Explicit Tate-to-Sutherland coordinates for
DefinitionRationalPointsN25SutherlandBridgemazur-n25
Starting from Tate parameters, the file defines raw coordinates and through their cleared numerators, and the optimized coordinates
It records Sutherland’s bidegree- equation, the raw bidegree- equation by coefficient polynomials, and the relevant homogenized and denominator data. This supplies the explicit model comparison between the primitive Tate locus and Sutherland’s plane model.
Definition code
import Mathlib
import Mathlib.AlgebraicGeometry.EllipticCurve.DivisionPolynomial.Degree
import Mathlib.AlgebraicGeometry.EllipticCurve.DivisionPolynomial.Basic
import Mathlib.NumberTheory.EllipticDivisibilitySequence
import Mathlib.Tactic
import Definitions.Def_TateOrder25Factor
/-!
# The Sutherland source model for the order-25 Tate locus
Andrew Sutherland's explicit models of `X₁(N)` use the raw coordinates
```
r = b / c, s = c² / (b - c),
```
and then the birational optimization
```
x = (s - r) / (r*s - 2*r + 1),
y = (r*s - 2*r + 1) / (s² - s - r + 1).
```
This file verifies that comparison for `N = 25` directly against the primitive
Tate division polynomial `F25`. After simplifying the composite map, it proves
an unconditional denominator-cleared polynomial identity and proves that both
remaining denominators are nonzero on the exact-order locus. The endpoint is
therefore an explicit rational map from every primitive Tate solution to
Sutherland's bidegree-`(8,8)` plane model.
The formulas are kept as literal polynomials, and every model comparison below
is certified by `ring`; no modular-curve identification is introduced as an
assumption. This closes the Tate-to-optimized-source part of the N25 comparison.
-/
namespace MazurProof.RationalPointsN25SutherlandBridge
open TateOrder25Factor
set_option linter.style.longLine false
/-! ## The simplified Tate-to-Sutherland coordinates -/
/-- Numerator `X = c³ - b(b-c)` of the optimized coordinate
`x = (s-r)/(r*s-2*r+1)` after substituting `r=b/c` and `s=c²/(b-c)`. -/
def tateToSutherlandXNum25 (b c : ℚ) : ℚ := -b ^ 2 + b * c + c ^ 3
/-- Common numerator `K` of `r*s-2*r+1`. On the Tate chart,
`r*s-2*r+1 = K / (c(b-c))`; it is also the denominator of optimized `x`
and a numerator factor of optimized `y`. -/
def tateToSutherlandCommon25 (b c : ℚ) : ℚ := -2 * b ^ 2 + b * c ^ 2 + 3 * b * c - c ^ 2
