Day 8 — Layer 1 vs Layer 2 (and why everyone talks about "scaling")
Welcome to Week 2. The first week was about building a mental model around the two foundational chains. This week is about the rest of the world that got built on top of (and alongside) them.
The first question you have to answer is the one almost every new participant in crypto eventually asks. If Bitcoin works and Ethereum works, why are there fifty other chains? Why does anyone use Solana, or Avalanche, or Base, or Arbitrum, or Aptos, or Berachain? Why isn't this winner-take-all?
The answer is a tradeoff every blockchain has to make. It is called the blockchain trilemma, and it was named (though not invented) by Vitalik Buterin, who you met yesterday.
The trilemma says that a blockchain wants to be three things at once:
- Secure. Hard to attack, hard to corrupt, expensive to rewrite.
- Decentralized. No single party can control or block participation.
- Scalable. Cheap, fast transactions for a lot of users.
The problem is that at the base layer, a chain can optimize for two of these but not all three. Push hard on decentralization and you slow everything down because every node has to verify everything. Push hard on scalability and you end up with a small number of powerful validators, which compromises decentralization. Push hard on security and you tend to slow throughput further to make attacks impractical.
Different chains have made different choices about which two to optimize.
Bitcoin picked security and decentralization. Its block time is roughly 10 minutes. Its throughput is roughly 7 transactions per second. It is unfashionably slow compared to a modern payments system. That is on purpose. Slow blocks and full validation by every node are what make it nearly impossible to attack at the base layer. Bitcoin trades speed for resilience.
Solana picked security and scalability. Its block time is about 400 milliseconds. Its throughput is in the tens of thousands of transactions per second. It is fast enough to feel like the regular internet. The cost is that running a Solana validator requires expensive, high-end hardware, which means there are fewer validators than on Bitcoin or Ethereum and the network is more centralized in practice.
Ethereum is the most interesting case. It picked security and decentralization at the base layer, accepting the same speed limitations Bitcoin lives with. Then it bet that scalability would come from a second layer of chains built on top, instead of from the base chain itself.
That bet is what made Layer 2s a thing.
A Layer 2 is a blockchain that runs on top of another blockchain (almost always Ethereum) and inherits the base chain's security while offering cheaper, faster execution. The mental model: Ethereum is the highway. Layer 2s are the local roads. Most actual driving happens on the local roads, but they all eventually connect back to the highway for the heavy stuff (settlement, dispute resolution, final record).
There are two main flavors of L2 worth knowing.
Optimistic rollups (Arbitrum, Optimism, Base) batch many transactions together off-chain, submit a summary to Ethereum, and assume the summary is honest. If anyone catches a fraudulent batch, they can challenge it during a built-in dispute window (usually about seven days) and have it rolled back. The optimism is the default-trust assumption. It works because cheating is detectable and punishable.
Zero-knowledge rollups (zkSync, Starknet, Linea, Scroll) batch transactions and submit a cryptographic proof to Ethereum that the batch was valid. No dispute window needed. The proof itself is the validation. ZK rollups are technically more elegant but harder to build. The technology is maturing fast.
Bitcoin's L2 story is different. Bitcoin's main L2 is the Lightning Network, which uses payment channels for fast, cheap micropayments. It does not have the same general-purpose smart-contract capability as Ethereum's L2s, but it works well for what it does. Lightning is how Bitcoin can plausibly be used for everyday payments without congesting the base chain.
A few more terms before we move on.
EVM-compatible describes any chain that can run Ethereum smart contracts. Polygon, Avalanche, Base, Arbitrum, BNB Chain, and most major L2s are EVM-compatible. This matters because developers can deploy the same code across all of them with minor changes. The EVM has become a de facto standard the way x86 became a de facto standard in computing. Non-EVM chains like Solana and Aptos run their own virtual machines (Solana Virtual Machine, Move VM) and require different code.
Sidechains are independent blockchains that connect to a base chain via a bridge but do not inherit its security directly. Polygon's PoS chain (different from its newer rollup) is technically a sidechain. The distinction matters when you're evaluating where your assets are most vulnerable.
What does this all mean for you as a participant? Two practical things.
First, when you read about "scaling solutions" or "Ethereum's L2 ecosystem," you should hear what is actually being said: developers and users are choosing where to do their transactions based on the tradeoffs they care about right now. Some L2s have hundreds of millions of dollars in daily volume. Some have almost none. The competition is real and it shapes which products win.
Second, when you move assets between chains and L2s, you are crossing bridges, and bridges are historically the most dangerous part of crypto (we cover this in Day 19). The good news is that bridges to and from rollups that inherit Ethereum's security are structurally safer than bridges to independent sidechains. The bad news is that most users do not know the difference. Be one of the users who does.
Tomorrow we look at the single most-used product in crypto, the one that even skeptics now grudgingly admit has product-market fit. Stablecoins.
