# Igra Labs | The Minimum Rollup Constraint Theory - Pavel Emdin | ETHDam III - 2025 - Channel: [CryptoCanal](https://streameth.org/cryptocanal) - Date: 2025-10-07 - Duration: 20:46 - Watch: https://streameth.org/watch/yt-zz5IF3p4YlE - YouTube: https://www.youtube.com/watch?v=zz5IF3p4YlE ## Description Welcome to the 3rd Edition of ETHDam, hosted May 9–11, 2025 in Amsterdam. This year, we brought together the brightest minds in privacy, security, and AI for a unique 48-hour hackathon + conference combo. 🌷 https://www.ethdam.com// 🌷 ------------------ Igra Labs | The Minimum Rollup Constraint Theory - Pavel Emdin | ETHDam III - 2025 About the Speaker: Serial founder and operator with over 20 years of experience ideating and delivering complex software projects. Until 2017, worked as a teams leader at two Berlin unicorns, Auto1 and DeliveryHero. Since then, building and growing web3 companies. Currently leading Igra Labs, building Igra Network β€” an EVM-compatible programmable layer (L2) on Kaspa blockDAG. Follow Pavel and Igra Labs: https://x.com/emdin https://igralabs.com/ https://x.com/Igra_Labs ------------------ About ETHDam & CryptoCanal ETHDam is powered by CryptoCanal, an education and events platform rooted in Amsterdam, expanding into Rotterdam and ZΓΌrich. Keep up with us to see updates on future events: https://www.cryptocanal.org/ Follow CryptoCanal on X: https://twitter.com/CryptoCanal Join CryptoCanal TG Community: https://t.me/CryptoCanalCommunity Join CryptoCanal Discord: https://discord.com/invite/XJVjpCqQBz CryptoCanal unites crypto enthusiasts committed to making a positive impact. Unapologetically political, we prioritize education, events, and services while championing cypherpunk values like privacy, sovereignty, and censorship resistance. ------------------ πŸŽ₯ Credits: Intro / outro by babyPRO - https://babypro.art/ ETHDam Photography by Paulus – https://concretestate.eu/ ------------------ Special thanks to our partners who made ETHDam possible: 🌹 Hackathon – Bouquet: Oasis Network https://oasisprotocol.org/ 🌷 Hackathon – Petal: Circles https://aboutcircles.com πŸ’› Conference – Gold: Zano https://zano.org/ Dash https://www.dash.org/ Bitvavo https://bitvavo.com/en 🩢 Conference – Silver: Igra Labs https://igralabs.com/hero πŸ’› Conference – Copper: Lido https://lido.fi/ DeTrip https://detrip.travel/ Cake Wallet https://cakewallet.com/ The Grid https://thegrid.id/ Calimero Network https://calimero.network/ 0xbow https://0xbow.io/ Mina https://minaprotocol.com/ JobStash https://jobstash.xyz/ Cyber Capital https://www.cyber.capital/ POAP https://poap.xyz/ Acronym Foundation (Supported our Top 10 Hackers) https://acronymfoundation.org/ 🌱 Sponsor: EF Ecosystem Support Program https://esp.ethereum.foundation ------------------ 0:00 – Opening Remarks & Why Decentralization Still Matters 1:45 – Fragmentation, Regulation & The Case Against Centralized Rollups 4:27 – Rollups Extracting Value from Ethereum 5:20 – The Quadrilemma: Why Interoperability Is the Missing Piece 7:03 – Why Shared Sequencers Enable True Communication 10:01 – Composability, Unified Asset Layers & VM Diversity 12:56 – Constraints: Why We Can’t Have Nice Things (Yet) 15:00 – Why Kaspa’s BlockDAG Might Be the Right Sequencer 18:12 – A Call to Expand the Modular Rollup Conversation 19:43 – How Kaspa DAG Can Help Ethereum Scale ## Transcript Welcome to Ethan to Ethan to Ethan to Ethan to Ethan. Yeah, they already told me my talk was too technical unfortunately like I I think it was super high level uh but yeah uh it was actually designed for beer market and Ethereum price back on like uh above 2,000. So I think we're sold back and we should start talking about like getting more VC money instead of uh speaking about better tech but here we are. So I'm paddle from eager labs. We're building eager network which is EVM uh zero knowledge based rollup uh on top of caspa blog duct decentralized sequencer. Uh our devet was recently launched and test net is coming this summer. Uh we have already 15 teams building on igra. So what's this talk is about and uh why we're here. So I think we're here because we uh want to build something decentralized. Uh I think it still makes sense. Uh recently we see big shift uh not just from uh uh making sense uh point of view but also from uh regulatory landscape actually legal requirements for teams building decentralized systems currently uh gently nudge us into the uh being decentralized uh I mean for all every team who is building anything related to uh crypto uh they know that you don't really want to control people funds and you should be decentralized just for the sake of uh compliance. It's for good and for bad. So it's but also yeah you know uh uh why would we build another uh centralized database like keep people asking the same question. I think that actually uh layer 1 and layer twos who are centralized in a way they have their own market. It's fine. It's just it's not uh super interesting. So again let's uh uh uh I try try to explain today what I think is uh minimal constraint for rollups is and why based rollups on decentralized sequencers decentralized shared sequencers can be solution for fragmentation of programmable blockchains. So it's not not only about uh scaling