/-- Numerator `B` of `s²-s-r+1` after the Tate substitution:
`s²-s-r+1 = B / (c(b-c)²)`. Hence the optimized second coordinate is
`y = K(b-c)/B`. -/
def tateToSutherlandYDen25 (b c : ℚ) : ℚ :=
-b ^ 3 + 3 * b ^ 2 * c - b * c ^ 3 - 3 * b * c ^ 2 + c ^ 5 + c ^ 4 + c ^ 3
/-- Numerator `c²-b+c` of `s-1 = (c²-b+c)/(b-c)`. It appears as the
third exceptional factor in the denominator-cleared model identity. -/
def tateToSutherlandSMinusOneNum25 (b c : ℚ) : ℚ := c ^ 2 - b + c
/-! ## Sutherland's optimized and raw equations -/
/-- Sutherland's bidegree-`(8,8)` affine equation for `X₁(25)`. The source
publishes this polynomial as `X1/FFFc25.txt`; the coordinate convention is the
optimized `(x,y)` convention stated in the module documentation. -/
def sutherlandX1Plane25 (x y : ℚ) : ℚ :=
x ^ 8 * y ^ 2 + x ^ 7 * y ^ 5 - 4 * x ^ 7 * y ^ 4 - x ^ 7 * y ^ 2 -
3 * x ^ 6 * y ^ 6 + 11 * x ^ 6 * y ^ 5 - 2 * x ^ 6 * y ^ 4 +
2 * x ^ 6 * y ^ 2 - 2 * x ^ 6 * y + 3 * x ^ 5 * y ^ 7 -
9 * x ^ 5 * y ^ 6 - 4 * x ^ 5 * y ^ 5 + 4 * x ^ 5 * y ^ 4 -
2 * x ^ 5 * y ^ 3 + 4 * x ^ 5 * y ^ 2 - x ^ 4 * y ^ 8 +
x ^ 4 * y ^ 7 + 9 * x ^ 4 * y ^ 6 - 5 * x ^ 4 * y ^ 5 +
6 * x ^ 4 * y ^ 4 - 15 * x ^ 4 * y ^ 3 + 7 * x ^ 4 * y ^ 2 -
2 * x ^ 4 * y + x ^ 4 + x ^ 3 * y ^ 8 - 4 * x ^ 3 * y ^ 7 +
3 * x ^ 3 * y ^ 6 - 14 * x ^ 3 * y ^ 5 + 26 * x ^ 3 * y ^ 4 -
15 * x ^ 3 * y ^ 3 + 7 * x ^ 3 * y ^ 2 - 5 * x ^ 3 * y + x ^ 3 -
x ^ 2 * y ^ 7 + 8 * x ^ 2 * y ^ 6 - 13 * x ^ 2 * y ^ 5 +
12 * x ^ 2 * y ^ 4 - 18 * x ^ 2 * y ^ 3 + 19 * x ^ 2 * y ^ 2 -
8 * x ^ 2 * y + x ^ 2 - 4 * x * y ^ 5 + 13 * x * y ^ 4 -
16 * x * y ^ 3 + 10 * x * y ^ 2 - 4 * x * y + x + y ^ 4 -
4 * y ^ 3 + 6 * y ^ 2 - 4 * y + 1
def sutherlandRawCoefficientTen (s : ℚ) : ℚ :=
1
def sutherlandRawCoefficientNine (s : ℚ) : ℚ :=
-s ^ 10 + 17 * s ^ 9 - 123 * s ^ 8 + 494 * s ^ 7 - 1205 * s ^ 6 + 1836 * s ^ 5 - 1732 * s ^ 4 + 968 * s ^ 3 - 294 * s ^ 2 + 35 * s - 5
def sutherlandRawCoefficientEight (s : ℚ) : ℚ :=
-6 * s ^ 10 + 74 * s ^ 9 - 345 * s ^ 8 + 690 * s ^ 7 - 185 * s ^ 6 - 1659 * s ^ 5 + 3051 * s ^ 4 - 2320 * s ^ 3 + 840 * s ^ 2 - 105 * s + 10
def sutherlandRawCoefficientSeven (s : ℚ) : ℚ :=
-21 * s ^ 10 + 161 * s ^ 9 - 351 * s ^ 8 - 144 * s ^ 7 + 1289 * s ^ 6 - 789 * s ^ 5 - 1551 * s ^ 4 + 2166 * s ^ 3 - 996 * s ^ 2 + 126 * s - 10
