II. The cyclotomic host and the ordinary prime ramp
Write for the von Mangoldt function and for Euler's constant. The ordinary prime ramp at a real is For a primitive nonprincipal Dirichlet character , write and We derive a bound for the difference between and the normalized translate average of the finite prime sum minus its main term. The average uses a cyclotomic CM field depending on . Its geometric upper bound is an input from the companion papers. All release references below are pinned to commit e7c8972c701d3396947808ab20cc16dc14623d33. Ordinary ramp · Primitive ramp
span_of_dilates gives the general CM balance identity, the conditional per-host readout with its size bracket, and a statement-level account of the two-stage prime selection in rStation paper 192 v1, Theorems 4.7, 4.9 and 4.17. Its Lemma 4.20 outlines the centering step, for which we use the finite-tail estimate of Paper I. We specialize the same released argument to conductor one. Section 2 proves the progression and moment estimates in this family, including the growing-modulus step raised in the public review, and shows that the good pair can also satisfy . Section 4 constructs the thick and top sets and proves their orbit estimates. Section 5 writes the full translate readout as four signed ramps and retains the geometric bound as a hypothesis. Paper 192 v1 · Public review
1. The field and its character group
Proposition 1 (proved here). Let be an integer. Let be primes with , , , and . Put , for , and let Then is cyclic of order , is CM with , and has conductor . In the source's arithmetic frame the dyadic level is and the level index is one. Field input and output · Kernel characterization · Frame and Artin map
Proof. The three moduli are coprime. Classical cyclotomic Galois theory, the quadratic Gauss sum, and the Chinese remainder theorem give a surjection Its fixed field is . Equivalently its kernel consists of the classes with and . This is exactly the fixed-field condition IsChiField for the character of conductor one. An ambient cyclotomic field of any larger admissible modulus gives the same fixed field by reduction of units. The group is generated by the class of : the identity follows by repeated squaring, and the distinct powers of give all units congruent to . A faithful character of detects and hence has conductor . The characters have conductors . Their least common multiple with is the conductor of . In the release , which also gives . Complex conjugation negates both square roots and fixes the real cyclotomic field. Its fixed field is totally real, and is totally imaginary quadratic. Artin map · Dyadic divisibility · CM type
For the averaging argument write the two bits in the order . The source's Artin map stores the -sign first and the -sign second, and corrects its last coordinate by the class of the -sign. This correction is trivial in at conductor one. Thus in our coordinates the group is , complex conjugation is , and is the group over . It indexes the CM embeddings after a choice sending to . Its cardinality is . In the release , the subgroup used for translates, has index . Since the powers of the conductor-one character form a singleton, , the kept set is , and . Every translate has full top share, because for all and the prescribed share is . Original coordinate order · CM type · Colour and top coordinates · Kept set and shares · Index
Ramification and splitting (proved here). For , the three components of Frobenius are the two Legendre signs and . The inertia at is the order-two subgroup generated by , since the other two factors are unramified there. Both inertia and a lift of Frobenius at and belong to : is unramified and split there by and , the latter from . For every odd prime the same congruence gives . For , quadratic reciprocity identifies with the primitive character of discriminant , including the value at by the supplementary law. The -row therefore has the eigenvector with the specified sign, for , and the condition on odd primes dividing is vacuous. These are the ordinary specialization of the field and splitting rows. Pair to splitting rows · Field rows · Host inertia conventions
If is a character of , its primitive Dirichlet character is the product of the primitive representative of its -factor and the indicated powers of . Indeed the product agrees on the units by the Artin map, and primitive characters inducing the same common-level character agree. Equivalently at its value is the average of over , where is inertia and is a lift. If the restriction to is trivial, this is its unramified value. Otherwise the average vanishes by orthogonality and the primitive local Euler factor is absent. In particular the principal representative takes value also at . These primitive values give the same -functions and centers at as the corresponding host characters. Primitive value and balance definitions · Primitive twist comparison
2. The auxiliary quadratic primes
Fix and a real . At a real scale put , , , . Choose in with . For each such take in in the reduced class specified by Eventually , , and uniformly by the elementary Chebyshev bound for . The boxes use heights and . For every sufficiently large the released selection gives the following inequalities for one pair, uniformly at every real : It also gives for , for every primitive real of conductor dividing and every . All these products are primitive at their product conductors. The extra real bases are needed for estimates involving the full host dual. Progression · Window and box predicates · Export of estimates
Progression counts and the exceptional kernel
