← All calls
ARBORIST#115

ZcashD retirement timeline and dynamic fees

A retirement roadmap anchored to ZcashD end-of-support dates and a dynamic fee design were presented.

Agenda summary 10

  1. Zebra update

    Zcash Foundation · Marek

    • The team released Zebra 3.1.0.
    • The main changes are an ARM64 Docker image fix, a dust filter for mempool transactions, and a setting for RPC response size.
    • It merged initial support for the ping RPC and is greatly expanding Grafana metrics.
    • It is adding the ability to verify and produce blocks with the pre-Canopy subsidy, which is used to increase parity with zcashd on regtest.
  2. ECC core update

    Electric Coin Company · Str4d, Kris

    • To release Zallet updates more often, the team is reviewing SLSA on top of reproducible builds, and when that is done Zallet alpha 3 will come soon.
    • It also finished reviewing a change to the lightwalletd protocol that optionally includes transparent data in compact blocks, and is reviewing the implementation.
    • It is working on an account disconnect feature in the core library so a Keystone hardware wallet can be disconnected and reconnected.
    • The Sapling and Orchard crate releases make it possible to add to a PCZT a signature made outside the minimized PCZT flow.
    • The Zcash dev tool created a wallet sync benchmark metric that measures the time to sync the most recent 1 day, 7 days, and the following 30 days of blocks.
    • For 1 day, most of the cost is the subtree update, and for 7 days is where most of the sync time goes.
    • It included the Halo 2 changes needed for the Orchard ZSA circuit in a release.
    • Proof generation became about 18 to 20% faster on modern multicore CPUs, and proof verification became 50% faster.
    • It has nearly finished reviewing the non-ZSA integration changes in the Orchard ZSA PR, with only part of the circuit code left.
    • For long-lived features, it plans to use a model in which maintainers keep a separate integration branch in the main repository up to date.
    • ECC discussed ideas for changing how the transaction format evolves to make smaller network upgrades possible.
  3. Zaino update

    Zingo Labs · Za, Nacho

    • The chain index integration and the project to retire the local cache are nearly complete.
    • Led by Nacho, the work is migrating so that all information comes from the chain index instead of the old modules.
    • Rollout of a new test framework for testing chain forks is also near completion, and Hazel is in charge.
  4. zcashd deprecation update

    Electric Coin Company, Zcash Foundation, Zingo Labs · Pacu, Alfredo

    • The team is doing an NU 6.1 retrospective with exchanges, mining pools and API providers and hearing about improvements, and is getting feedback on using Zallet and the z3 stack from one or two large exchanges and API providers.
    • It uses the zcashd end of support (EoS) as the anchor for the schedule.
    • The end-of-support date for the current version 6.10 is January 21.
    • The schedule slipped somewhat because of high market interest.
    • After announcing that all the features needed for deprecation are in place, it plans to have a roughly 16-week test period running zcashd and the new stack in parallel.
    • It shared a rough estimate that if things are ready in early Q2 2026, NU 7 would be possible on the next zcashd end-of-support date, September 9, 2026.
    • All of this is tentative and is also affected by community sentiment and the NU 7 feature composition.
    • Alfredo is creating OpenRPC documentation for Zebra and applying it to Zaino as well, and is building in the z3 repository a Rust router that routes Zallet, Zebra and Zaino calls to a single endpoint.
  5. ZSA update

    Qedit · Pablo, Vivek

    • All the Halo 2 related changes have been merged and some released.
    • For Orchard ZSA, Str4d reviewed the changes other than the circuit, and the base branch was changed from main to an integration branch in the Orchard repository.
    • Following the review to separate the parts unrelated to ZSA (such as cargo fmt), zcash-test-vectors PR 108 was split into several PRs.
    • While merging with the already merged NSM v6 implementation, it will also check that it includes all of the v6 fields.
    • In Zebra there is a test node that supports transfer, burn and finalize.
    • It is working on global state management for ZSA consensus checks, and after the Orchard review the focus shifts to librustzcash and Zebra.
    • Pacu asked about the grant Qedit submitted for the last item, the code's completeness and the testnet period.
    • Pablo answered that there is a public informal single-node testnet and that after the Orchard and librustzcash review a feature testnet can be opened.
    • Str4d explained that the main testnet follows mainnet ahead of it, so it cannot be deployed before consensus rules are finalized, and that a separate feature testnet that can easily be discarded is advantageous.
    • Nate said they would arrange a discussion in January or February about a feature testnet and a staging net shared by ZSA, Crosslink and Tachyon.
  6. Network Sustainability Mechanism (NSM) update

