I have gotten a few questions about value accrual on layers. Hate to break it to you:
- Execution layer: highly competitive, tight margins, a race to the bottom. We can see this from alt-L1s, where the sole gimmick is accruing as little value as possible.
Rollups have a significant advantage as they pay a small fraction or zero of alt-L1s on inflation budgets, but any significant value accrual over that is unclear. MEV mitigation will be an area of innovation, and the rollup with the strongest MEV mitigations will be advantaged.
- Settlement layer: nearly zero for optimistic rollups in the optimistic case; tending towards zero for ZK rollups for anything above 100 TPS. Low-activity ZKRs become recursive rollups (L3s) or sharing proofs.
- DA layers: requires very little compute, can use consumer hard drives for extremely cheap storage. Possible network effects as the most decentralized DA layer have the highest capacity, but also a competitive space with many players over the next couple of years.
In my opinion, 99% of crypto assets today are grotesquely overvalued, and derive their value entirely from speculation & gambling, rather than any actual value accrual or projections of value accrual. So, where will the value accrue to? It's pretty obvious - the application layer
People have seen this graph a hundred times, and it's bleeding obvious what's going on. Despite all the hype and rage, L1s barely accrue value, and a small fraction of their marketing budgets and farming incentives worth billions of dollars collectively. All about applications!
As an example: dYdX earned $67M in October, while paying $1.5M in gas fees for verifying; $0.5M for DA. As activity goes up, the settlement fees become a small fraction, as do the DA fees as data shards & DAS roll out. It's clear to me applications will accrue most value.
Anyone who has been in this space and paying attention will know we go through waves of speculative hype, where tokens pump for very little fundamental reason. There's only one thing that endures: M. O. N. E. Y.
Both settlement & execution layers (maybe DA?) have an opportunity to develop quality money, and transcend their actually very limited value accrual properties. Going back to the Token Terminal graph - there's a large elephant in the room, and it's that way because MONEY.
I know everything I have said above is highly unpopular, and I'm prepared to be trashed by every single influencer you know and love. But it's the truth, it's the way the world has always worked, and it will continue to, long after speculative games have faded away. Peace out.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
I'm delighted to see a lot of anticipation and excitement for rollups in 2022, but I'm only cautiously optimistic. With some exceptions (Immutable X), rollup teams haven't yet proved they are capable of outcompeting alt L1s on marketing, public relations & business development.
Some would argue they are in beta - but so are Solana, Cardano, Polkadot etc.
I think smart contract rollups - particularly StarkNet & zkSync 2.0 - will need most of 2022 to become a mature product out of beta. Arbitrum will be first, but Nitro & FSS will still take time.
Which is not to say there won't be adoption - but that depends entirely on how well their teams & communities market the rollups, and how their tokens turn out. But true blockchain scalability ready for mass adoption for the very first time - unlikely to happen in 2022.
Showerthought: ZKRs should be loss-leading, with max ~$0.10 transaction fees* for the end user. Verification price capped at ~250 gas. Cost to rollup: $12,000/day, decreasing rapidly with growing activity, with break-even at 3.33 TPS*. From there, tx fees will keep decreasing.
This solves for the chicken-and-egg problem of low activity = high costs and thus low demand. The $12K/day can either come from the rollup team's marketing budget, or from token inflation. For perspective: SOL, DOT, AVAX, ADA etc. are subsidizing >$10M per day - $12K is peanuts!
The way I see it: if a ZKR cannot have 3.33 TPS activity soon, then it's surplus to requirements anyway. So, rollup teams should just consider it a small fraction of what should be much larger marketing & bizdev budgets, even if there's no token.
Recently, we have seen some revisionist memes around decentralization & security. I think we need better awareness of what these have always meant. Attempt:
Decentralization: ease & cost of verifying transactions
Security: risk-adjusted difficulty to compromise the network
Possible metrics:
- Decentralization: how much money do you have to spend on hardware that can verify transactions reliably? Lower is better.
- Security: PoS - how many entities own enough tokens to compromise the network? PoW - cost of acquiring hardware. Higher is better.
Of course, there are many nuances: multi-clients? slashing? nature of delegation (plutocratic? randomized?)? Staking queues? Battle-tested? Etc. Which is why people can exploit the complexity with populist revisionist memes.
Thoughts? Is the above over-simplification helpful?
Haven't listened to UCC, but seeing a lot of comments about it. Firstly, I'm delighted we have moved past the "rollups bad" phase to "rollups awesome but only on L1 bags I hold". However, the whole point of a modular architecture is the concept of L1s melt away.
1/
Rollups will always verify on the most secure, decentralized, robust, battle-tested settlement layer, backed by the soundest money and with the widest range and depth of liquidity. The cost is negligible, and the risks of having a bridge to a less secure SL isn't worth it.
2/
I believe Bitcoin would have been the best SL, but given it doesn't support rollups, Ethereum is the next best option. For a PLONK rollup doing 1,000 TPS, gas fees per transaction is only $0.00003. With 10,000 TPS it's 10x lower still etc. These are negligible numbers.
3/
Alternatively, implement an actual pricing system (I know there's some work happening on that) and set a transaction fee floor to $0.01 or so like Polygon PoS has done.
For negligible / 0 fees, a modular approach is the only way out.
Is it acceptable for an L1 to reject transactions it determines to be spam or DDoS attack?
With EIP-4488 now coming post-Merge, here are other avenues to lower fees on rollups:
- Optimizations & calldata compression: Arbitrum Nitro will halve fees, Optimism working on it
- Wallets supporting BLS signatures: Arbitrum can enable it: easy ~1K gas reduction
(1/7)
- Socially & financially incentivize developers to deploy on rollups. Particularly memecoins & NFTs, which have been the majority culprit for excessive gas fees on Ethereum
- Incentives for dapps on rollups - Uniswap liquidity mining coming to A1 & OE is a big deal
(2/7)
- Greater usage on rollups -> lower fees. Currently, a swap on zkSync / ZigZag costs $0.61, will go lower. (Source: L2fees)
- Volitions: StarkNet & zkSync will both have alt-DA options. zkPorter estimates $0.01-$0.03 tx fees. Lower security, but still higher than alt-L1s.