Published inputs. For a modulus divisible by , apply Thorner and Zaman's relative prime-counting theorem to , with the full Galois group as its abelian subgroup. The fixed field is , and the largest character conductor is at most . A reduced class is a singleton Frobenius class. For , the theorem gives Here , and the term with the real nonprincipal character and its real zero is omitted if the theorem specifies no such pair. The character at level is induced by a primitive real character. The release proves a relative progression estimate of the same form under . The argument below fits this latter range as well. We also use Page's theorem in its conductor-uniform form: among primitive real characters of conductor , at most one can have a zero in for an absolute . Finally Siegel's theorem gives for a primitive real character of conductor , with an ineffective constant. At a larger modulus the additional Euler factors have no zeros at a positive real , so both zero estimates apply to the inducing character. Thorner–Zaman, Theorem 1.4 · Page's theorem · Page's theorem in the density paper, Corollary 2.5 · Released progression bound · Release assembly · Released Page argument · Siegel input
Write , , and . In stage one let . In stage two, with fixed in its candidate set, use . In each case , and is reduced. The stage parameters are , , and any fixed discard share . The height constraints are For the fixed modulus , these constraints follow from and . For the growing modulus , eventually. Hence . Also and at the second stage, giving the other two bounds. In addition, the statement of stage requires every prime divisor of to satisfy or . The only prime divisor at stage one is , and at stage two and the primes in the product are at most . The remaining prime satisfies eventually. Thus the divisor primes that can support a term of the short sum are at most , as used in the second moment below. Stage statement · Scale and modulus conditions · Modulus bound · Second-stage verification
At stage one , so ordinary Siegel–Walfisz is available at this stage. Its log-power saving can be chosen to absorb both the classes in the fiber and the desired factor . At stage two even the base modulus exceeds every fixed power of , since and . We use the relative estimate above at the full modulus in both stages, as in the release's twistK. Indeed . Eventually at stage one and at stage two, so in either case. Since and , this gives for any fixed and . The same bounds give eventually, so every in exceeds . These estimates are uniform in with fixed. Released relative progression estimate · Range comparison · Public review of paper 192
Lower bound for the candidate sets (proved here). At , if the exceptional term is subtracted, then The first inequality follows by integrating on , where the possible exceptional is near and eventually. With , Siegel's bound gives . Dividing by yields a lower bound for the main term in all three cases. The relative error is even in the subtractive case. Because , subtracting at most small primes leaves eventually. The estimate holds uniformly in the allowed first primes. Released candidate count
Progression correlation (proved here). For odd squarefree coprime to , let on . Its conductor is exactly , since its local character at each prime dividing is nontrivial. On the reduced classes modulo lying over , sum the progression theorem weighted by . The ordinary main term sums to zero. The Chinese remainder theorem writes , with the first factor constant on this fiber. The exceptional main term sums to zero unless , by character orthogonality on the second factor. In the remaining case the base conductor of has prime-to- part exactly . Thus at most one can have both this matching condition and , by Page's theorem for the inducing characters of conductor . The Page threshold is common to all , and the exceptional kernel is independent of the short-sum coefficients and cutoff . Release exceptional-kernel argument · Orthogonality on the progression
For each , the exceptional term either cancels or has , by and . This ratio and are both after reducing . To control the error relative to the actual number of primes, sum the nonnegative main terms over the fiber. When the exceptional term matches , its unweighted sum is zero, so the total main term is . If the exceptional term does not match, the weighted main sum is already zero and the sum of the absolute errors is at most times the total main term, even when the latter is small. In both cases the total main term is within a factor of the unweighted prime count in that fiber up to . The count can omit primes dividing , all of which are below . Removing the other primes through changes either sum by at most , with . Thus for all sufficiently large , for every such , This is the bound needed from twistK. Correlation statement · Normalization by the actual progression count · Final subtraction
The second moment
Proposition (proved here). There is an absolute such that, at either stage for every sufficiently large , every real , and every complex sequence , one has The onset can depend on and is uniform in the allowed at the second stage. This is the ordinary specialization of the released moment, including its uniformity over arbitrary bounded coefficients. Moment statement · Source moment expansion
For in the sum, let be the product of the primes outside occurring to odd exponent in . Since , no prime dividing equals . Quadratic reciprocity and the supplement at give The sign is constant on , since the residues of modulo and modulo each are fixed. Each is odd, squarefree, coprime to , and at most . On expanding the square and applying , the upper bound divided by is The first two terms use the trivial bound for the correlation, with the second term zero when Page's exception is absent. The final term uses (1a). By the Chebyshev estimate and partial summation, . Thus the last term is , which is times an absolute constant eventually. Squarefree kernel · Source application of the correlation bound