    Shielded Labs · Jason

    • Least Authority completed the ZIP 233 audit and shared the report with the ECC and ZF engineering teams.
    • The main recommendation is to add a requirements section and a privacy impact section to the ZIP, and the team is preparing it for next week's ZIP editors meeting.
    • After the contract ended at the end of November, Shielded Labs took over the engineering.
    • Judah is updating the code and rebasing the ZIP 234 and 235 PRs, and plans to finish before the end of December.
  7. Crosslink update (Trailing Finality Layer)

    Shielded Labs, Electric Coin Company · Nate

    • The team finished workshop 4B in early December.
    • For the first time it linked Crosslink-specific testnet ZEC to PoS staking weight, and Zingo built an iOS demo wallet and Shielded Labs also integrated wallet functionality into its visualization tool, so two wallets, desktop and mobile, are available.
    • The next milestone (mid to late January) implements the basic design, such as the unbonding period and power-of-10 amount quantization.
    • It weighs UX experiments more heavily than completeness, and also improves the persistent testnet infrastructure for real PoW mining and test asset distribution.
    • It is discussing participant engagement tasks and rewards such as badges with ZecHub.
    • The goal is a feature-complete prototype in which all functions are accessible by the end of Q1 2026, after which it will ask the community whether to continue.
    • It will start a design review with independent mechanism design researcher Nicholas della Pena.
    • The approach is to fix specific git revisions and review questions in advance and make all reviews public.
    • The promised ZIP PR has not been submitted yet because it was pushed back by preparing the review documents.
  8. Dynamic fees proposal

    Shielded Labs · Mark

    • The team published a dynamic fees lab at fees.shieldedinfra.net covering history, prior art and design goals.
    • The goals are preventing information leakage, using public information, a consistent UX, fast response to congestion, raising the cost of block-filling attacks, and ease of implementation.
    • The proposal is a comparison-based approach that recalculates the ZIP 317 marginal fee from the median of the preceding blocks (the 50 blocks from the 5th to the 55th before).
    • It uses a 10x fast lane during congestion and rounds fees to powers of 10 to reduce entropy.
    • It goes through policy stages before consensus, and consensus changes are minimized to roughly an expiry height, power-of-10 fees and a floor.
    • Threat modeling and privacy analysis follow.
    • Str4d said that the expiry height requirement eliminates disabling transaction expiry (setting it to 0), so it is a big decision separate from fees, and suggested measuring the proportion of affected transactions.
    • Nate said fee constraints should be communicated to the Tachyon team.
  9. Zec.rocks and flyclient discussion

    Blake, Kris, Str4d, Nate

    • Zec.rocks said that in its experiment of rewarding light wallet server operators, the number of servers grew from 12 a few months ago to 112.
    • It will continue ZEC rewards of about 300 dollars a month for at least six more months, and is envisioning an experiment to route user traffic to community servers.
    • Kris said that to make this trust-minimized, flyclient proofs are needed.
    • Str4d said it can be implemented easily with the zcash_history crate in a Zaino-based lightwalletd and that multiple servers would be cross-verified.
    • Nate said flyclient could also benefit Crosslink.
    • Kris tagged light wallet protocol 0.4.0, which adds transparent information to compact blocks.
    • There is no lightwalletd release yet, but it is gRPC backward compatible, and a protocol version field was added to light wallet info so that servers can report supported versions.
  10. Inverse Tech: private light wallet research proposal

    Inverse Tech · Daniel, Aurel

    • Inverse Tech has experience with Sapling and Orchard ZSA work and ZF's private programmability SDK work, and is in early exploration of improving the light wallet protocol.
    • The current approach becomes a bottleneck in proportion to the number of transactions, and servers can collect metadata.
    • It is exploring applying private information retrieval (PIR) to witness updates, homomorphic encryption (FHE) or MPC to detecting incoming transactions, and distributed storage and onion routing for send notifications and secret distribution.
    • Because it is a big topic, it said it wants to start with research that does benchmarks and simulations rather than immediately making a protocol proposal.
    • It plans to post on the forum.
    • Str4d said they would like to see results on whether FHE-based detection is now feasible, and that, like the fuzzy message detection variant Penumbra deployed, it would need padded traffic.
    • They recommended the brainstorming section of the Zcash R&D Discord as a place to discuss.

This summary is based on the original meeting notes; check them for exact details.