True.
The only remaining debate is what constitutes “toxic” ordering. It is my view that any ordering rule except FCFS is toxic to market structure long term.
High-frequency traders and limit-order updates contribute materially to price discovery. If a mechanism suppresses or filters latency-sensitive informed flow, the likely result is that discovery migrates elsewhere.
@will__price@0xasrequired We also need proof that censorship resistance and fair ordering was maintained by the sequencer.
Execution only occurs *given an ordered series of transactions *
TL;DR:
> Type of takers responsible for price discovery: HFT/algorithmic and latency-sensitive traders.
> Type of venues dominating price discovery: CEXes and DEXes with low-latency FCFS microstructure.
Cancel priority, RFQ, and FBA microstructures deter the very category of trades responsible for the bulk of price discovery.
Today, we're releasing the Fermi Trade SDK.
Programmatic Access to Solana's fastest perps DEX is now open.
To access it, simply generate an API key on the frontend with your approved wallet.
More info and latency benchmarks below 👇
We've had a lot of questions on how @fermitrade enforces deterministic ordering on Solana.
Today we're releasing the full technical documentation, covering the design and working.
Dive in, or drop the link into your favourite LLM and ask away! (Link below)
This is not about countries. Global finance & trading can run out of any neutral location; the key is whether we can trust the behaviour of the exchange, re sequencing, execution, liveness, and CR.
“Global Decentralization” is not strictly necessary for this, and comes with a significant penalty re latency that can’t be brushed under the carpet either
In an ideal world all software and hardware would have "nutrition labels" that provide a full list of trust dependencies - what math and which actors' honest behavior (and on what time scale) the system is relying on to provide its core functionality and implied guarantees.
@0xasrequired Not rhetorical, it’s actually possible.
The key is determinism. If you remove leader discretion in ordering, you can perfectly simulate all transactions in sub 10ms. Execution still lives fully onchain.
https://t.co/UoRuniAPC2
cancel priority clobs differ from classic CLOBs in the nature of the deal they represent.
In a classical CLOB, the orderbook reprsents open offers that anyone can accept at any time.
In a cancel priority book, there’s another level of confirmation required - after you take an offer, the makers still has a final, “last look” decision available - whether to accept the trade or to pull a last-minute cancel.
This subtly erodes the credibility of the represented depth on orderbooks. In many senses, its not as “real” as a traditional CLOB.
Consider a larget taker order. in traditional CLOBs, there’s no way to filter by taker, so they get access to the full displayed depth. However, in a cancel-prior book, the makers may use the additional info (eg. large size order), as a signal of an informed trader - and pull their quotes after the fact. This means that the large taker gets a worse fill, or not at all.
Now, where do you think this taker would prefer to trade? a classical CLOB or cancel-priority CLOB? The remaining question is what percentage of taker flow is adversely selected on (and thus avoids) cancel-priority vennues.