Ethereum, it's also about like uh building the better uh designs for uh programmable blockchains uh in in general. So this is yeah and uh let's talk about roll-ups. Uh rollups is a scaling solution for Ethereum and actually Ethereum fun found its product market fit as a settlement layer. You can see just by how many different teams built uh on Ethereum. I think it's a great match like you have finally uh real product as an infrastructure. I'm not really sure I understand uh ultrasound money narrative for Ethereum uh because Ethereum as a uh infrastructure make much more sense. It's just works. uh you again you can see by amount of uh uh roll-ups which are already built on Ethereum and roll-ups initially made a lot of sense as a scalable execution uh environment but they completely doesn't make sense as a standalone consensus because like building your consensus is hard and Ethereum took several years to get where where it is now. Same goes for several other projects I mean also Bitcoin like to get this credible neutrality. So why should we build it? But still that's that's what people did. They built uh L tools with centralized components in this especially uh centralized sequencers. And this is what we have currently this is this famous tweet which states the uh like inform us that it's actually a good thing that L2 base actually paid 1% of its revenue like back to layer 1. Uh I don't think like I would argue that it's it's it's uh because B or any other rollup is malicious and diverge from uh Ethereum just because they had it. No, just because it's there is no better solution for them than to build whatever they build currently. I will touch it in a second what I exactly mean by that. So that's uh this drama around uh extracting value by L2s from L1. I think it's it's it's like kids growing up in parents' house and not paying any rent despite having their own salary. Uh it yeah it's kind of uh you can apply social pressure to this to solve it maybe but uh it doesn't scale the solution. I believe we should have combination of uh different incentives. We should build architectures which uh gently enforce uh systems, rollups, whatever just to behave uh in rational and eventually uh good way like this the sharing is caring kind of things not uh not be isolated but this so like build being decentralized should make sense and fragmented shouldn't be uh why but why we think like why I think this the whole thing happened is because we don't have uh blockchain trlemma, we have quadrillemma with interoperability being the central ingredient, the central missing ingredient. Uh because without interconnectivity between systems, you just everyone's go stray and start building uh their own thing and we don't have any alignment in the in the ecosystem. So what's uh but like what's missed like why why interpretability is a missing ingredient? First of I think because it's hard to build like consensus is hard to build as we uh as we all know to build like critical neutrality consensus decentralized in terms of valid validator diversification is hard problem and then synchronizing between them it's even harder like because you have different ecosystems different everything so syncing between them is a hard engineering problem. So this is uh this is one thing and also like uh this uh uh like I I keep saying like people people telling me like why uh why you think it's uh Ethereum is not replaceable actually like why can't we use it anything else for communication layer essentially rollup is just a web service and Ethereum is AWS we can replace AWS with whatever we like and then build on this foundation Well, no, you can't because again, Ethereum is is a brilliant uh product, I believe. Uh and I still have backs in Ethereum and I'm build building on completely different infrastructure now. Uh so I'm kind of my bucks now. So, but without coordination mechanisms, it's just not not not going to scale. So, this is uh so Ethereum make a lot of sense. uh we should uh improve it but without coordination without intercommunication it's not it's just not interesting uh not going to work so this uh there are also several different uh credible consensuses on the market there is obviously bitcoin but yeah uh and there is another solution which I will tell you about like in in a second so just saying like what's the like how would you create this communication layer between rollups. I think it's just here like when we talk about modular architecture for uh programmable blockchains specifically Ethereum compatible blockchains but this all makes sense for other chains as well uh consensus is actually a sequencer like your sequencing defines your state. So when you and and that make a lot of sense to externalize your sequencer and make it shared and that's what's important but then unfortunately what we see in all existing K tools instead of externalizing sequencer and make it shared they have it centralized and it's a single point of failure. It's really weak solution by any regard and this is what we car currently have like we have silos we have islands which connected with very fragile very non-reliable bridges uh it doesn't scale it's uh permissioned it doesn't make any sense because there is no other solution