, just because the infrastructure and the fees associated with it, that didn't pan out. But I think you can actually do that with high-throughput blockchains. You can have them get basic financial services, but you can also allow them to like build wealth, whether that's speculating with meme coins or hopefully tokenized equities in the future. To me, those things get me pretty excited. And then also just more and more of these like decentralized physical infrastructure networks. I don't think those are uniquely— some of them actually may require more compute, but I don't know if those uniquely require high throughput in terms of like Hey, we need like financial information and we need it now. But if more and more start to coalesce on Solana or other high-throughput blockchains, then I think you just go the user mode. And that's where it comes back to, in my mind, the shared state aspect. People underappreciate the shared stateness.

Absolutely. Anyone that's been in finance appreciates. I mean, there was a time in history where people just tried to build closer and closer to the exchange because they had an information advantage because they would just get the information faster.

Exactly.

Yeah, for me, tokenized equities, like it's going to happen. And I think you're already starting to see teasers of banks appreciating, like moving in the T+1 instead of T+3, which feels like, you know, you're going from, yeah, very rudimentary to somewhat less rudimentary. But we know that there's a better way to do that than tokenizing equities. On the deep end angle, totally agree with you on the moats. You know, once you, you know, from a user onboarding standpoint, just feels so obvious that like if you start, you know, the next 100 million users probably come from a DePIN service that is very compelling. And then once they're using Helium, then it's an incentive to, you know, then understand how the wallet works. And then once they have the wallet, then very quickly they can discover other DePIN applications. So you not only get mobile, but you can get, you know, home wireless, and you can get, you know, earn different types of income streams and and pay for other plans. And so it just becomes a very sticky product as a category for Solana. And in my opinion, you're seeing the best founders and teams quickly realize that Solana is the place where just more activity's happening. And so you're just going to go and build there. Why would you bother building another chain?

You have your money there, you can borrow against that, you can interact with your favorite applications. At that point, it becomes more like of a black hole. And to me, that's, that's the really interesting part is where you actually start getting these applications where they're, they're building there because the users are there. And I think Ethereum did have that. It just got unfortunate when you paid $50 or $500 for gas fees. It just wasn't economical for the vast majority of the world. Now, if you can increase kind of overall throughput, if you can continue to scale the hardware so that you get more computation, The goal is to create, to make sure that you can continue to scale and support all these applications while simultaneously keeping these gas costs where they're negligible. And I think that's the magic.

Hey everyone, Jason Yanowitz here. I wanted to talk about something that's been on my mind a lot recently, and that is where are all of crypto's killer apps? We often ask this question on the show, so I'm excited to tell you about Empire's latest partner, Anoma. The Universal Intent Machine. Intents are a new way to build dApps, enabling users to say what they want without needing to specify how it's done. Let's say you have an NFT on one chain, but you want to swap it for a token on another chain. To make things even worse, you also don't know which bridge to trust or how to navigate different DEXs on each chain. Disaster, right? But here's the good news. Thanks to intents, you can just tell the app what you want to do and Anoma's Solver Network takes care of all the complexity in the background. We're excited to see what cool dApps Anoma's architecture will unlock. To stay up to date with Anoma, you can follow them at x.com/anoma or sign up for the newsletter at anoma.net. Special thanks to Anoma for sponsoring today's episode. Let's face it, building on crowded L1s or expensive L2s is just not going to unlock the next generation of applications in crypto. That's where Supra Containers come in, offering a completely new solution that transforms your development control. Imagine dedicated, customizable app spaces running on Supra's 500K TPS Layer 1 blockchain. You get complete control over gas tokens and fees with the freedom to use your own token, create local fee markets, or even go gasless. Scale on demand and build with better customization than AppChains. All at a fraction of the cost. Plus, with multi-VM compatibility, you can easily deploy your EVM, Move, and SVM smart contracts. It's time to rescue your projects from the costs, complexities, and fragmentation of traditional solutions. Get notified when Supercontainers are ready for builders at super.com/containers.

Two questions, but I'll start with the first. What surprised you the most? Or what has surprised you the most in these, as you have examined under the hood more closely, these high throughput chains or even Ethereum, like what has been the thing that has surprised you the most in this journey?

I think, I mean, the last like 8 months has surprised me a lot. I mean, it's one thing to like say that you think high throughput blockchains are going to win, but then it like starts to happen and you're like, Wow, maybe I'm not crazy because you feel crazy for a long time.