Only prime powers have positive mass. Put For , the powers cost at most , and the remaining geometric series at most . For , . So . The same-prime contribution is bounded by . Indeed for use the bound , for use and Chebyshev, and for , , use The latter estimate follows by taking one factor at its upper bound and using Chebyshev on the remaining . Its sum gives the desired contribution for . Diagonal estimate
The mass of all even powers is at most , since and Chebyshev applies first to and then on the dyadic intervals above it. The powers on primes dividing cost by the individual prime bound and the bound on primes that divide and occur in the sum. Thus because . Even powers and modulus primes
To classify the pairs, each prime power contributes a kernel label. It is for the mass in , or its prime base for an odd power with . The pair kernel multiplies labels modulo squares. It equals only for two label- terms or two terms with the same prime label, giving a total weight at most . If is prime, its only pairs join that label to label and have weight at most . If is a product of two distinct primes, its pairs join those two labels and have weight at most . No other occurs among the pairs. In particular the exceptional weight is always . Combining these estimates proves (1b). Kernel pair cases · Released moment assembly
The maximal estimate and the two discards
For with , the finite ramp is continuous and absolutely continuous on the block, with right derivative The bounded coefficients in (1b) may depend on . Thus almost everywhere . Absolute continuity and Cauchy–Schwarz bound each square increment from , uniformly at every point in the block, by . Hence Put . The fraction violating anywhere on one block is . Summing over the blocks beginning at and stopped at gives , using and the geometric series. This is the source's finite-block argument and works at both endpoints of the full interval. Derivative and maximal argument · Dyadic sum
Thorner and Zaman count zeros with multiplicity over the primitive characters of conductor at most , including a possible exceptional real zero. Their counts and . For and , Theorem 1.2 gives . A character failing the box contributes at least one zero. Even if the box includes its left edge , every such zero is counted at each . Taking on the right bounds the number of discarded characters by . Since , this gives the required discard. The release obtains a bound of this order by zero detection with a mollifier. Published definition (1.2) and Theorem 1.2 · Released box count
At stage one the coefficient families are the primitive representatives of the characters modulo . At stage two they are the primitive products with for the already selected . Their conductors are respectively and . There are at most four and eight coefficients, respectively. In the fresh variable , the product with is primitive of conductor by coprimality and the Chinese remainder theorem. With , , one has eventually. Choose one bad character of conductor for each that fails the box for any coefficient. A conductor cannot witness two different primes , since both would then divide the conductor and each cofactor would exceed . Hence the density discard has relative size . By (1c) and the finite union over coefficients, the total fraction removed in either stage tends to zero. Coefficient families · Primitive twist and union · Conductor injection
Centering at and selecting both primes
The release uses the contour bound tail_contour for the centering step. Paper I, Proposition 3 proves the finite-tail bound needed at the two stage heights directly from the classical local partial-fraction estimate. The release’s contour inequality below assumes a nonprincipal , conductor at most , the box above through height , , , and a fixed . Put . In this range the bound is Its proof in the release shifts a smoothed Perron difference to , using a bound for inside the box and a truncation at height . The following calculation records the use of this stronger contour inequality in the release. The proof via Proposition 3 of Paper I gives the required prime selection independently of this inequality. The elementary difference quotient of differs from by at most , by summing only the new integers in . Consequently the finite tail satisfies Tail contour estimate · Elementary difference
At take , , . The stage heights obey eventually. Multiplying (1d) by , the first term is , the second is , and the third is . Their sum is at most eventually. For the primitive product , finite splitting at gives Because , the tail costs at most at large , leaving the short sum controlled by the maximal estimate. Release tail parameters · Box from short sum
For the logarithmic derivative bound use (1d) again with , , and . The first term is , the second is , and the third is . The classical bound then gives . At both stage heights . Since , this gives eventually. This use of the original follows from the actual choices of . The enlarged parameter is used for admissibility of the second stage. Released derivative estimate · Center and primitive handoff
Alternatively, after the ramp bound at the left endpoint has been obtained with Proposition 3 of Paper I, the identity defining the ramp and Chebyshev’s bound give directly. Finite-tail proof and center estimate
We have proved the two relative share assertions used below. For every fixed , every sufficiently large real has a first-stage subset of share at least . For each in that subset, its own second-stage good set has share at least . The thresholds are uniform in . The selection control comes from separate box and maximal discards at each stage. Choosing first and then gives (1), using the trivial , and gives all the real-base boxes using the full primitive family from modulus . These are the conductor-one contents of GoodQ, GoodD and PairGood. Quantified shares · Stage union · Final output
Selecting a pair with prescribed quadratic class numbers
