Original selected-negative tests beyond an inner cutoff retain the complete tail balance
ProvedConnesGreen.canonical_negative_tests_after_half_cutSuppose characterizes the successful windows of the original finite packet . For every positive , we prove existence of an ORIGINAL supported test such that, with ,
The accepted exact equivalence between the original Picard half-bound and nonnegativity on all original supported tests makes a failed window supply a selected-negative test. The complete original tail is retained: the closed negative-analysis partition gives
Original actor-column summability and Cauchy-Schwarz justify these convergent squared analysis sums. Accepted full signed actor arithmetic then identifies
This proves the asserted strict full-background balance. The cutoff characterization is an explicit premise of this reduction and is the finite branch supplied by the separate unconditional dichotomy. All original actor, carrier, actual-zero, analytic multiplicity and reflected-pair normalizations are retained. The result is not an unconditional proof that a given packet has a finite cutoff, nor a critical-endpoint jump budget or RH proof.
import Definitions.Def_ConnesGreen_small_support_constants import Definitions.Def_ConnesGreen_RG0_original_inner_marker open Complex MeasureTheory ConnesRZ ConnesRZFrontier ConnesGreen WeilDefect WeilDefect.ConnesNative Set open scoped BigOperators InnerProductSpace lp ENNReal Classical ComplexOrder Topology set_option autoImplicit false set_option maxHeartbeats 3000000 noncomputable section
theorem ConnesGreen.canonical_negative_tests_after_half_cut (S : Finset CriticalZeros)
(c : ℝ) (hcut : (∀ T : ℝ, ∀ hT : 0 < T, ((1 / 2 : ℝ) • (1 : ℓ²({ρ : CriticalZeros // ρ ∈ S}, ℂ) →L[ℂ] ℓ²({ρ : CriticalZeros // ρ ∈ S}, ℂ)) ≤ canonicalPicardMarker T hT S) ↔ T ≤ c))
(T : ℝ) (hT : 0 < T) (hcT : c < T) :
∃ g : ℝ → ℂ, SupportedTest T g ∧
‖(canonicalPositiveSynthesis T hT).adjoint (sourceEmbed T (problemOneL g))‖ ^ 2 -
‖(canonicalSelectedSynthesis T hT S).adjoint (sourceEmbed T (problemOneL g))‖ ^ 2 < 0 ∧
‖(canonicalBackgroundSynthesis T hT S).adjoint (sourceEmbed T (problemOneL g))‖ ^ 2 <
-(weilDistribution (conv g (starInv g))).re := by sorry