I know you well enough.

You feel crazy for a long time when you kind of go against the grain. And that is not easy. I think it's something that I've just kind of had to come into my own skin with. But I think the Ethereum community is actually extremely brilliant. I think the hard part that they're going against is kind of the discussions that are happening now on the timeline is, do we prioritize the solo stakers? And at the end of this day, at the end of the day, to me, it comes back to that question of physics. They're trying to fight physics in my mind, which I think is a losing battle, where if you're prioritizing these stakers that only have a certain amount of compute capacity or have a certain amount of bandwidth capacity, and the entire system is bottlenecked because you're designing the constraints about around these relatively limited actors within the system. I just don't think that is the correct thing to optimize for. It's not to say it's wrong. I just, I think if you truly want scale, if you truly want products that can hit Web2 type scale, the solo staker is going to be hard to, um, I would say keep around long term. If you want mass adoption, there's different ways to kind of get the benefits of solo stakers. Maybe you do something like Celestia pioneered with data availability sampling. I think that's very novel.

But I think at that point, anyone can like validate very easily.

Yeah, you can validate the header of the blocks and enough of these data availability sampling clients can essentially flag that an invalid state change has happened if you need so. And so there's, I would say, as this, as the entire blockchain industry is continuing to mature, I think we're just finding out better ways to do things. And I think even with me, like, you want to be told, my point of view is that you're wrong. And you should be less wrong over time. Um, and I think the Ethereum community rightfully innovated on a lot. Uh, they were one of the early, uh, networks to proof of stake. They were the first blockchain to enable Turing completeness, to enable smart contracts. But they are unrelenting and changing their mind about the SolarStaker, and maybe they're right. But I think that is the thing that is today holding them back to actually more mass adoption and why they'll probably lose to high-throughput blockchains because they are explicitly focused on these solo stakers.

Say you wave a wand and you can change that.

What—

there's a lot of things that Ethereum can change, like sort of backwards compatibility issues or, you know, And then so it gives rise to potentially someone like Monad to kind of from the nowhere, like from scratch, like just improve the EVM. But Ethereum can't necessarily do that. So you have to work within certain constraints. But what could, if you're in a position to change, like perhaps a solo staker, but like what specifically would you change and how do you think that would manifest downstream?

I'm a big fan of what Mac Reznick has been pushing. Where it's kind of multi-leader, make it so that the solo staker can be run on like the last two generations of like modern computing, which is like, say, a MacBook. You would actually get a lot more out of Ethereum if you could just do those things. The multi-leader, I think, is probably a separate conversation, but very intellectually interesting that I think most blockchains will have in the future. It's just kind of keeping pace with modern hardware requirements. It's kind of what Solana has said, like compute generally over time will get better. You'll go from 2 cores to 4 cores to 8 cores to 16 cores, and that will just generally get faster and better over time. And then generally your internet speeds will get better over time. You'll go from 1 megabyte to 10 megabytes to 100 megabytes to 1 gigabyte to 10 gigabytes. It's just kind of the progression of technology. So the hard part for the Ethereum community in my mind is it's kind of stuck in a rock and a hard place because it's almost trying to be competitive with Bitcoin and like the moneyness. And I think generally Bitcoin has kind of won that. And then it's also trying to compete on the smart contract application standpoint. And generally it's not as high throughput or low latency as a high-throughput blockchain. And so I think making sure you, you know what game you're playing— and Santi and I had a long podcast talking about making sure you know what, what game you're playing and how to actually win that game— I think is very important. And I think somewhere along the line, Ethereum just got distracted by trying to have two priorities, which was money and smart contract platforms. And I think that lack of focus has really kind of put Ethereum in its spot today where it's perhaps slightly worse money than Bitcoin and slightly worse smart contracts than the high-throughput blockchains.

Yeah. And, you know, I guess how much probability would you ascribe to this change happening in Ethereum? You know, very low.

Yeah.

I mean, it's hard, right? I mean, you could, I think imagine like, you know, as product gets decentralized and there's different contingents within the Ethereum ecosystems, like it's quite hard to coordinate that. Which makes it even all the more impressive to your point that they were able to transition to proof of stake and implement EIP-1559. I mean, that as a whole, to your point, like the airplane's flying and making these transitions might be one of the greatest, most Herculean like human coordination events ever. It's just pretty impressive, but it presents challenges from implementing these changes. I mean, it is, to me, it sounds like ideology is taking precedence over practicality.

