Credibility
Publishable comparisons need traceable evaluation: what was fit, what was held out, what metric moved, and what stayed internal by design.
GNSS clock comparison · holdout discipline · results you can defend
Your clocks disagree. Tare makes that disagreement defensible.
Long-baseline timing work lives or dies on whether two analysis centers’ clocks can be compared fairly. Tare is post-processing that is fitted on training windows, proves itself on held-out days you did not fit, and returns corrected outputs plus RMS-class summaries—not a paper recipe, not a black-box promise.
Typical held-out: low-30s ns RMS before → 17–22 ns after (~45% reduction in tested configurations). Not universal until we run your panel.
Tare Program · Banlys, LLC · patent pending · invite-only pilot (≤10 seats) · evaluation only—not for live production timing, navigation, or trading.
ORBIT → AC → EVAL
illustrative
Artwork: constellation → analysis-center clocks → holdout evaluation (not live data)
01 · Why this matters
The number is not the hard part. The argument is.
Labs, product teams, and review panels do not fail on having clocks—they fail on being unable to show that the comparison was fair after the fact. A biased window, a leaky train/test split, or an undisclosed fit can invalidate months of work without anyone noticing until review day.
WITHOUT HOLDOUT
WITH TARE
Operations
Wrong timing bleeds into product specs, campaign planning, and legal traceability. Fixing it late is expensive; proving it early is cheaper.
Control
You should not have to reverse-engineer someone else’s method to use corrected clocks. Tare ships outputs and summaries; the correction path stays in the evaluation envelope.
Stakes first. Spreadsheets second.
Read how we define fair before you read how far RMS moved.
02 · What Tare is
Evaluation + correction deliverable. Not a method paper on your site.
Tare is a holdout-validated post-processing service for multi-AC GNSS clock comparison. You provide agreed inputs (comparison series, windows, product definitions). We return corrected clock products and RMS-class statistics computed the same way before and after—so the improvement is measurable, not narrated.
Accepted inputs: multi-AC clock-comparison series and agreed IGS-class clock products. Raw receiver logs and streaming feeds are not accepted unless agreed in writing.
STEP 01
You define the panel
AC products, baselines, train/holdout windows, comparison metric.
STEP 02
We fit on train only
Correction is calibrated inside the pilot envelope—not on the days you will test.
STEP 03
We score on holdout
RMS (and agreed summaries) on days your team did not use to fit.
STEP 04
You receive outputs
Corrected series + evaluation report. Method remains confidential.
train ∩ test = ∅ · ship outputs, not recipes · typical ≠ universal until your pilot runs
WINDOW DISCIPLINE
train ∩ holdout = ∅
The product is the split. If we fit on holdout, the number is not yours anymore.
03 · Why you would want this
Less arguing about whose clock is wrong. More signing off a number you can defend.
+
A comparison you can defend
Holdout discipline is the product—not a footnote in someone else’s README.
+
Material RMS movement (when it works)
Typical reductions on the order of ~45% in tested configs; remainder commonly 17–22 ns, not zero.
+
No method archaeology
You integrate deliverables; you do not rebuild their entire research stack.
+
Evidence you can open
Pinned DOY ranges, null battery, strict-window examples—for reviewers who read footnotes.
+
Pilot-shaped risk
Structured evaluation before a bigger commitment—not a perpetual SaaS gamble.
04 · Where corrected time lands
One holdout spine. Different clocks on the line.
Tare is not “an HFT product” or “a telecom product.” It is defensible GNSS clock comparison that many downstream desks, SLAs, and review panels need to trust. The verticals below are who feels the residual first—not separate miracle claims.
Same deliverable today: train/holdout post-processing, corrected outputs, RMS-class summaries. Maturity tags tell you what is pilot-ready vs software-path vs R&D.
Pilot today
TareHFT · software path
Capital markets & sync desks
Co-lo timing and event ordering live on whose clock you trust.
When residual noise is tens of nanoseconds, sequence calls and latency budgets amplify small disagreements. Tare gives you a holdout-defensible comparison before you bet desk architecture on it. Desk visuals here are illustrative—not field claims.
I/O shape: recorded CLK / comparison panels → corrected series + holdout RMS (batch pilot; streaming envelope in TareHFT phase 2).
Pilot today
Telecom & network sync
5G and backbone sync need traceable time, not heroic holdover stories.