def sutherlandRawCoefficientSix (s : ℚ) : ℚ :=
s ^ 15 - 18 * s ^ 14 + 151 * s ^ 13 - 770 * s ^ 12 + 2655 * s ^ 11 - 6558 * s ^ 10 + 11834 * s ^ 9 - 15408 * s ^ 8 + 14630 * s ^ 7 - 11195 * s ^ 6 + 7227 * s ^ 5 - 2441 * s ^ 4 - 388 * s ^ 3 + 555 * s ^ 2 - 70 * s + 5
def sutherlandRawCoefficientFive (s : ℚ) : ℚ :=
s ^ 15 - 15 * s ^ 14 + 90 * s ^ 13 - 245 * s ^ 12 + 90 * s ^ 11 + 1587 * s ^ 10 - 6145 * s ^ 9 + 12270 * s ^ 8 - 15060 * s ^ 7 + 12520 * s ^ 6 - 8214 * s ^ 5 + 3660 * s ^ 4 - 685 * s ^ 3 - 120 * s ^ 2 + 15 * s - 1
def sutherlandRawCoefficientFour (s : ℚ) : ℚ :=
s ^ 3 * (s ^ 12 - 12 * s ^ 11 + 48 * s ^ 10 - 49 * s ^ 9 - 165 * s ^ 8 + 609 * s ^ 7 - 433 * s ^ 6 - 1623 * s ^ 5 + 4299 * s ^ 4 - 4615 * s ^ 3 + 3435 * s ^ 2 - 1740 * s + 455)
def sutherlandRawCoefficientThree (s : ℚ) : ℚ :=
s ^ 3 * (s ^ 12 - 9 * s ^ 11 + 25 * s ^ 10 - 35 * s ^ 9 + 45 * s ^ 8 - 181 * s ^ 7 + 569 * s ^ 6 - 705 * s ^ 5 + 5 * s ^ 4 + 470 * s ^ 3 - 540 * s ^ 2 + 340 * s - 105)
def sutherlandRawCoefficientTwo (s : ℚ) : ℚ :=
s ^ 3 * (s ^ 12 - 6 * s ^ 11 + 21 * s ^ 10 - 56 * s ^ 9 + 126 * s ^ 8 - 231 * s ^ 7 + 266 * s ^ 6 - 126 * s ^ 5 + 96 * s ^ 4 - 91 * s ^ 3 + 75 * s ^ 2 - 45 * s + 15)
def sutherlandRawCoefficientOne (s : ℚ) : ℚ :=
s ^ 3 * (6 * s ^ 7 - 28 * s ^ 6 + 21 * s ^ 5 - 15 * s ^ 4 + 10 * s ^ 3 - 6 * s ^ 2 + 3 * s - 1)
def sutherlandRawCoefficientZero (s : ℚ) : ℚ :=
s ^ 10
/-- Sutherland's raw equation `F(r,s)=0` for `X₁(25)`, before the universal
birational optimization. Its bidegrees are `10` in `r` and `15` in `s`.
This is deliberately distinct from the repository's Tate polynomial `F25`. -/
def sutherlandRawX1Plane25 (r s : ℚ) : ℚ :=
r ^ 10 * sutherlandRawCoefficientTen s +
r ^ 9 * sutherlandRawCoefficientNine s +
r ^ 8 * sutherlandRawCoefficientEight s +
r ^ 7 * sutherlandRawCoefficientSeven s +
r ^ 6 * sutherlandRawCoefficientSix s +
r ^ 5 * sutherlandRawCoefficientFive s +
r ^ 4 * sutherlandRawCoefficientFour s +
r ^ 3 * sutherlandRawCoefficientThree s +
r ^ 2 * sutherlandRawCoefficientTwo s +
r ^ 1 * sutherlandRawCoefficientOne s +
r ^ 0 * sutherlandRawCoefficientZero s
def sutherlandHomogeneousCoefficientEight (K D B : ℚ) : ℚ :=
B ^ 6 * D ^ 2 * K ^ 2
def sutherlandHomogeneousCoefficientSeven (K D B : ℚ) : ℚ :=
-B ^ 3 * D ^ 2 * K ^ 3 * (B ^ 3 + 4 * B * D ^ 2 * K ^ 2 - D ^ 3 * K ^ 3)
def sutherlandHomogeneousCoefficientSix (K D B : ℚ) : ℚ :=