Yeah. And that's the hard part. It's not that there are many, many, many brilliant people in the Ethereum community. There's a lot of dollars that have gone into making Ethereum a better ecosystem and continuing to advance Ethereum. It's not dollars, it's not the people, it's the desire to change. And that is actually the thing that I think is holding them back. And it is, in my mind, an ideology thing. And again, if you want to say, hey, we're prioritizing solo staking and that is our bread and butter, that's fine. I just don't think you're going to have the market cap that you want long term. And I think more than, more than a kind of ideology, a lot of people just want to make money. And so if you're not constantly pushing at the forefront, I just don't think you're, you're going to, you're going to lose market share. And I would say that areas that you don't.

So maybe Logan, just rounding out the discussion on solo stakers, I'm curious, what reasonable argument do the hardcore, let's call them hardcore Ethereans, stick to? Because I believe that when you look at what actually happens is that very few, there are very few solo, like there's a natural tendency for centralization and like people delegate, right? And so is there a compelling argument that you might give credence to? Like, okay, this actually would be a reason why you would want to preserve that feature?

I mean, I think generally people incorrectly state the amount of Ethereum nodes to begin with. The big thing that matters in my mind is the full nodes. On any given day, I think Ethereum has between like 4,500 and 8,500 full nodes. So that actually have a copy of the ledger. Generally people incorrectly state that it's like millions. Solana is about 3,000-ish today, or last I recall. Then SUI, Aptos hover around 100 or so. I think Monad and Sui are kind of in that range as well of 100 full nodes. Full nodes. The argument to be made is, hey, if these are truly decentralized systems, if you run, if it costs a node $3,000 or $5,000, maybe the average person cannot actually afford to run a one of those nodes. And so we're going to purposely limit, uh, the hardware requirements or the physical internet connections, the bandwidth requirements, because we want more and more people to participate. And I've always found that noble. And like, I think again, like, if you're opinionated in your design choice, that's fine. I, I just don't think that is ultimately going to scale as hard as the systems that lean more into increased hardware requirements. The argument though that I've always kind of pushed is, hey, theoretically, if you have $5,000 of hardware requirements and a certain amount of bandwidth speeds, if the true limiting factor for people to run a node is that they— is cost, if you reduce that, um, uh, the cost by 10x, uh, to $500 then you— my, my assumption would be you see 10x number of the nodes join the system because now you reduce the cost by 10x. But when in Ethereum, and specifically Solana's case, like, that's not true. Like, Ethereum does not have 10x the amount of full nodes that Solana does, and Solana's node costs are 10x more than Ethereum.

And so that's just the hardware though, not the bandwidth.

That's just the hardware. The bandwidth costs, I think, right now are not a bunch, but over time there'll be more and more and probably one of the most costly aspects of these blockchains.

But you could imagine bandwidth. And I think I credit you for this realization that I had, which is bandwidth is over the next 10 years, probably a bigger cost item than the hardware.

Bandwidth is going to— when you actually crank these systems to like their full capacities, bandwidth actually gets very expensive. So you'll probably locate the nodes in areas of the world that have lesser bandwidth costs. That will probably take some time to play out. But when you're doing like tens of gigabytes per second of throughput, you're going to, you're going to look for ways to reduce your bandwidth costs. And so ultimately, I think the argument is sound. It's just, you are explicitly making the choice to prioritize one certain group of people over, in my opinion, kind of the broader masses that actually want scale. And that's a fine decision to make. Again, I think that's going to be reflected in your market cap long term. If you're not actually trying to do scale, you're, you're going after a much smaller market segment. And that is okay. But I just don't know if the Ethereum community wants to sacrifice their market cap for this ideology. We'll see.

Feels very limited in their thinking, whereas maybe rooted in this, you know, we keep referencing the scalability trilemma and we think of it as a kind of like binary trade-offs. And whereas in reality, and this was my core argument in the discussion that I had with Vitalik, and all credit actually to Mert for this, when he actually said that trilemma is actually— we should, we should reconsider it, um, because after a certain degree of central— after a certain degree of security and threshold of decentralization— and we could argue all day long what that means— but the key is if people are willing to join the network and build on top of it, then that's probably your proof.

