Suzuki's y₁ = D^(1/(β+1)).
Equations
- MathlibNt.SieveTheory.SwitchingPrinciple.suzukiYOne D β = D ^ (1 / (β + 1))
Instances For
The exact finite normalized base object V₁(D,z)/V(z) from Lemma 7.1:
∑_{y₁ ≤ p < z} ν(p) V(p)/V(z).
Only supported primes occur, as required by the finite BoundingSieve model.
The D argument in the production prime-sum is the power-coordinate parameter;
for the constant test function used here it does not affect the value.
Equations
Instances For
The finite Euler product V(z) on the production support.
Equations
- MathlibNt.SieveTheory.SwitchingPrinciple.suzukiVProduct S z = ∏ p ∈ S.prodPrimes.primeFactors with ↑p < z, (1 - S.nu p)
Instances For
The unnormalized finite base object V₁(D,z). Expanding the normalized
prime sum gives exactly ∑_{y₁≤p<z} ν(p)V(p).
Equations
Instances For
Suzuki Lemma 8.4 at the exact real base cutoff. The proof uses the production real-node Abel identity and prime-atom assembly. Lemma 8.6 is not needed in the base case: its integral estimate would be weaker than this exact telescoping identity.
The one-term source-parity sum is exactly f₁.
Replacing a real lower cutoff by max(w,2) does not change the finite
Euler ratio: every member of prodPrimes.primeFactors is a prime and hence at
least two. This is the exact endpoint repair used in Suzuki's base case.
N=1, κ=1 local-product closure.
This is the finite content of Suzuki (14.8), before the paper-specific
T̂⁺ absorption: the main term is the genuine continuous f₁(s), and the
remaining loss is the explicit dimension-one local-product error
K(β+1)²/(s log D).
The production local-product contract starts at 2. As in Suzuki's proof, we
therefore replace y₁ by max(y₁,2), use the exact carrier equality above,
and then enlarge the elementary logarithmic bound back to y₁.
Unnormalized form of the N=1 base estimate, with the exact source V(z)
factor displayed.
The discrete base object vanishes in the support branch s ≥ β+1.
Source support branch: for s ≥ β+1, the continuous f₁ vanishes.