What each method actually claims
Armilla computes rather than foretells — and this page exists so that you can verify it. For every tool you can see what the result is derived from and the limits the method does not cross. The arithmetic is described so you can reproduce it on paper.
They do not predict events and they name no future dates. A birth date and a moon phase are arithmetic, and arithmetic does not know what will happen to you. Everything you read here concerns structure and rhythm, not chronicle.
My personal code
- The month, day and year of your date are each reduced to a single digit — separately, so the intermediate steps stay visible.
- Master numbers 11, 22 and 33 survive this step: the method treats them as meaningful rather than “not yet reduced”.
- The three numbers are then added and reduced the same way. That is the life path — it stays the same for life.
- The personal year works differently: birth day and month plus the CURRENT year. That is why it changes annually.
- The challenge is the difference between the fully reduced month and day, an axis of 0–8. Master numbers are not preserved here, otherwise a position 10 would appear that the method does not have.
14.06.1997 → 6 + 5 + 8 = 19 → 1+9 = 10 → 1. Year 2026: 6 + 5 + 1 (from 2026) → 3. Challenge: |6 − 5| = 1.
- computed from the birth date alone — no other input is used
- describes patterns of behaviour, not events on a calendar
- the same date always gives the same code: it is arithmetic, not a reading
Compatibility
- Both dates are reduced to zodiac signs by calendar boundaries, then the element and modality of each are compared.
- The three dimensions come from that pair rather than from a single score: one number would hide what actually differs.
- The timing block computes each personal year with the same function as Personal Code — otherwise two tools would tell you different years.
- In sign mode the calculation is deliberately coarser: there is less input, and no timing at all.
Water + air → different reaction speed; fixed + mutable modality → different attitude to plans.
- no single compatibility score — three named dimensions instead
- no assumptions about relationship type
- sun sign boundaries may shift by a day
Relationship Tarot
- The deck is shuffled with Fisher–Yates from a seed that Web Crypto issues once per session.
- The index is drawn by rejection rather than modulo: modulo biases the distribution toward small indices.
- The seed lives in session state, so reloading the page gives the same cards — a reading is never silently re-rolled.
- Picking a card by hand is not decorative: the position indexes an already shuffled deck.
seed a3f19c04… → a permutation of 22 arcana → positions 3, 17, 5 read as context, tension, next step.
- the draw is a seeded shuffle: the same session always gives the same cards
- major arcana only for now
- names a situation and its tension — does not state what will happen
Today's focus
- The number of the day is the date itself reduced to a digit. None of your data.
- The card of the day is deterministic from the date, so everyone gets the same one that day.
- The moon phase comes from the same model as the moon page.
18.08.2026 → 1+8 + 8 + 2+0+2+6 = 27 → 9.
- moon phase is an approximation, not ephemeris
- no personalization without explicit opt-in
Yes or no
- One bit from Web Crypto entropy.
- The verdict is drawn by rejection over 0–1 rather than by modulo: with a coin, bias shows faster than anywhere else.
- Your question is not passed into the calculation at all — which is why the answer does not depend on it. That is not a shortfall but the definition of a fair draw.
4000 draws in the test keep the share of “yes” between 47% and 53%.
- an unbiased 50/50 draw from Web Crypto entropy
- the answer does not depend on the question asked
- use it to break a tie you have already thought through, not to decide for you
Moon today
- The age of the moon is counted from the known new moon of 6 January 2000 using a uniform synodic cycle of 29.530588853 days.
- Illumination is the cosine of the phase, so it is a number rather than an estimate.
- Eight phases split the cycle into equal sectors, offset by half a sector: otherwise a new moon would begin half a sector before itself.
- The next full moon is found in cycle days, not by walking the calendar.
Days to full: (14.765 − age) mod 29.53. No times are shown: the model does not know hours.
- uniform synodic cycle, not ephemeris
- phase dates are given without times
- near a phase boundary the name can differ from an exact calculation