Yeah.

Then after a certain threshold of those two parameters, then the most important thing is throughput, because then that— if you have really high throughput that means you have more applications, more applications, more fees, more MEV, emphasis on non-toxic MEV, which needs to be fixed, then that really, that is the flywheel that keeps this going, right? Because that is, you got to look at the unit economics of the validator. And if broadband is going to continue to increase, then, you know, you're going to have to make that up with top line. And that ultimately is if more applications get built, more demand for block space, then they can make more fees and then recoup the cost faster. And it becomes a profitable, like, you know, business to run. If you, if you think of it as a business, which it is, it's a very profitable business.

Yeah.

If you, if you have a validator or stake, you, you can make quite a bit of money.

Yeah. And people in Ethereum, and this is the last thing I'll say in this, in this part of the segment is when you hear, you know, the maxis say, but Solana is subsidizing it and, you know, like all this stuff, which is, you know, I don't actually, but, you know, I'll just acknowledge it and preempt it because it's not true.

Logan, if you were Vitalik today or you were crowned CEO of Ethereum and given 2 years to fix Ethereum, do you like— A, do you think that it's fixable at this point? And B, like, what would you do?

They're in a hard spot. They are in a hard spot. Um, because if you say, hey, I want to compete on the blockchain kind of smart contract more, I mean, it's not only just increasing throughput and adding more cores, it's also all the other service providers in the stack. Uh, when you go from low throughput to high throughput, you need faster oracles, you need faster, faster, um, um, You need kind of like what Mert is doing with better RPCs. There's a variety of things that—

Well, that's not— but don't you have like you have Infura and Alchemy and QuickNode, you have, you know, like Pyth is doing a great job on the Oracle side. Maybe if Ethereum made some changes, Pyth could do more in Ethereum.

Like, I'm just highlighting that, like, even if you say, hey, we want to start doing high throughput, we want to start doing parallelization. There are also many like other factors that I think are hard to do. And so maybe it is best for Ethereum to just continue to prioritize the solo stakers and say, hey, we're trying to focus on just being the quote unquote most decentralized, continue to broaden more full nodes on the network instead of going from like the 4,500 to 8,500. What if you went to 20,000? Or 50,000 and just really trying to push that extreme because it doesn't seem to me on the surface today that even if they wanted to hard pivot, it would be very hard in my mind to continue to catch up to the high throughput chain just because I think they already have such a massive head start and the lack of tech debt. I don't know. It's, it's a hard decision. They're a rock and a hard place.

Does that make you more excited about Monad than Ethereum?

Yes.

Nice. Okay.

But is that just because you're an investor in Monad, Logan? Like, like, like if you were, I noticed you didn't mention like Bearachain. So there's like 3 hot things right now for the new chains. It's like Bearachain, Monad, and MegaETH, I would say. I'm curious, like—

We're not investors in any of those.

Full disclosure, investor in all 3.

Ladies and gentlemen, sit down, Santi, sit down. I'll get out.

Out now.

Yeah, he's just gonna go home. Um, I am home. Logan, are you— I have the power to boot you out of this. Uh, Logan, are you guys an investor in Moned?

No.

Oh, you're not? Oh, interesting. Okay, never mind, I take back that.

I mean, we're very close with the Moned team. I think I, I like what they're doing from a technology standpoint. I think if you go— if you could go back in time I think Ethereum would look much more like Monad than it does what it currently looks like today. It would kind of be a blend of Ethereum and Solana where you focus on high throughput. You also focus on parallelization. I just think both of those are going to be requirements.

Is parallelization the most important thing, which Ethereum, like, is, you know, it's not backwards compatible?

