In this episode, I had the opportunity to sit down with Eli Ben-Sasson of Starkware. If you listened to my last episode with Su where we talked about the scaling approaches of different L1s blockchains compared to Ethereum, I argued that L1 blockchains do not scale and that the only way to create true scalability is to perform all of the computation off-chain and only post the results of that computation on-chain. STARKs are a technology that allows huge amounts of computation to be compressed into succinct proofs that anyone can easily verify. Our goal for this episode was to create the most approachable and comprehensive audio resource on how STARKs work and how they will scale blockchains in the future. We start by explaining inclusive accountability and the true meaning of scalability. Then we dive into STARKs, how proof systems work in general, and where they fit into the context of unbundling blockchains. Next, we use DYDX as a comprehensive case study to learn about the StarkEx system, before diving into StarkNet and its tradeoffs to StarkEx. Finally, we talk about StarkWare’s programming language Cairo and how the different costs of proving, verifying, and storage are going to scale into the future. If you’re a developer, you should also gain a very good idea of the tradeoffs between building on a regular L1 blockchain, the general-purpose StarkNet blockchain, and a StarkEx application-specific chain. Enjoy! 00:00 - Intro 2:03 - Eli Ben-Sasson background 5:29 - The goal: Computational Integrity 10:46 - What is true scalability? 14:19 - What are STARKs? 17:06 - Proof Systems / How can we trust them? 20:18 - How are the proofs generated? 26:36 - False-positive proofs 29:01 - How validity proofs fit into scaling blockchains 34:12 - DYDX & StarkEx: How do we get from 300,000 gas to 300 gas? 43:33 - How can DYDX provide instant finality? 46:20 - How long does a batch take? 47:52 - Differences between Fraud proofs & Validity proofs 50:27 - Other tradeoffs using StarkEx instead of L1 Ethereum 52:40 - What if StarkWare stopped making proofs for DYDX? 56:21 - StarkNet: Differences between StarkNet & StarkEx 59:12 - StarkNet roadmap 1:02:46 - Cairo 1:07:22 - Why a new programming language? 1:12:03 - Early feedback & steps to enter StarkNet as a dev 1:13:50 - How does batch size affect proving cost & verification cost? 1:17:42 - What can be done to lower these costs in the future? (Verification, proving & storing data) 1:32:27 - Understanding proving cost 1:36:28 - End Listen to conversations between Su Zhu, the CEO and CIO of Three Arrows Capital, and Hasu, an experienced crypto researcher and writer. Together with occasional guests, we explore the transformative nature of trust-minimized currency and financial services. SUBSCRIBE to the Podcast Apple Podcast https://podcasts.apple.com/us/podcast/uncommon-core/id1517659188?uo=4 Spotify https://open.spotify.com/show/3vuV292Him90EjQ5YL4XIw Youtube https://podcasts.google.com/feed/aHR0cHM6Ly9hbmNob3IuZm0vcy8yNTc4ZDVhMC9wb2RjYXN0L3Jzcw== Other https://anchor.fm/uncommoncore FOLLOW your hosts Su Zhu https://twitter.com/zhusu Hasu https://twitter.com/hasufl Our homepage and mailing list https://uncommoncore.co/podcast/ Transcripts, if available https://uncommoncore.co/blog/ FOLLOW the guest Eli Ben-Sasson https://twitter.com/EliBenSasson Learn about StarkNet: https://starknet.io/ Learn about Cairo programming language: https://www.cairo-lang.org/ Learn about ZK-STARK math: https://starkware.co/stark/ StarkNet Discord https://discord.gg/uJ9HZTUk2Y