unfortunately and all this obviously islands are like kind of drifting away so they because they don't have like any any any any proper communication layer and not just communication layer they don't have a way to communicate one with each other. Like think we we it's crazy. We as an industry like web 3, we kind of we we have web servers essentially if you think roll up as a web server and you can you should wait seven days for another web server for you to confirm your call or you should collocate your APIs on the same machine so you can call them because that's that's that's how currently it works like if you want to call smart contract from your smart contract they have to be collocated on the same on the same rollup. it's not good. Uh so there is a solution we think this there is a solution in the form of based rollup uh on shared sequencer. So this way uh every rollup actually which uh share the sequencer can uh like effectively communicate with another rollup share even state uh they both they all rollups on top of the same sequencer will inherit its uh livveness and security. So again it's it have to be strong uh strong uh fast and credible sequencer. uh but then uh once you have it it's uh like you you can have several really really cool features uh in your ecosystem. So like the coolest thing I think about based rollups again based rollup it's a rollup which use uh another base layer as a sequencer. I think that's good enough uh definition. So when you share a sequencer you can have more or less out of the box uh atomic composibility not just syn synchronous atomic composibility meaning one rollup can call another rollup in the same transaction and this several calls will fail or execute simultaneously like in atomic manner. this can create like some really really cool applications uh and this can create actually a network effect which we're all looking for in uh current programmable chains uh uh world. So uh also you can build if you want you can build uh unified asset layer similar to Salana. So every rollup will actually reference the same assets which also uh def fragment the ecosystem uh helps to to fight uh fragmentation of liquidity which is kind of a good thing. Uh you can also plug in uh diverse VMs. So think that your rollup for example uses Solana VM or whatever vasom or uh I actually talked to the real guy who want to build uh on lisp uh and think that this lisp rollup can call another salana rollup in the same transaction. So this is this this is a solution for fragmentation if we shouldn't look elsewhere. That's uh something which can really move industry forward because I'd argue we stagnate a bit. uh if you look at layer 2 TVL it's kind of stucked right you see it was it was a really strong growth uh in first years like 20 uh 2022 2023 and then 2024 down like why I'd argue like why because it's not like we're completely out of money there is still TVL but it's not growing anymore uh yes it's partially because of VC money they don't come anymore for rollups but it's also because uh there's nowhere to grow like all this L2s they not super interesting it's kind of like you can okay you can live in your small Kraken roll up you can live in your small base rollup or like no disto arbitum or optimism like was using them a lot but again you have kind of you have limits you're in your small kingdom it's not how internet works like in internet you build with each other you cross-pollinate you call whatever you like you take your data in one second and it just works for you. So why should we have something different in web 3? Like can can't we can't we do like any better? So that's uh yeah but that's that's a nice vision right like why can't we build it tomorrow? Why can't we have it why can't we have nice things? Well, because there is a hard constraint in this architecture and this is a sequencing speed for for sequencer for shared sequencer to be proper communication uh and interoperability solution. You should have really fast not just credible but really fast sequencer. It's not just about like uh security, it's also about sub-second finality of your base layer. Also you should fix you should you should have enough block space and yes I said there should be security and credible neutrality all this very important. So for Ethereum layer 1 for example uh there are several initiatives watching them really closely happy to see like these things evolved because to be honest when I heard 5 years timeline uh in Bangkok last year I was kind of astonished not in a good way but then finally like some better like improved uh solutions came up and then like so now latency probably will be partially solved by based preconfost post for Saka upgrade end of this year. Then by dunk sharding the blob space uh will be will be increased. So just more more space to save your uh uh to sequence your transactions and like for interrop I've seen several really interesting uh uh cross rollup messaging protocols like they're there. Uh you also need adoption but yeah adoption it's a social coordination thing. It's hard hard to build. It just should be there. Still I think we're looking uh for at least two three years time until Ethereum can be this perfect sequencer. I might be wrong but it's just