-B ^ 2 * D * K ^ 3 * (2 * B ^ 5 - 2 * B ^ 4 * D * K + 2 * B ^ 2 * D ^ 3 * K ^ 3 - 11 * B * D ^ 4 * K ^ 4 + 3 * D ^ 5 * K ^ 5)
def sutherlandHomogeneousCoefficientFive (K D B : ℚ) : ℚ :=
B * D ^ 2 * K ^ 5 * (4 * B ^ 5 - 2 * B ^ 4 * D * K + 4 * B ^ 3 * D ^ 2 * K ^ 2 - 4 * B ^ 2 * D ^ 3 * K ^ 3 - 9 * B * D ^ 4 * K ^ 4 + 3 * D ^ 5 * K ^ 5)
def sutherlandHomogeneousCoefficientFour (K D B : ℚ) : ℚ :=
K ^ 4 * (B ^ 8 - 2 * B ^ 7 * D * K + 7 * B ^ 6 * D ^ 2 * K ^ 2 - 15 * B ^ 5 * D ^ 3 * K ^ 3 + 6 * B ^ 4 * D ^ 4 * K ^ 4 - 5 * B ^ 3 * D ^ 5 * K ^ 5 + 9 * B ^ 2 * D ^ 6 * K ^ 6 + B * D ^ 7 * K ^ 7 - D ^ 8 * K ^ 8)
def sutherlandHomogeneousCoefficientThree (K D B : ℚ) : ℚ :=
K ^ 5 * (B - D * K) * (B ^ 7 - 4 * B ^ 6 * D * K + 3 * B ^ 5 * D ^ 2 * K ^ 2 - 12 * B ^ 4 * D ^ 3 * K ^ 3 + 14 * B ^ 3 * D ^ 4 * K ^ 4 + 3 * B * D ^ 6 * K ^ 6 - D ^ 7 * K ^ 7)
def sutherlandHomogeneousCoefficientTwo (K D B : ℚ) : ℚ :=
B * K ^ 6 * (B - D * K) ^ 2 * (B ^ 5 - 6 * B ^ 4 * D * K + 6 * B ^ 3 * D ^ 2 * K ^ 2 + 6 * B * D ^ 4 * K ^ 4 - D ^ 5 * K ^ 5)
def sutherlandHomogeneousCoefficientOne (K D B : ℚ) : ℚ :=
B ^ 3 * K ^ 7 * (B - D * K) ^ 3 * (B ^ 2 - B * D * K + 4 * D ^ 2 * K ^ 2)
def sutherlandHomogeneousCoefficientZero (K D B : ℚ) : ℚ :=
B ^ 4 * K ^ 8 * (B - D * K) ^ 4
/-- The `(8,8)` bihomogenization of `sutherlandX1Plane25` evaluated at
projective pairs `[X:K]` and `[K*D:B]`. It is polynomial even on the boundary
where either affine denominator vanishes. -/
def sutherlandBihomogenized25 (X K D B : ℚ) : ℚ :=
X ^ 8 * sutherlandHomogeneousCoefficientEight K D B +
X ^ 7 * sutherlandHomogeneousCoefficientSeven K D B +
X ^ 6 * sutherlandHomogeneousCoefficientSix K D B +
X ^ 5 * sutherlandHomogeneousCoefficientFive K D B +
X ^ 4 * sutherlandHomogeneousCoefficientFour K D B +
X ^ 3 * sutherlandHomogeneousCoefficientThree K D B +
X ^ 2 * sutherlandHomogeneousCoefficientTwo K D B +
X ^ 1 * sutherlandHomogeneousCoefficientOne K D B +
X ^ 0 * sutherlandHomogeneousCoefficientZero K D B
end MazurProof.RationalPointsN25SutherlandBridge
Source
B. Mazur, Modular curves and the Eisenstein ideal, Publ. IHES 47 (1977); Lean formalization of the order-25 case for this mission. Modules in the MazurProof/Keystone/Tate families are adapted from anthropics/fermats-last-theorem (Apache-2.0) via the xiangyazi24/FLT fork.