Carriers and timing-as-a-service vendors defend SLAs with comparison data across products and windows. Tare is how you show the panel was scored fairly—before it becomes a field incident.
Same pilot envelope: AC clock products, agreed baselines, holdout report you can attach to a review packet.
Pilot today
Academic track
Labs, NMIs & product integrators
Publishable comparisons and shippable clock specs share one bar: fair evaluation.
National labs and GNSS timing vendors need reviewers to see what was fit, what was held out, and what moved—without handing over the correction recipe. Academic pilots use the same discipline at a different fee tier.
Evidence room + technical note for footnotes; pilot for your panel.
Stakes vary. The split does not.
Pick your lane above; see what else we’re building or jump to adoption.
05 · How you adopt it
Hard science. Practical envelope.
Implementation is intentionally boring: files in, evaluated outputs out. No rip-and-replace of your entire stack on day one.
1
Apply / scope the pilot
Tell us your AC products, baselines, and what “fair comparison” means for your program. We align on metrics and windows before touching data.
2
Secure data handoff
Sneakernet/USB or agreed transfer—your security rules lead. We do not need your method; we need the comparison inputs you already use.
3
Train / holdout run
We fit on train, report on holdout, document nulls and pinned exhibits. You get corrected outputs + summaries.
4
Decide on merit
If holdout holds on your panel, expand. If not, you still have a documented evaluation—not a marketing slide.
90-day pilot track
90-day evaluation · May–July 2026 cohort · $45k academic / $195k commercial · invite-only · ≤10 concurrent seats. Details on the pilot page.
06 · Proof (not promises)
The excitement is real. The numbers are qualified.
We show typical holdout performance on the home page because it orients you. We show pinned runs and null tests in the evidence room because that is what survives review.
RMS · BEFORE / AFTER · HOLDOUT WINDOW
animated · illustrative
Illustrative trace — shape only; pinned DOY tables live in Evidence
RMS_before · typical
low 30s
ns · window-dependent
RMS_after · holdout
17–22
ns · not remainder = 0
Δ_RMS · typical
~45%
not a guarantee on your data
Pinned strict-window example
80.8%
Labeled reduction on DOY 2024-250–279 (IGS/CODE/ESA context in technical note). One published exhibit—not a claim that every window behaves identically.
EVIDENCE · PINNED_DOYNull battery
holdout ≠ luck
Time-shift and synthetic controls documented in the evidence room. If the effect were trivial leakage, you would see it here.
EVIDENCE · NULLSFootnotes welcome.
Console layout, DOY tables, scope limits, no independent lab claim (yet). Full definitions in the technical note.
07 · What we’re building
Multiple Tare programs. One timing bench.
The pilot is the product you can buy today. The family below shares holdout culture and Banlys’ timing hardware heritage—each line has its own maturity; we do not fold R&D claims into pilot numbers.
North star: a sealed corrector at the timing boundary (live or near-live I/O, method inside the box)—only after holdout proof on your panel and a streaming latency envelope, not before.
Pilot today
Tare Program
Holdout-validated GNSS clock-comparison post-processing: corrected outputs, RMS-class summaries, evidence room. The 90-day pilot track is the commercial and academic entry point.
Software path
Phase 2 queued
TareHFT
Batch validation closed on recorded CLK panels (harness path; no live co-lo claim). Next: streaming envelope, per-epoch latency shape, CGGTTS I/O spec for desk integration—not a claim that live co-lo is shipping yet.
R&D
TareGEO
Does corrected timing show up in PPP / position? Baseline vs IGS20 runs exist; optional step is correction inside the PPP chain. Separate from core clock-comparison proof tables.
Sim + desk prep
TareShield
Trusted reference vs incoming GNSS stream—flag when delta exceeds threshold (spoof / anomaly). Toy sim and two-dongle scripts on the bench; deployable reference at customer sites is a later milestone.
08 · Who runs it
Built in the workshop that already ships timing products.
Tare Program is operated by Banlys, LLC—a small shop with more than ten years of delivered products on real benches (timing hardware, precision tools, and field-hardened gear). We are not a slide-deck licensing someone else’s paper. We built holdout discipline because we needed it on real panels.
We do not claim independent lab certification yet. We do claim documented evaluation, honest scope limits, and outputs you can integrate without adopting our research program.