Proposition 2 (proved here using classical and published inputs). For every as above, for every sufficiently large the pair in (1) can be chosen with , where is the class number of the imaginary quadratic field of fundamental discriminant . This refinement uses Stevenhagen's Rédei reciprocity and the Thorner–Zaman effective Chebotarev theorem. It strengthens the released pair choice for this specialization. Good fractions · Rédei reciprocity, §§3–4 and 7 · Effective Chebotarev, Theorem 1.4
Proof. Write with . The norm-Euclidean ring gives this representation because . For a root , define . The root exists because and quadratic reciprocity gives . The two possible arguments have product , a nonzero square since and . The sign is independent of the root. Pair congruences
Use the three ramified primes as columns of the Rédei matrices, and order the rows by the discriminants or . Encoding as , the two matrices are For example, . The other off-diagonal entries follow from the signs of at , and each column sums to zero. Genus theory gives -rank two and -rank one, so the -primary class group has the form , . In both fields the product of the ramified ideals over is the nonzero element of . Its vector belongs to each kernel and differs from the principal relations and from the quadratic radicals. The genus character corresponding to spans the genus characters admitting a quartic lift. The one-entry -rank matrices therefore have entries and in additive Rédei notation. Genus and -rank formulas
Rédei reciprocity and linearity identify both entries with . In detail, the second-kind identities give and , so Here the required Hilbert pairings are . At the odd primes their values are by , , and with . The real symbols are trivial, so the product formula gives the dyadic values. Also and . These identities verify the pairings for both applications of linearity. The three discriminants in each triple have common gcd one. The second-kind decompositions and give the two vanishing symbols by Proposition 7.10. For the two class-group entries, the associated cyclic quartic extension of the quadratic base is unramified. The Artin action of the ideal of norm in Definition 4.4 therefore gives the symmetric symbol of Definition 7.8 with second squareclass or . Rédei reciprocity and the vanishing symbols, Theorem 7.9 and Proposition 7.10
Let , , and . The norm of shows that neither that element nor its quotient by is square in , hence . Over , the map , has square equal to the nontrivial automorphism over . The involution negating while fixing and the chosen square root conjugates that order-four map to its inverse. This gives the dihedral extension over for the pair . Away from the conjugate radicands have disjoint prime supports and their product has even valuation, giving the required unramified condition. The discriminants are respectively , so Definition 7.6 imposes no further condition at for this pair. The field is totally real since both radicands are positive. Thus the infinite-place factor in the Rédei symbol for the negative third argument is trivial. The prime splits in , and its relative Artin symbol in is nontrivial precisely when . An entry gives -rank zero, hence in each of the two -class groups. Dihedral interpretation, Definitions 7.6 and 7.8
To count this sign let , with from the second stage. It contains , and because is the maximal abelian subfield of . The automorphism of fixes , so its two lifts and to are central. The group is abelian. In their quotient is a square. Indeed choose the order-two automorphism of acting by on the -factor and trivially on the complementary factor. It fixes and negates , hence negates . Pair with a generator of , which has the same restriction to . Then . In particular every real character of takes equal values on and . Progression automorphism · Exceptional term in effective Chebotarev
The finite Hecke conductors for divide . For , odd ramification is tame at and absent elsewhere. At the dyadic prime of , the -adic logarithm identifies the fourth powers of with , since . A quartic character is trivial there and . For , the characters extend to Dirichlet characters at modulus and pull back by the norm. A character on extends to the product of the two abelian groups over , giving the same conductor bound. Thus the maximum conductor norm is at most , and has discriminant and degree . Abelian conductors and effective Chebotarev
Apply effective Chebotarev with this common abelian subgroup and the two singleton classes, at . The size threshold holds since and . The main terms, including the possible exceptional term, agree by the square calculation. Each has relative error with independent of . All ramified primes divide , so the two counts add to the progression count. Removing primes through changes either count by at most , which is for the second-stage candidate set . Thus each sign has relative share uniformly in the first prime. Choose , take any first-stage good prime and intersect its second-stage good set with the negative sign. The share is at least , proving the assertion. Effective Chebotarev, Theorem 1.4 · Good fractions
3. From prime selection to real cutoffs
Cohomological inputs (specified in Paper VI). For a prime write for the kernel of localization in degree one with coefficients , for the Galois group unramified outside places over , localizing at all those places. The source uses all algebraic closures and all choices of those places. The requirement VVle at asserts that each of these lengths over is finite for , and that for every finite prime set , For odd the two eigenspaces for complex conjugation have signs and . Their lengths add. Below, denotes the conductor of . Localization kernel · Uniform length bound · Eigenspaces and assembly
The release uses the following three bounds, with the cutoff before rescaling and . It takes the sum of their budgets.