Glossary
| Term | Definition |
|---|---|
| Blockchain trilemma | The tradeoff every base-layer blockchain makes between security, decentralization, and scalability. You can optimize for two; the third gives. |
| Layer 1 (L1) | A base-layer blockchain. Bitcoin, Ethereum, Solana, Avalanche, etc. |
| Layer 2 (L2) | A blockchain that runs on top of an L1, inheriting its security while offering cheaper or faster execution. |
| Rollup | A type of L2 that batches transactions off-chain and posts a summary or proof back to the L1. Two main types: optimistic and zero-knowledge. |
| Optimistic rollup | Posts batched transactions to L1 with a default-trust assumption. Includes a dispute window during which fraud can be challenged. (Arbitrum, Optimism, Base) |
| Zero-knowledge (ZK) rollup | Posts a cryptographic proof of transaction validity to L1. No dispute window needed. (zkSync, Starknet, Linea) |
| EVM (Ethereum Virtual Machine) | The runtime that executes Ethereum smart contracts. "EVM-compatible" chains can run the same code. |
| Sidechain | An independent chain connected to an L1 via a bridge. Does not inherit L1 security directly. |
| Bitcoin Lightning Network | Bitcoin's L2 for fast, cheap micropayments via payment channels. |
| Bridge | The infrastructure that moves assets between chains or layers. Often the most-attacked part of crypto. (Covered in Day 19.) |
Reality check
Pick two of these chains and explain which two of the three trilemma properties each one prioritized: Bitcoin, Ethereum, Solana, an Optimistic Rollup like Arbitrum.
If you can name the tradeoffs without looking back at the lesson, you have the framework. If not, re-read the trilemma section. It is the single most useful lens for evaluating any new chain you encounter for the rest of your career in this space.
Read deeper
1. What is a Layer 2 network for blockchains? by The Block
The clean overview of the L2 category.
Read on IMPCT (curated commentary) | Read original (theblock.co)
Deven's take. Read this when you want the architectural picture without the optimism-vs-ZK weeds. The Block does a good job of explaining why L2s exist at all and what they trade against. The single most important thing to internalize: an L2 is not an alternative to Ethereum. It is a part of Ethereum's strategy. Every dollar of activity on Arbitrum or Base is also (eventually, in summarized form) Ethereum activity. The "Ethereum killer" framing you sometimes hear in crypto media is misleading. Most of the L2s are not trying to kill Ethereum. They are trying to extend it.
2. What is the blockchain trilemma? by The Block
The framework that explains why every chain feels different.
Read on IMPCT (curated commentary) | Read original (theblock.co)
Deven's take. If you only read one piece from today's recommendations, make it this one. The trilemma is the single most-useful evaluative lens in crypto. Any time someone tries to sell you on a new chain, ask which two of the three they picked. The answer will tell you almost everything about what the chain is good for, what its weak points are, and what it's competing against. Most "this chain is special" pitches collapse when you push on them with the trilemma.
3. What is the Bitcoin Lightning Network and how does it work? by The Block
Bitcoin's answer to scaling.
Read on IMPCT (curated commentary) | Read original (theblock.co)
Deven's take. Lightning is underrated outside of Bitcoin-maximalist circles. It is the reason Bitcoin can credibly be used for everyday payments (people in El Salvador, Argentina, and Lebanon do this in volume) without congesting the base chain. The user experience is still rough by mainstream consumer standards, but it is improving fast. If you want to understand Bitcoin as a payments network rather than only as a store of value, this is the technology that bridges the gap.
4. What is Berachain? and What is Monad? by The Block
Two of the more interesting new Layer 1 entrants.
Read on IMPCT (curated commentary on both) | Read original Berachain (theblock.co) | Read original Monad (theblock.co)
Deven's take. Skim both. Not because you need to use either chain, but because they represent two different bets about how to compete with Ethereum and Solana. Berachain bets that you can build a chain around an entirely new consensus mechanism (proof-of-liquidity) and a tri-token ecosystem. Monad bets that an EVM-compatible chain can win by being radically faster than Ethereum without abandoning the EVM. Both are interesting case studies in how the competitive landscape thinks about differentiation in 2026. Watch which ones get real usage and which ones don't.
5. What are EVM-compatible blockchains? by The Block
The de facto standard that connects most of the ecosystem.
Read on IMPCT (curated commentary) | Read original (theblock.co)
Deven's take. Useful when you want to understand why some chains feel interoperable and others feel like islands. EVM-compatibility is the closest thing crypto has to a USB standard. It is also why your MetaMask wallet works on a dozen different chains with just a network change in the settings. Non-EVM chains (Solana, Aptos, Sui, Cosmos chains, etc.) require different wallets and different developer tools, which fragments the experience. The EVM is not strictly the best technical design, but it is the standard, and standards have their own gravitational pull.
Tomorrow
Stablecoins. The product that does the most actual work in crypto, settles the most actual dollars, and increasingly underwrites the case for crypto's role in the financial system. We look at how a "stable" coin is actually stable, the three structural designs, why USDC didn't blow up the way UST did, and where stablecoin volume is taking the industry next.
See you in the morning.
Preview reader
You are reading a private preview of IMPCT Institute. If something landed, didn't land, or felt confusing on this lesson, tell us. Short notes are useful. Long notes are useful. No notes are also fine.