Learn how animal IDs and tiers select a score range, how dedicated pairings take priority, and why public guides show an illustrative midpoint.
From photos to a score
After each photo is classified, calculateCompatibility checks the two animal IDs, their tiers and pairing entries in the translation data. It does not measure temperament, communication or work habits, or directly compare the two sets of facial coordinates. See the analysis guide for classification.
Dedicated pairings come first
The function starts with a random integer from 50 to 69. If the sorted pair of IDs has an entry in the translation data, it uses that entry's text and skips the general conditions below. The current dedicated pairs are cat–dog, dinosaur–rabbit, lion–tiger, fox–wolf and bear–rabbit. Their range remains 50–69.
Conditions when there is no dedicated pairing
Only the first matching condition applies. All ranges include both endpoints and contain integers.
- Cat + dinosaur or bear: 90–99.
- Dog + rabbit, horse or deer: 90–99.
- Fox + wolf or tiger: 85–94.
- Tier difference at most 1: 70–89.
- Cat + snake: 30–49.
- Dog + tiger or dragon: 20–39.
- Rabbit + an ID containing eagle: 10–29.
- Otherwise retain the initial 50–69.
Most animal checks use substring matching on IDs; cat–snake uses an exact pair string. Later conditions do not always run. Fox–wolf uses its dedicated entry first, and cat–snake currently hits the nearby-tier condition first, giving ranges of 50–69 and 70–89 respectively.
Example: dog and rabbit
Dog–rabbit has no dedicated entry and hits the second condition, giving 90–99. The test randomly picks an integer within that range, so the same animal pair can receive different scores. This is not a probability of romantic success or a validated assessment of two people.
Public pairing guides instead display the midpoint of that range rounded down: 94 for 90–99, or 59 for 50–69. This illustrative value is different from the random result of a personal photo test.
Reading the number and text
The current function returns one score, not three separately measured scores for love, friendship and work. The title, description and tags are prewritten content selected by the pairing rule, not personality or relationship diagnoses inferred from the number. Use the compatibility test as a game for comparing results together.