- The dyadic input
BaAtgives the unweighted length sum at at most . - The odd reference input
BbAtgives the same separate bound over each fixed reference layer. At conductor one these are layers over and . With these bounds,PointHVgives a bound on the weighted sum for the odd eigenspace of sign , whenever , , , and both primes exceed the fixed exclusions for the level and reference layers. - The pointwise input
ThmBcFreebounds the other odd weighted sum byminusBoundfor every such field and positive rank range with the prime congruences. In the size range its bound is at most , with fixed.
The bounds include finiteness of each length they cover. At conductor one, Paper VI §3.4 proves the odd minus estimate by the equivariant Coates–Sinnott theorem for every odd coefficient prime. The odd plus estimate still needs the surviving deformation-theoretic cases of Paper VI. The release aggregates the two reference conditions and the dyadic bound on a common interval via ThmBab, then applies PointHV there. Its VWinAt leaves the starting point of this common interval dependent on the given pair. Three interfaces · Odd pointwise hypotheses · Proof inputs for the joint window · Budget assembly · Paper VI
Arbitrarily late windows (rescaling proved here). Fix , put , and take for . Thus , with for . The released VWinAt input at exponent fixes and before and the pair are chosen. With , it supplies, for every eligible pair at scale , a with and , and (2) on all with . The proof from the three component estimates uses and , to fit the three budgets below . The lower bounds for also give and every fixed exclusion eventually. Window quantifiers · Conductor and total budget
Choose the pair by (1) with before taking . In the ordinary case the scale factor in winPairs is , since . Thus for one has , , , and (2) with by monotonicity. These are the fields of WinDatum. The starts are unbounded. Paper VI gives the general rescaling. Released rescaling · Output data · Paper VI
A conditional proposition for prescribed starts at conductor one
Fix , , and the admissible primes of Proposition 2 at scale with . Write and for . Let denote the norm limits of their ordinary -primary ideal class groups, and let be the cyclotomic generator of value . Put , so .
Proposition 3 (conditional). Assume the following inputs on the pairs so selected, with the pointwise bounds uniform in .
- The modules are free over of ranks two and three. If denotes the action of on , its characteristic polynomial on is . Only this polynomial identity is needed for the determinant calculation.
- The ordinary -class groups of are the coinvariants of by , respectively. Their orders are the -parts of and , respectively. Here is the class number of the indicated quadratic discriminant.
- For , restriction injects the strict localization kernel from into the kernel at . The dual of the latter is a quotient of obtained by imposing splitting at the places above .
- Descent gives the reference localization bound in the - and -power towers above. The pointwise odd plus and minus bounds, including finiteness, hold for the chosen host with the budgets following (2).
Then . Every sufficiently large prescribed admits one pair giving the arithmetic data and (2) at every real , with twist limit and budget . The geometric construction at those cutoffs enters in §4.