To me, Um, I, I think the most important thing is the lack of fragmentation. That I, I've actually started to appreciate more the social coordination than the technical stuff. I, I came at this very from a technical lens, but the more and more that I kind of like dive in, the technical nuance does matter, but the social coordination of it all is also extremely important. The more that you fragment the different state, I think it's just harder to coordinate. You have different standards. It's not that Optimism, Arbitrum, and Base can't do a shared interoperability solution between them. They definitely can from a technical standpoint. Now, do they want to? Are they incentivized to? That is the hard problem. And when you have one giant blockchain, sure, you are explicitly making the trade-offs that over time we are going to increase the hardware requirements, we are going to increase the the bandwidth requirements. But when you use this blockchain, you're implicitly saying, hey, I'm opting into the shared set of standards that allow me to coordinate with everybody much easier. It's like, there's a bunch of different exchanges around the world. And they don't really talk to each other. And you have all the different antiquated banking systems on the backend. But if you could collapse those all and make them one giant kind of like pricing engine. That'd be rather interesting. And so does that dictate that the primary use case is going to be financial and like composability is the most important thing and therefore, you know, I think financial, I mean, at the end of these things, at the end of the day, these are just accounting ledgers keeping track of who owns what. I think it gets a little bit more expressive with things like rollups or ephemeral rollups. Again, like the, the details of those matter. And like, I think where the state actually lives matter. But then if you wanted to say, do an ephemeral rollup where you do an arbitrarily large amount of compute, you run a neural network in an L2, and then you settle whatever state back to the L1, and that state lives in the L1. That's like a new novel, interesting idea. I don't know, like what the unique product case is for that, but when you now unlock arbitrary amounts of compute, that could potentially be interesting.

Yeah.

Logan, can you, uh, can we shift gears a little bit and can you tell me what's happening in the Move ecosystem? Because I've got two charts up. I've got the chart of SUI and I've got the chart of Aptos. And you mentioned both of them earlier and SUI's up 280%, it looks like on the year and like over 100% on the month. And then Aptos is up like 50% on the year and, you know, 20 or 30% on the month. So What is actually happening on the ground inside of these ecosystems that's causing this divergence?

From a technology standpoint, there's some slight nuances. I think that's probably less important. I think for the new communities, what's actually really important is kind of taking the playbook that Solana did where they really focused on hacker houses and they were just available. Like being on the timeline. I've talked with quite a few of, well, both of these teams now, and my explicit feedback to both of them is like, be available, be on the timeline. And it doesn't always seem like high ROI because you're just in like nerd battles or like talking with different people on Twitter. But I think trying to explicitly tell why you're differentiated, why, why you matter, I think the SUI team, at least the feedback that I've gotten from engineers, is that they're very responsive. People really like Sam Blackshear. He is very approachable. He talks— Sam Blackshear is the CTO of SUI. Generally, because I think these platforms, all blockchains, are really engineering-driven, they're kind of developer-centric, you need to have the head engineer be out there, be in the open, answering questions. You now see Avery on the timeline, the CTO of Aptos, doing something similar. To me, it ultimately boils down to a race for engineers, which I think now, if there's not another 10x improvement or 100x improvement, I think Toly said this on the timeline and kind of got some flak. If there's these more minute details of like slight engineering changes, but it's not as big of a jump from Ethereum to Solana, then it's like, how can you actually build the engineering and developer community from the early days? And I think the Sui team has just been on the ground floor and has done fairly well at that. I think the Aptos team, perhaps, and this I think goes to every blockchain, when they initially come into the ecosystem, they kind of say, hey, crypto is small, let's run the Polygon playbook, let's do the big partnerships. Let's try to do the Googles, the Meta, the Facebooks. And generally, we haven't seen that work. It's generally the zero to one with the crypto native, and then you continue to kind of expand and potentially do things more over time as you actually get adoption. And so I think Sui has kind of focused maybe slightly earlier on like the crypto native community, and I think Aptos is starting to do that as well. Yeah, I would say it's probably less of a technical thing, in my opinion.

Yeah. Do you think it's too— do you think it's solvable when, like, when someone goes down, like an L1 or L2 goes down that rabbit hole of, like, focusing on the big enterprises? That's obviously not the winning strategy. I think it's been proven out. So then is it too late to focus on trying to win the crypto-native crowd, or has that brand already been supply?

I don't think so. I think, uh, I mean, we always say this, but it still feels early. Like, if you look at like the applications that actually make revenues, uh, we don't do that in crypto, but if you actually did, uh, you'd be surprised.

And so that's right, Logan, there's one group who does, you know who does that? Yes, Blockworks Research, baby!

And shout out to Blockworks Research, uh, you guys are actually doing a lot of really great things. Uh, I'm not just saying that. I, I think the dashboards that you guys have put— pushed out, the metrics that you guys are starting to really hone in on, it collectively— going back to why I'm interested in integrated blockchains is because that shared standards that Blockworks Research is pushing really helps move everybody forward. Um, so yeah, I, I think, um, It's not too early to focus on crypto natives. I think Solana has a head start, but you have to focus on not only attracting, uh, kind of quote unquote the crypto natives, but the engineers. At the end of the day, these are engineering platforms. You need engineers to build on your ecosystem and why people will actually use your blockchain long term versus others.

