PAXAL
devlog
Notes on building. For one-line "what shipped" entries see /changelog; for "what's next" see /roadmap. Both are short. The pieces below are the longer "why."
2026-05-20
HLS for point clouds — why splat streaming looks like video streaming
Captures got too big to ship as one file. The fix wasn't a new compressor — it was borrowing the delivery protocol that already solved this problem for video.
read on →2026-05-18
What the unit tests caught
The spike-track tooling ships with 98 pure-logic tests, 5 seconds wall-clock for the whole suite. Most of them just pin behaviour that's already correct. Two of them found bugs nobody had noticed.
read on →2026-05-18
Reading the trainer's source before writing the runbook
We had a recipe for validating PUP-3DGS pruning on our own captures. Building the harness to support the recipe revealed the recipe was wrong about which trainer to use.
read on →2026-05-16
Engineering quality as feature velocity
The §12 gates landed as guardrails, not chores. Each one bought more feature time than it spent. Here is the bookkeeping.
read on →2026-05-16
Photoreal-first in practice — what we said no to
The photoreal-first stance is easy to state. It only matters at the points where we said no to a feature that would have shipped in another engine. Here are the concrete ones.
read on →2026-05-16
The three-channel rewrite — heard, seen, felt
A game is heard, seen, and felt. v3 shipped one of those. v4 ships all three deliberately — and the pairings between them are where most of the headroom lived.
read on →2026-05-16
Why splats — the photoreal-first stance
Gaussian splats are the moat. Everything else (post-process, audio, mode-design) is in service of not getting between the capture and the eye.
read on →2026-05-14
Three workers, one main thread
The streaming pipeline is the engineering proof of concept. Why three workers, and why not one or seven.
read on →