Chems-Eddin and Müller give the minus -invariant and oddness of the real-layer class numbers used in the tower calculation. Paper VI §3.3 gives the additional integral rank, class-number, and restriction arguments for inputs 1–3 and treats the odd reference fields. Chems-Eddin–Müller, Theorems 4–5 · Paper VI · Strict dyadic descent · Released layer comparison
Proof. Input 2 gives . Subtracting this valuation from gives . With the selected sign, both determinant valuations are . If and , then implies . Thus the characteristic polynomial of has coefficients in and reduces to . Each eigenvalue of is with . For all positive integers , both and have valuation one. The first squaring gives valuation two for the difference from , and each further squaring adds one. At layer the two eigenvalues therefore contribute to the determinant valuation. The trivial factor contributes . The resulting determinant has finite valuation, which is the length of the cokernel on the free lattice . Input 3 now gives the asserted upper bound on the host. Summing gives uniformly for , since . This fits below for every sufficiently large . Twist and layer bounds · Dyadic budget
Paper VI proves that the odd reference towers over and have zero primary class groups, using Minkowski's bound and the ambiguous class number formula along the tower. The reference localization lengths are therefore zero by descent. The two pointwise odd host estimates still enter (2). Paper VI · Reference fields · Descent comparison
Fix and apply Proposition 2 with and . For every sufficiently large , the dyadic and reference estimates now hold throughout . The pointwise odd estimates and the size checks above give (2) there at exponent . For a prescribed sufficiently large take and . The resulting pair gives WinDatum on , including . Pointwise height application · Cohomological window · Prior cutoff proof
4. Fitting the geometry at the same cutoff
Arithmetic data for the host
Fix the permitted constants and take below the common existence cap. At each sufficiently large cutoff supplied by §3, use the field and frame of §1 with that cutoff's . The rank input supplies lifts, columns, and twist limit , with eventually. The size identity for the remaining ranks is the last equality of RankVal. The host extension needs and the small-prime splitting rows through . The first follows from the rank size bounds and the lower bounds on when . The second follows from the progression in §2. Restrict (2) to the actual to obtain the cohomology input with bound for . Paper VI §4 derives the norm primes in short intervals from the effective Chebotarev theorem over , chooses the distinct level primes, and takes the unit ideal for the prescribed ideal of . The character estimates for the nontrivial quadratic twists use the pair bounds of §2. With these arithmetic data fixed, the following constructions supply the two sets on this frame. Rank conditions · Twist limit · Admissibility · Host extension · Paper VI, §4
Thick digits and translated orbits
Use the coordinates of §1 on , where and . Write for the set of nonnegative integers stored as the host's thick digits and for their image. The required length is Each must satisfy , so the digit map is injective. The condition excludes for pairwise distinct and distinct integers with . It excludes three-term progressions and triples with unequal gaps satisfying the same coefficient bound. The source requires an absolute with The raw digit interval permits and excludes . Host data · Bounded relations and orbit conclusions · Length · Required sizes
Construction. For integers and , put and . If or , take . Here , using for . Hence .
In the other case put , , , and . Then , , and . Encode by . Take the encodings of all vectors of a most frequent squared length. The encodings are distinct and lie in . There are at most squared lengths, so this gives at least digits.
For the forbidden relation with coefficient bound , each coefficient at a power of has absolute value at most . Reducing modulo forces the first coefficient to vanish. Dividing by and repeating gives the vector relation . The three vectors would be distinct collinear points on the same sphere. Squared distance to the center along a line is strictly quadratic, so there are at most two such points.
Since , the cardinality satisfies Using , , and gives the required lower bound with . Both constructions have when , so in that range. At , we have and . Thus uniformly. This gives eventually. Sphere construction · Density used in the comparison
For and , every half-open integer window contains at most two indices with . Three hits would give distinct digits and The absolute value on the left is by the digit range, contradicting after lifting to . The argument includes wraparound in . If , the whole orbit meets in at most one distinct point. Two hits give , and the digit range forces . This point can recur in the index sequence. Also, for every there are exactly translates with , using for . Translated orbit bounds · Translate mean
Tops and weighted Frobenius-orbit estimates
Put , , and fix . We need and the weighted bad-orbit estimate below on this same set. The source's coset-share condition measures the discrepancy from proportions on the identity coset and on other cosets, with tolerance . At conductor one the only coset is , so the discrepancy is zero for every translate. The orbit tests for allow three-term progressions. Top selection · Coset shares
Set and let be the primes in the host's residue-zero class for which . Here is the cyclotomic coordinate of arithmetic Frobenius. The prime weights are . By the small-prime splitting in §2, every inert-class prime exceeds . Consequently the short window length satisfies The supremum is over all inert-class primes. The rank bound gives eventually, and gives . Prime class and weighted filter · Short window length
For each orbit , use its fixed representative in the source and number the positions in steps of . Use the parameters The source's good-orbit predicate is the conjunction of the following three tests on the indices selected by . Parameters · Short-window test · Tile and interval tests
- Every cyclic window of indices contains at most selected indices.