What, from a developer ease, you know, onboarding, tooling, how would you compare kind of these high-throughput blockchains?

Uh, from, I think people like Move programming language. It's, I mean, now I guess there's three, there's Move, there's, uh, Solana's programming language, which is generally Rust. And then the EVM with both Monad and Sei. So the EVM still has quite a bit of bugs, or I won't say bugs. The EVM has potential flaws where even if you're the best engineer, we still see general hacks from time to time. And I think maybe it's not the engineers, maybe it's just the underlying programming language. Rust is a pretty predominant programming language. And so I think it's still fairly well understood. Sam Blackshear, the CTO of Sui, was actually the main engineer, as far as my knowledge, at Meta or Facebook that really pushed forward Move. And so Move perhaps may be the next iteration or the final iteration of kind of blockchain smart contract design where it implicitly bakes in some of these design principles that makes it harder for engineers to write bad code or to write flaws in their logic. Um, Solana uses, uh, compiles down, uh, their into LLVM bytecode. And so it's a little bit more program agnostic. You could use Python, you could use Rust, you could potentially even use Move in the future if Move does become the predominant programming language. Um, I think it's more of like a safety aspect, and Move is probably the highest up there.

Yeah, I was going to ask you if Solana could, you know, transition or just implement Move. And it sounds like— yeah, I've had some conversations, but it sounds like it could.

Do you guys have, uh, any spicy opinions that you want answered? Or, uh, as we round up the podcast, uh, Spicy questions.

Well, spicy takes. I mean, I don't, uh, I mean, what would make you like, I guess one question would be like, what would make you totally change your thesis, which we've kind of like teased out here in some capacity, but I just want adoption.

I mean, like, look, at the end of the day, if it's high throughput or not, if there are people where they find immense amounts of value, like, let's go. I just like, and I saw that in the early days of Ethereum. That's why it got me excited. Like people were building. You had DeFi Summer, uh, there was all these promises of building really interesting applications. My personal experience, and the experience that I think others experienced as well, was just higher gas fees as people wanted to do more and more interesting things, which personally made me not want to use Ethereum. Because even if you're a multi-millionaire, paying $500 to $1,000 for gas costs, it sucks. I mean, if your economic value is greater than the gas cost, sure, you'll do it, but not very many people have that opportunity. And so I really approach this from a product standpoint. I will change my mind if Ethereum starts building better products.

Yeah. Category-wise, anyone, any particular category that you're just bearish on, just to think, does make absolute sense?

It's a good question. I mean, games is kind of like the easy one to kick.

Yeah, it was like dancing on the grave, which I coincidentally think it's wrong take.

But yeah, TBD, like if games work, I think the timing of these things work or are hard. Games will work. Like it to me, it also seems inevitable that we'll have like a full Ready Player One metaverse metaverse, uh, style world. Um, as a venture investor though, I just don't know how to underwrite different games. Like, is the game fun is the criteria, and I don't know if I have any unique alpha in underwriting games. I think where our— and Frictionless is uniquely well suited— is focusing on the things that will continue to compound. I think you can actually reason about the space much more than people care to try You have to probably be slightly technical, but you can really get into the details of these things. And then generally, once these things start working, people underappreciate the nature to which they will continue to compound in the future.

How worried are you about toxic AMV and that perhaps not being fixed to the degree that we want it to? Because it is, it is perhaps, I think in my mind, the thing that might— that was my counterargument to Solana not working. It's like, it's not that toxic AMV doesn't happen in other chains, but because the activity is happening in Solana, it might just create a perception problem that can, can hurt it. Um, I don't think it totally derails the value proposition of Solana, but it just can set it back.

