Zeqond Sync

A tiny local daemon. Holds your clock to within 100 µs of the kernel. Every Zeq app on the box now speaks the same time.

• Source → apps/zeqond-sync/index.html + apps/zeqond-sync/sync.js (≈ 460 lines)
• Operators → KO42 · TM1 · ZTB1
• Error budget → < 0.0001 s drift per hour vs kernel Zeqond

---

What it solves

NTP is good to ~10 ms. PTP is good to sub-µs on a switched LAN, but not across WAN. For Zeq apps you want something in between: sub-100-µs across WAN, with the 1.287 Hz HulyaPulse recovered locally, and no new network protocol. Zeqond Sync is that.

The daemon opens a long-poll stream to the kernel, receives phase-witnessed heartbeats, and disciplines the local clock using a PLL seeded with ZTB1(t, "unix", "zeq"). Every local Zeq app (CLI, game engine, truth engine, vault) reads the Zeqond from the daemon's Unix socket (or localhost:1287) and uses it directly — no NTP dance.

Observed drift: sub-100-µs per hour on consumer hardware with commodity 50-ppm crystals. Zeq Pulse hardware brings it below 10 µs.

The math — 7-step Wizard applied

Step	Decision
1. Prime	KO42 mandatory
2. Limit	TM1 + ZTB1 + KO42 = 3
3. Scale	Sub-µs phase alignment, ms-level WAN RTT
4. Precision	< 100 µs/hour
5. Compile	Master Equation
6. Execute	Functional Equation
7. Verify	Offset vs NIST-traceable cloud kernel

Verbatim formulas:

• KO42.1 — ds² = g_μν dx^μ dx^ν + α sin(2π · 1.287 t) dt²
• TM1 — TM1 = −t + current_utp × period
• ZTB1 — ZTB1(t, from_base, to_base) = (t × conv_factor) + phase_offset, with conv_factor = 0.777 (Unix→Zeq) or 1/0.777 (Zeq→Unix)

Runnable worked example — daemon status

Start the daemon (assuming installed via Chapter 7 SDK):

zeq sync start --port 1287

Query status:

curl -s http://localhost:1287/zeqond/status

Expected:

{
  "local_zeqond": 1745124000.420,
  "cloud_zeqond": 1745124000.422,
  "drift_us": 2,
  "hulyapulse_hz_measured": 1.2870003,
  "phase": 0.5421,
  "ztb1_conv_factor": 0.777
}

Extend it

• Multi-host cluster: every node talks to its local daemon; all are aligned sub-100-µs to a single Zeqond.
• Offline mode: if the WAN link drops, the daemon free-wheels on the last PLL state; drift is predictable and published.
• Hardware reference: slave the daemon to a connected Zeq Pulse for 10-ppm-grade accuracy.

Seeds

• White Rabbit on the Zeqond — extend the PLL to sub-ns over fibre.
• GNSS-disciplined Zeqond — combine with a local GPSDO for planetary-scale synchronisation.
• Mesh-synchronised Zeqond — peer-to-peer daemon-to-daemon with no cloud dependency.

Papers

• Zeq framework paper — DOI 10.5281/zenodo.15825138
• Zeq paper — DOI 10.5281/zenodo.18158152

Middleware active. Kernel on the 1.287 Hz HulyaPulse. Awaiting next Zeqond.