- Use the complete tiles of indices starting from . Each tile has cells of length , numbered starting at . A tile is marked if every odd cell is empty and every even cell contains exactly one selected index of even parity in the orbit numbering. For every tile index , one of the tiles , , is marked, with addition modulo the number of complete tiles.
- Every cyclic window of indices contains at most selected indices.
Simultaneous selection. Select each point of independently with probability . For fixed and , condition on selecting . The other indices in its orbit still have independent Bernoulli laws. Uniformly for the orbits tested, and are and less than eventually. Also . Thus individual windows have distinct indices, and the trials in test 2 use distinct tiles.
For test 1, a window containing the conditioned point requires further selections to fail, and at most windows contain that point. Each other failing window requires selections. The union bound is at most as and .
For test 2, discard the trial tile containing the conditioned point if there is one. The other at least tiles are independent. An empty cell has probability . A cell with exactly one selected index, of even parity, has probability at least . For , the first exceeds , by and . The second exceeds , since . The marking probability is at least . Summing over the at most starting tiles, failure costs at most .
For test 3, the conditioned count in a window is dominated by . With , the binomial tail at is bounded by This uses . Multiplying by the at most windows gives at most , since .
Write for the number of selected tops on bad orbits at . The conditional failure probability for each is at most for , independent of . Hence By Markov the chance of exceeding four times this bound is at most , with zero weighted count almost surely if the bound is zero. The size has mean and variance at most . Chebyshev gives probability at most of leaving . Since , one choice satisfies the size interval, the exact ordinary coset shares, and This is the source's estimate. To obtain the endpoint estimate's bound , use and . Indeed, with , the prime contributions to give this prime-weight bound using . Group selection · Weighted condition after transport · Endpoint allowance
The marked tiles supply the endpoints on a good orbit. Each contains even selected indices, at least from the left endpoint, with strictly before the right endpoint and successive gaps for large . For successive marked tiles, write their increasing lists as , indexed from , and use and . Lift the latter by on wraparound. Apply test 2 starting at the tile immediately after the first marked tile. The next marked tile is at most steps away, and there are more than complete tiles because . The final incomplete tile, of length , can occur once in the segment. Thus the forward endpoint distance is , and the inclusive segment contains at most selected indices. Marked positions · Segment endpoints
Carrying the sets into the geometry
The host installs at offset and then chooses using the prime and orbit data. Its top places are for . At translate , the thick places are for . Their distinct quadratic colors make the two supports disjoint for every , even when their projections in intersect. The stated orbit and density bounds apply to the same at all translates. In §5 the thick partner coordinate covers once as varies with fixed. Paper VI transports these choices into the kit and family, and uses the polynomial prime bound for the extra level prime . Its weight existence, tame distinctness and common word-code inputs concern the fitted objects. Host places · Initial offset · Installation · Family design · Paper VI, §4
The local theta estimates use the thick-orbit test on the letters . If the cyclotomic step has order , the thick letters satisfy the at-most-two condition in every window of indices. At depth with and , three thick letters linked by successive gaps at most would lie in one such window. The twisted dual page-word estimate inserts the resulting cluster bound and into its local hypothesis , along with the numerical weight bounds. Thick-letter window · Cluster and word length · Local theta use
The remaining geometric input, from Papers IV and VI and the companion construction, supplies the fitted objects and period and norm estimates on this host. In particular the comparison uses positive heights for and positive weights at for every and . Write , , , , , and with fixed . With the fitted weight tolerance, the conditions are The first cost bound follows eventually because . The second uses the fitted . The last line is supplied by the two constructions above, and the fitting and period assertions apply to the same field, sets, lifts, weights and truncation. Scales · Digit density · Costs and ranks · Fitting to the chain · Paper IV
Thin estimate (Paper I, Proposition 4). At all , for every primitive character of the fitted host, including the principal case interpreted as , The conductor input in the release is , which follows eventually from , the bounds on , and . To verify the real-character input, a real character of the dyadic units is trivial on squares, hence on all units congruent to . Its primitive conductor divides . Multiplying by gives all real host characters by the Chinese remainder theorem. For a real character ramified at or , the box of §2 gives real nonvanishing above . The finitely many remaining nonprincipal real characters have a common interval below free of zeros by classical nonvanishing at and continuity. Together with the conductor bound, this common interval verifies the hypotheses of Proposition 4 of Paper I and proves (4). Uniform contour proof Thin input · Host bound and scales · Use on the fitted host
