# Pinball — sourced facts

Every line is either SOURCED (URL + what it says) or RECONSTRUCTED (my own model / calibration).
Nothing here is a quotation I could not open myself. Common expectations the sources do not
bear out are marked CORRECTION.

## A. Competition rules — IFPA / PAPA unified rule set

Primary document actually downloaded and read:
`PAPA_IFPA-Complete-Competition-Rules-2023.09.pdf` (16 pages, 307,152 bytes), linked from the
PAPA 23 rules page and described there as "The controlling rule set for play at this event is
maintained and published jointly by IFPA and Replay".
  - https://lfs.papa.org/papa23/PAPA_IFPA-Complete-Competition-Rules-2023.09.pdf
  - https://lfs.papa.org/papa23/rules.php  (PAPA 23 Official Rules, event-specific page)

CORRECTION: `https://www.ifpapinball.com/rules/` is behind a Cloudflare interstitial and
returns 403 to every non-browser client, and `https://papa.org/learning-center/` no longer exists —
`papa.org` 301s to `https://lfs.papa.org/papa23/` and has no HTTPS listener of its own. The
PAPA "Learning Center" could NOT be opened; I did not cite it.

A1. Machine software settings for tournament play (unified rules, "Machine Settings / Software
    Settings", PDF line 510-527):
      * Tournament Mode, Free Play
      * "3 Balls"
      * "Extra Balls disabled (may award score instead)"
      * "2 Tilt Warnings (may be 0 on older machines)"
      * "Flipper AutoLaunch disabled", "Timed AutoLaunch disabled"
      * "Standard Factory Settings for Ball Savers, Difficulty, Timers, etc"
    and: "older machines may have different settings, such as allowable extra balls, five-ball
    play, or a Tilt penalty of 'entire game' rather than 'current ball'."

A2. Tilt penalty (unified rules, "Player Errors"): "Any player who tilts his or her ball in play
    will not receive any penalty other than the normal loss of ball." Slam tilt: "Any player who
    slam tilts a machine, thereby ending play for all players, will receive a score of zero for
    that game."

A3. Hardware setup (unified rules, "Machine Settings / Hardware Settings", PDF line 532), IN FULL:
    "Each machine will be properly leveled left-to-right and inclined front-to-back."
    CORRECTION: one might expect competition setups to state a range for playfield pitch.
    They do not. The unified IFPA/PAPA rules give NO pitch number and NO range — the words
    "degree", "pitch" and "angle" do not appear in the document at all (grep of the extracted
    text). The 6.5 deg figure is a manufacturer/encyclopaedia convention, not a competition rule.

A4. Stuck balls (unified rules + PAPA 23 page, "Stuck Balls"): the player must wait for automatic
    ball searches; "If the stuck ball has not been freed after four such searches, or if the
    machine is not performing searches for some reason, the player must alert the scorekeeper".
    A freed ball is replaced "in either the plunger lane, or on either upraised flipper".

A5. Ball save is NOT defined by the rules as a mechanism — it appears only as an exception:
    "an unexpected software ball save" is listed as an example of a BENEFICIAL MALFUNCTION which
    "is allowed once per game". Ball savers themselves are left at "Standard Factory Settings".
    CORRECTION: one might expect the rulebook to define ball-save. It does not.

A6. Intentional delay (unified rules, "Intentional Delays"): "time during which the ball is left in
    the plunger lane, held on a flipper, or passed from one flipper to another" counts as delay;
    the PAPA 23 page sets the limit at 30 seconds. "A player is permitted to wait a reasonable
    amount of time ... for a tilt mechanism to settle."
    CORRECTION: there is NO rule defining "a legal plunge". The rules only say autolaunch
    is disabled and that a ball left in the plunger lane is an intentional delay.

## B. Machine specifications

B1. PLAYFIELD PITCH = 6.5 degrees (solid-state), ~3 to 3.5 degrees (electromechanical).
    - Wikipedia, Pinball: "The playfield is a planar surface inclined upward, usually at six and a
      half degrees, away from the player" — https://en.wikipedia.org/wiki/Pinball
    - Williams' own "Ask Uncle Willy #14", July 29 1997, fetched in full:
      https://www.planetarypinball.com/mm5/Williams/archives/willy14.htm
      Q: "Why is 6.5 degree the preferred playfield angle? Did it used to be 3 degrees?"
      A: "...In the early days of pinball, however, the mechanical devices on the playfield were not
      up to the task of fighting that much gravity. A slope of about 3 degrees struck a good
      balance. Once pins started including more powerful and robust mechanical assemblies, the
      playfield slope could be increased..." — and the column then says outright that the origin of
      the precise 6.5 figure is folklore ("Or maybe that was just the empirically determined angle
      of one machine once that 'sure felt like it played well.'").
    - PinWiki, "General", section "Adjusting the Game Pitch":
      https://www.pinwiki.com/wiki/index.php?title=General
      "When a recommended pitch is not indicated, the 'Rule of Thumb' is 3 1/2 degrees for EM games
      and 6 1/2 to 7 degrees for solid state games."  Also: "For a Williams game, the manufacturer
      recommends getting the nose of the bubble between the first and second mark. The marks
      represent one-half degree."
    DECISION: this app declares 6.5 deg and lets the player set 3.0 - 8.0 deg.

