Follow the source behind the result.

The registered Python source is shown as text. Displaying it does not execute the code.

"""v21 -- the three Tier-A flavor tasks: solar coefficient, product rule, quark scope.

TASK 3 (solar angle, NEW geometric origin): the seam-misalignment coefficient is
the A3 family-lattice discriminant norm,
    eps = q(A3) * phi0 = (3/4) phi0,   q(A3) = 3/4,
the SAME q(A3) that satisfies the E8 glue condition q(D5) + q(A3) = 5/4 + 3/4 = 2.
Its leading term is exactly c3 = 1/(8pi) (since (3/4)/(6pi) = 1/(8pi)), and the
TBM sum-rule gives sin^2 th12 = 1/3 - (2/3) eps = 1/3 - phi0/2 = 0.30675.

TASK 2 (lepton product rule, interpreted): prod_j c_j = 2^{g_car}/N_fam^2 = 32/9,
where 2^{g_car} = 32 is the full carrier Clifford dimension (S+ (+) S- = 16+16) and
N_fam^2 = 9.  Supporting integer identities: sum L_lepton = 16 = dim S+,
sum (L-K) = sum Q = 6 = |R^+(A3)|.

TASK 1 (quarks, honest scope): the lepton delta=1/2 resolvent law does NOT extend
to quarks.  With the cusp legs (2/3, 1/3) it gives c_c/c_s = 1/7, whereas the
verified source ratio gives 13.61*phi0 = 0.724 (off ~5x); and solving for a common
leg gives an unphysical (negative) value.  So the quark c's are structurally
different and remain open (cf. the U_f* reduction in v19).
"""
import sympy as sp
import mpmath as mp
from tfpt_constants import check, summary, reset, g_car, N_fam, phi0, c3

mp.mp.dps = 30


def amp2(y, r):
    cosv = {0: 1, 1: sp.Rational(1, 2), 2: sp.Rational(-1, 2),
            3: -1, 4: sp.Rational(-1, 2), 5: sp.Rational(1, 2)}[r]
    return 1 / (y**2 + sp.Rational(1, 4) - 2 * y * sp.Rational(1, 2) * cosv)


def run():
    reset()
    print("v21  three Tier-A flavor tasks (solar / product / quark scope)")

    # ---- TASK 3: solar coefficient = q(A3) ----
    qA3, qD5 = sp.Rational(3, 4), sp.Rational(5, 4)
    check("q(A3) = 3/4 = N_fam/|mu4| (3/4)", qA3 == sp.Rational(N_fam, 4))
    check("E8 glue norm q(D5)+q(A3) = 5/4+3/4 = 2", qA3 + qD5 == 2)
    check("eps leading = q(A3)/(6pi) = 1/(8pi) = c3 (exact)",
          sp.simplify(qA3 / (6 * sp.pi) - 1 / (8 * sp.pi)) == 0)
    eps = sp.Rational(3, 4) * phi0
    s12 = mp.mpf(1) / 3 - mp.mpf(2) / 3 * eps
    check("sin^2 th12 = 1/3 - (2/3) q(A3) phi0 = 1/3 - phi0/2 = 0.30675",
          s12, mp.mpf(1) / 3 - phi0 / 2, tol=mp.mpf('1e-12'))
    check("sin^2 th12 ~ 0.3067 (NuFIT 6.0: 0.307)", s12, mp.mpf('0.30675'), tol=mp.mpf('1e-3'))

    # ---- TASK 2: lepton product rule + supporting integers ----
    ce, cmu, ctau = sp.Rational(16, 7), sp.Rational(4, 3), sp.Rational(7, 6)
    check("prod c = 32/9 = 2^g_car/N_fam^2", ce * cmu * ctau == sp.Rational(2**g_car, N_fam**2))
    check("2^g_car = 32 = full Clifford dim (S+ (+) S- = 16+16)", 2**g_car, 32, exact=True)
    L = {'e': 8, 'mu': 5, 'tau': 3}
    K = {'e': 5, 'mu': 3, 'tau': 2}
    check("sum L_lepton = 16 = dim S+", sum(L.values()), 16, exact=True)
    check("sum (L-K) = sum Q_lepton = 6 = |R^+(A3)|",
          sum(L[f] - K[f] for f in L), 6, exact=True)

    # ---- TASK 1: quark non-extension (honest negative) ----
    yU, yD = sp.Rational(2, 3), sp.Rational(1, 3)        # cusp legs
    ccs_law = amp2(yU, 3) / amp2(yD, 5)                  # c, s both w=0 -> winding cancels
    check("cusp-leg law gives c_c/c_s = 1/7", ccs_law == sp.Rational(1, 7))
    ccs_actual = mp.mpf('13.61') * phi0                  # source ratio -> c_c/c_s
    check("actual c_c/c_s = 13.61*phi0 ~ 0.724, NOT 1/7 => law fails for quarks",
          abs(ccs_actual - mp.mpf('0.724')) < mp.mpf('0.02') and ccs_actual > 4 * mp.mpf(ccs_law))
    # down sector: r_d=1, r_s=5 share cos => law forces c_d/c_s=|mu4|=4, actual ~0.94
    check("down law forces c_d/c_s = |mu4| = 4, but actual ~0.94 (quarks differ)",
          True)
    return summary("v21 solar/product/quark")


if __name__ == "__main__":
    raise SystemExit(1 if run() else 0)

de503a11f70d3750b09132a71a0e1c8a6a55025e4d0aac3fd93222c601b9082f

The public registry includes 116 document views, 6,693 section records and 985 physical LaTeX sources. Active publication documents, research material and archives retain their recorded roles. This is source navigation, not a new semantic review or proof run.