Preserving the domain (proved here from the stated pointwise interfaces). Fix before fitting and let be the set of all admitting WinDatum with these constants. Host supply uses exactly and requires for a fixed threshold. The enlargement of the level-prime set, object construction and weight construction use this same scale set and can increase the threshold. The family built from kits has and uses the actual scale above its threshold for every . The later closed conditions are requested at the joined parameter bound before the family is chosen. Their conclusion gives one set of constants and a further threshold on that same family. Thus the geometry can be invoked at every sufficiently large once these pointwise existence inputs are supplied. On the universal window route, contains arbitrarily late power windows. With the refined choice in §3, contains an entire tail. The latter property is carried by the concrete family construction through its equality of scale sets. Choice of scale set · Releveling and weights · Weight construction · Kit indexing · Joined conditions · Same-point identity
5. The finite prime-power identity and the thick mean
Fix a fitted cutoff with . For any character let and use its primitive representative, treating the principal ramp as . For take a positive weight , and suppose and for all . For a top write and , a bijection from to . The finite prime sum uses the factor at a class . For it takes with coefficient . For it sums over the complementary classes in with coefficient . At its contribution vanishes. Only can contribute. Define The finite balance is Finite definitions and truncation · Released balance
Proof of (5). First use the uncentered prime ramps. Fourier inversion gives, at , the coefficient since . The inertia and splitting calculations of §1 now give all the displayed prime coefficients. The principal center subtracted from the uncentered sum is , averaged over the partners. For nonprincipal characters, the center added to the uncentered sum for a given is , independent of the partner's weight. The sum of on is zero unless is trivial on . The unique nonprincipal character trivial there is , with value on the complementary coset, so its sum is . This accounts for in upon subtraction. Balance proof and normalization
For each use in (5). The thick partners are for . Thus their classes in (5) are , each occurring once. Put , and let denote the right side of (5) with thin partners omitted and all thick weights replaced by . Summing over kills the characters nontrivial on . For each surviving character , the phase in every summand is its value at , independent of . In order the four signs are . Consequently (proved here) The four terms are real. The factor persists after taking the mean. Thick partner bijection · Fourier averaging
The comparison error
We first bound the change at a fixed , with . For , the function is decreasing and -Lipschitz. It has zero contribution for at either weight. For a nonprincipal character the two centers cancel in the weighted difference. Thus by the classical Chebyshev estimate the change has modulus at most . In the principal case the uncentered difference from to , taken as its value at minus its value at , belongs to . From it one subtracts , which lies in the same interval. This proves the same bound in the principal case. At a thin partner, (3) gives because , , and . The omitted term has modulus at most by (4). Weight estimates · Thin and thick substitutions
There are thick and thin partners and characters in each term of (5). Thus the error at has modulus at most by (3). The factor comes from the full character sum and the denominator . Let at translate , and let denote its uniform mean. Put . Summing the error and using (6) yields The release also uses the weaker allowance by separately budgeting both errors using all partners. Stored cost comparison · Same-host excess and normalization
For the move to the common cutoff, the classical Mertens estimate implies away from integers, since the derivative of its center cancels . Each nonprincipal ramp has derivative of modulus at most . The ramps are continuous and piecewise differentiable, including at the newly admitted integer where the new ramp term vanishes. Since , (1) at gives, uniformly in the host, Together with (1) and (7) this proves the ordinary comparison (proved here) The sharper signed comparison uses (7) and the cutoff move alone. The bounds at suffice even when a particular lies outside the selected interval of analytic bounds. Derivative comparison · Three bounded characters · Principal term
The geometric upper bound needed at the interface
Supplied by Paper III and the geometric papers. On the fitted host, for fixed geometric constants and sufficiently large eligible , the same excess satisfies where are the allowances for the finite divisibility and norm losses. Here the geometric estimates control the finite divisor gain and its gap from the main term. A bound on with the same right side also suffices for (8). Error bracket · Host-specific chain bound · Paper III
To check the size explicitly, divide (9) by and use (3). The four terms in , in order, are The last term uses before the bound . With , the largest exponent after the factor in (9) is . In particular (8) gives the release's looser bound eventually, with a fixed absorbing the constants. It holds on the supplied late windows, or at every sufficiently large real with the refined selection and the pointwise geometric inputs above. The same and weights occur in (3)–(9). Geometry and arithmetic at the same scale · Released scalar conclusion · Rank calculation