Dev Update 2026-04-06

News and Announcements
1

Dev Update | Neptune & Triton Projects

1. Last Week by the Numbers

neptune-core

  • Issues: 1 raised, 2 closed
  • Pull Requests: 2 opened, 1 merged
  • Commits: 4 merged into master

triton-vm

  • Pull Requests: 1 merged
  • Commits: 1 merged into master

2. Stand-up Summary

Alan Szepieniec

  • Last Week: Focused on prover efficiency. Conducted local profiling to identify bottlenecks, discovering that the “extend” step accounts for 60% of computation time.
  • Coming Week: Integrate informative logging tools esp. wrt. timing into neptune-core for hardcore users, and conduct PR reviews.

Thorkil Værge

  • Last Week: Published a new blog post and released version 301 of the web wallet, which introduces an automatic update checking mechanism. Managed the architectural discussion regarding the ustration barrier and block header modifications.
  • Coming Week: Aims to complete the programming for the next fork’s block header changes and evaluate the code diff size between the proposed architectural paths.

3. Technical Discussion

Prover & Profiling

Profiling of the prover indicates that the “extend” step—filling auxiliary table columns—is the primary bottleneck (60% of compute). While not “embarrassingly parallel,” the team noted that the underlying operations are somewhat parallelizable. Comparisons between “frugal mode” and other memory-saving strategies showed performance parity (within 3%), validating the mathematical efficiency of the frugal approach.

Block Header Architecture

An architectural disagreement persists regarding the block header, specifically involving a “pseudo field.” The team has opted for a “parallel implementation” strategy: both proposed changes will be developed to compare the resulting code diff size. The goal is to minimize the impact on the codebase outside of blockheader.rs.

4. Updates and Announcements