my hunch. So I would say we should look look elsewhere and there are several really interesting projects uh on the market. So obviously there are in Ethereum ccentric world there is Espresso, Aztec uh radius, Astria like this really great projects. Uh it's like there's lots to learn but like for us specifically uh Kaspa blog duck stood out. It's a uh blog duck essentially. So it's a block it's a system which built as a direct asylic graph. You can think about this as a very performant bitcoin. It was based on bitcoin script and its core logic. Uh it's proof of work, right? So I happy not so many people here because I thought I could be really burned here on a stick like not so many people. I can fight all of you but yeah it's proof of work and I'm talking about proof of work uh on Ethereum conference. I'm I'm that reckless but we think that it's has a really interesting unique uh properties. Uh it's not just speed although speed is there it's uh recent upgrade just unlocked 10 bps 10 blocks per second and it means roughly converts into five uh uh thousand transactions per second but it's not just interesting like just speed is not interesting uh like my my local computer can do million transactions per second but it's interesting because it's a really credible and secure uh base layer secure consensus so this is so any layer tool which built on top of uh uh blog duck will inherit its livveness 5 5,000 TPS and security today it's secured by one exahash so for the context one exahash is what bitcoin had eight years ago in 2016 so it's not that it's not that far so fingers crossed uh we'll get there this on our blog doc as well and yes it's uh has several other great properties. For example, since there is no complex uh builder proposal separation, miners actually like these are entities which sequence your transactions. Uh it's MAV resilient meaning like you cannot easily extract value. Uh it's much harder. Let's it's resilient. I'm not saying like there is no MAV you can do if you try but it's much more resilient than other systems and uh it opens the way to interesting mechanics like fair auctions for example. So miners actually compete for your transaction, not you paying uh paying rent or uh fee for you know uh for using this specific uh L2 blockchain and also there is uh possibility to use actually miners as oracles. So you can use this uh consensus uh blog d consensus for uh serving as oracle network which is really also very interesting cool design and like for people who like super uh sarcastic uh skeptic and I completely understand that it's three years for three years running in production and it's secured by uh two billion of liquidity. So uh I invite you to research it to ask hard questions about that. We think that its uh security and credible neutrality properties make it excellent shared decentralized sequencer uh at least for us. So I cannot uh cover eager network architecture in deep don't have time but if you like reach out if you're interested I can give you details. So this is uh like re really quick overview. uh and we are currently at like we in the middle of uh very interesting debate about programmable chain architecture and this debate is mostly uh Ethereum ccentric because yeah the whole thing about roll-ups came from Ethereum foundation also based roll-ups idea came from Ethereum foundation so it just makes sense it's still kind of here so we want to expand this debate uh and we invite you to look elsewhere just at better simpler more modular systems with high degree of decentralization because that matters and we believe we can build blueprints that not only like benefit uh again some new ecosystems but they eventually can be uh absorbed into Ethereum itself right so once it's ready uh so yeah we came here to build systems that's decentralized performant secure and interconnected uh and bring value to people yes I'd love to open this up to discussion uh there are lots of open questions here lots of unfinished thoughts. I know uh and I think this uh the kind of conversation will benefit a lot from experiences and different perspectives which we have in this room. So please reach out. My DMs are open like super interested in uh your thoughts and thank you for cannon. Yes. Yes. A big round of applause for Pavo please. Thank you very much. Um um yeah, if you have any questions, please use slido for that. I see you do have one question. I think you already kind of touched about it, but how exactly does Casper DAG uh architecture help Ethereum scale? It's if you use it so high uh high level if you use uh Caspa blog duck as a shared decentralized sequencer for your two and also do settlement on Ethereum with zero knowledge. By the way, I like I left out all parts which touch zero knowledge because it's will be just talk will be just twice as long. Uh but then if you do your proofs on Ethereum, you enjoy from Ethereum security and credibility and you you have super fast sequencer and then you can have uh synchronous atomic composibility and finally have interconnectivity solution which we're all desperately looking for. All right, thank you. Um well that's pretty much it. No uh one more time please thank uh Pavl for very technical breakdown how to scale Aarium. Thank you.