B2. BALL DIAMETER = 1-1/16 inch = 26.9875 mm. SOURCED as a part number, not as an encyclopaedia
    claim: Williams/Bally part 20-6500, listed by a parts distributor as
    "Budget Pinball Standard 20-6500 steel 1-1/16" / "Ball steel 1-1/16" —
    https://www.pinballspareparts.com.au/20-6500-b.html
    BALL MASS = 80.8 g. NOT sourced; DERIVED here from the sourced diameter and the density of
    carbon steel (7850 kg/m^3): V = (4/3)pi r^3 = 1.0291e-5 m^3, m = 80.78 g. The commonly quoted
    "about 80 g" agrees with that derivation, which is why I did not go looking for a citation for
    a number I can compute. The engine derives the mass this way at runtime; the harness re-derives
    it independently. Density is the only free input and it is declared.

B3. FLIPPER BAT LENGTH = 3 inch (76.2 mm) standard; 2 inch (50.8 mm) on earlier games.
    - Both lengths are sold as distinct parts today: 'flipper Bat - white 3"' and
      'flipper Bat - Plastic Flipper 2" Round Top White Gottlieb 5093' —
      https://www.pinballspareparts.com.au/catalogsearch/result/?q=flipper+bat
    - German Wikipedia, Flipperautomat: "Ungefaehr Anfang der 1970er Jahre setzten sich laengere
      Flipperhebel (3 Zoll (7,62 cm) statt 2 Zoll (5,08 cm)) durch."
    - English Wikipedia gives a looser "roughly 3 to 7 cm (1+1/4 to 2+3/4 in) in length", which is
      the RUBBER-to-tip working span rather than the bat part length. NOTE the two encyclopaedias
      disagree; I used the part length, 76.2 mm, because it is the number a part is sold by.

B4. POP BUMPER is a solenoid, and the published mechanism says the impulse comes from the COIL:
    Wikipedia, Pinball: "Pop bumpers are operated by a switch connected to a ring surrounding the
    bottom circumference of the bumper that is suspended several millimeters above the playfield
    surface. When the ball rolls over this ring and forces one side of it down, a switch is closed
    that activates the bumper's solenoid. This pulls down a tapered ring surrounding the central
    post of the bumper that pushes downward and outward on the ball, propelling it away."
    The trigger is a POSITION switch, not a force switch: nothing in the mechanism reads the
    incoming speed. That is the published basis for the speed-independent impulse this app models.

B5. FLIPPER COIL is two windings (power + hold) with an end-of-stroke switch. Wikipedia, Pinball:
    "Flipper solenoids contain two coil windings in one package; a short, heavy gage 'power' winding
    to give the flipper its initial thrust up, and a long, light gage 'hold' winding... As the
    flipper nears the end of its upward travel, a switch under the flipper disconnects the
    power-winding". Also: solenoids "can draw up to 500 W momentarily and operate at voltages up to
    50 Vdc."  This app models the power winding as a constant torque to the end-of-stroke angle and
    the hold winding as a weaker holding torque, which is the published two-stage behaviour.

B6. TILT PLUMB BOB is a pendulum hanging through a ring; the published service guidance is about
    its PERIOD and its settling time, which is exactly what a pendulum model produces.
    PinWiki, "General", "Tweaking Tilt Mechanisms":
    "the position you put the plumb bob in makes a huge difference in how the tilt operates. Almost
    all parts manuals picture the plumb bob inverted (small side down) ABOVE the actuator ring, but
    almost all games have the bob BELOW the ring instead."
    "The inverted position allows quicker recovery after a tilt; the rod stops moving quicker as
    the weight is closer to its pivot point."
    "Because the weight is farther along the rod, the tilt adjusted this way takes longer to
    recover after hits and may in fact swing for a very long time (2 minutes or more...)."
    "you want the bob to be centered in the ring assembly."
    This is a physical-pendulum statement: moving the mass along the rod changes the period AND the
    decay. The engine models the bob as a rod-plus-bob physical pendulum with the bob position
    adjustable, and the harness checks that moving the bob up shortens the period and the ring-out
    time, matching the quoted service behaviour.

## C. What I derived, calibrated and chose (and where the full accounting lives)

C1. DERIVED, not sourced: the ball's mass (B2), and every consequence of the sphere's moment of
    inertia. The engine never writes the rolling factor 5/7 anywhere in code - tools-harness.js
    greps the shipped engine, with comments stripped, to prove it - so that a wrong inertia shows
    up as a wrong playfield pitch when the second oracle recovers the pitch from an emitted ball
    track. That is the one blind spot the energy account cannot cover.

C2. CALIBRATED (my numbers, no publisher states them): kinetic friction, rolling resistance, wall
    and rubber restitution, the pop and slingshot coil impulses, the flipper's moment of inertia
    and its three coil torques, the plunger spring rate, the tilt bob's masses and lengths and the
    restitution of its strike against the ring, and the rail's groove angle.

C3. CHOSEN (design, not physics): the whole table layout, every dimension of it, the shot names,
    the scoring, the ball-save duration, the colours and the cameras.

C4. THE FULL ACCOUNTING, with what each source says and what it does not, is in CREDITS.txt -
    including the five corrections above restated in the form a reader can check, and one
    unresolved difference between the service literature's "a bob can swing for two minutes" and
    what a machine could possibly do with the tilt warnings that would produce.