Yeah. I mean, as you mentioned, I think because Solana is the first high-throughput blockchain, it's running into a lot of these issues like the spam. Like Solana was the first to— once you go from high fees to low fees, there's just a new set of problems that ultimately arise from the engineering side. I think as you go from like AMMs to order books and go from like 10 TPS to 10,000, uh, you can do more interesting things, but, and you can get more trading volumes. And Slana is kind of like front-running, uh, or spearheading all these like unique insights. Um, the toxic MEV is difficult. I think one way that we think ultimately potentially can help mitigate this is just doing different order flow segmentation. We invested in a company called DeFlow. What they're ultimately doing is really focusing on the maker side of the order book. So a lot of people in the industry only focus on the taker side of the order book. So you can think of that as like Jito, Flashbots. They're taking liquidity. I think over the long term, as volumes continue to increase, you'll have more and more market makers want to come online and start adding liquidity to the markets. And this is where order flow segmentation actually becomes more and more important. Uh, you want to know, hey, is this toxic or non-toxic? Um, and then perhaps in the end state you could actually create markets for toxic versus non-toxic flow. Um, I don't know, like, we'll see how they ultimately do and how Dflow does. Like, we're obviously very excited about them. I think this is going to be an iterative process because, uh, the toxic MEV is still, I would say, be worked on. But again, this is not an issue to Solana. I think just because Solana was the first high-throughput blockchain, they're speedrunning this faster than anybody else.

That's right. Yeah, absolutely. I think perhaps allowing dApps and also users, which will not be as sophisticated, like just right now it's the, the block proc— like MEV is captured because like the block producers have like ultimate control of like how fees are being shared and whatnot. But as you might create like a different layer there that can allow for that flow to come up, it might fix— like I've seen different approaches, obviously excited about DeFlow and investor there as well, but there's, I've seen other kind of iterations at this. I think smart people are trying to fix this. And so it will be fixed probably.

I feel like Santiago has kind of changed his mind in the last like 8 to 12 months. But Janno, how has your mind changed? Maybe flipping the script and the book on you. Have you changed your mind on high-throughput blockchains over kind of Ethereum?

Yeah, I've got more excited about— I don't know what we're calling them these days, high-throughput or monolithic or integrated, but from a user's perspective, I mean, I tweeted this out the other day, like, I think I was on— I forget the wallet that I was using, but I think I was using Fantom, but I was using Fantom on Ethereum, or I think I was trying to bridge— basically, I was trying to get money on the Polymarket. And it was like, okay, I need to send this token and I basically needed to get a token. I think it was USDC on Polygon into Polymarket, but then I needed gas and then I couldn't get like basically the user experience was an absolute nightmare. And I posted this on Twitter and then the natural reaction was just use Solana and it's become a bit of a meme at this point. Like it's a bit like, you know, kind of meme-y, which I think is good marketing, but like it really does work better. And so from a user's perspective, I tend to like it. Now, I'm probably more optimistic about some of the experiments happening in modular Ethereum land than you are, just because I, I do also spend a lot of time in that world. And like, there are— to me, it doesn't matter as much about the technology and it matters a lot about the founders building in that area. And there are still a lot of incredible tenets.

They're very smart.

Founders building in that area. As soon as that changes, I'm bearish on it. But as long as there are really brilliant, amazing founders building in that ecosystem, I will remain optimistic about it.

I think that's totally fair.

Yeah.

Logan, I'm going to count on you to tell us when that transition might happen and when you flip bearish and bullish on another chain. So please let us know.

The only— I think generally I've done probably the most technical podcast with the CTOs across the industry. The only person team that I have not talked to is the parachain team. So, Smokey, get on there, goddamn it. If you want to do a deep dive, let me know.

Okay, I'll make that happen. Smokey, I'm putting you on the spot here. Or Papa Bear. But anyways, Logan, very glad that you were able to take time.

Thank you guys.

From the act of in Asia, we spent some time there, so it's always a treat to have you on and you know, have you come on and give you— give us your brain dump. Because as I said, I don't think there's a person in the space that is as good as you to compile all these different concepts in a digestible format without compromising the essence. So really appreciate you coming on, sharing your insights. We'll have to— we'll have to have you on more often.

Yeah. Long, better products. At the end of the day, we need to build cool things that people are going to use. There's a lot of semantics in terms of the underlying technology. But let's just build people things that people will use.

Products, not crypto products.

Products, not crypto products.

Let's leave it at that. Thanks so much, Logan. Appreciate it.

Thanks for watching today's episode. Just wanted to quickly mention that the wait for crypto's killer app wave is nearly over thanks to today's sponsor, Anoma, the universal intent machine. To stay up to date with Anoma, you can follow them at x.com/anoma or sign up for the newsletter at Anoma.net. Thanks again to Anoma for sponsoring today's episode.