What does Diamond Magic Sizes and the Lab-Grown Slide conclude, in numbers, about “The Cutter's Arithmetic: Bunching, Holes and Left-Digit Attention”?
Answer: Chapter 3 uses a model of the cutter to show how a step leaves a hole and a pile in the weight distribution, with 0.650 of the 1.00 ct class pushed up in the illustrative model, and compares left-digit attention with focal-size conformity.
Scope of the numbers: Chapter 3 uses a model of the cutter to show how a step leaves a hole and a pile in the weight distribution, with 0.650 of the 1.00 ct class pushed up in the illustrative model, and compares left-digit attention with focal-size conformity. Marked boxes separate published evidence, the author's hypotheses and classical sources, and the book contains 20 programs.
Data table
The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 0.650, 1.00
| lowest rough weight pushed to 1.00 (ample material), by C: | |
| C | 2.0: 0.9434 |
| C | 3.0: 0.9525 |
| C | 5.0: 0.9803 |
| C | 8.0: 0.9896 |
| bin no-step with-step pushed | |
| 0.93 | 562 562 0 |
| 0.94 | 573 573 0 |
| 0.95 | 554 554 0 |
| 0.96 | 557 544 0 |
| 0.97 | 555 469 0 |
| 0.98 | 537 283 0 |
| 0.99 | 518 58 0 |
| 1.00 | 437 1250 813 |
| 1.01 | 501 501 0 |
| 1.02 | 428 428 0 |
| 1.03 | 464 464 0 |
| share of the 1.00 bin that was pushed | 0.650 |
| mean cut-value discount of a pushed stone | 0.0418 |
| expected discount of a random 1.00 stone | 0.0272 |
Show the full program
# Chapter 3. p31: The cutter's choice: keep weight or keep the cut?
# Each rough crystal yields weight w0 when cut to ideal proportions
# (ln w0 normal, median 0.95 ct, sd 0.15; illustrative). The cutter may
# keep extra weight w = w0 * (1 + r), 0 <= r <= R, at a cut-grade
# discount of C * r; C varies by crystal, uniform on 2 to 8 (C = 3:
# each 1 % of kept weight costs 3 % of value; illustrative). Value =
# w ** 1.94 * (1 + J if w >= 1.00) * (1 - C * r), J = 0.07 (inside the
# 5 to 10 % of Scott and Yelowitz 2010). Since C > 1.94 nobody keeps
# weight without the step. R, the spare material a crystal allows,
# is uniform on 0 to 0.04 (illustrative). Weights truncated to 0.01 ct.
import math
import random
from collections import Counter
E, J, N = 1.94, 0.07, 20000
def value(w, r, c):
v = w ** E * (1.0 - c * r)
return v * (1.0 + J) if w >= 1.0 - 1e-12 else v
def choose(w0, c, rmax=0.04):
best = (value(w0, 0.0, c), w0, 0.0)
if w0 < 1.0:
r = 1.0 / w0 - 1.0
if r <= rmax:
cand = (value(1.0, r, c), 1.0, r)
if cand[0] > best[0]:
best = cand
return best
rng = random.Random(20261112)
rough = [(math.exp(rng.gauss(math.log(0.95), 0.15)),
rng.uniform(2.0, 8.0), rng.uniform(0.0, 0.04)) for _ in range(N)]
hist0, hist1, pen = Counter(), Counter(), Counter()
cuts = []
for w0, c, rm in rough:
_, w, r = choose(w0, c, rm)
hist0[int(math.floor(w0 * 100.0 + 1e-9))] += 1
k1 = int(math.floor(w * 100.0 + 1e-9))
hist1[k1] += 1
if r > 0:
pen[k1] += 1
cuts.append(c * r)
print("lowest rough weight pushed to 1.00 (ample material), by C:")
for c in (2.0, 3.0, 5.0, 8.0):
lo = 1.0
while choose(lo - 0.0001, c, 0.06)[1] == 1.0:
lo -= 0.0001
print(" C = {:.1f}: {:.4f}".format(c, lo))
print("bin no-step with-step pushed")
for k in range(93, 104):
print("{:.2f} {:6d} {:8d} {:6d}".format(
k / 100.0, hist0[k], hist1[k], pen[k]))
share = pen[100] / hist1[100]
print("share of the 1.00 bin that was pushed: {:.3f}".format(share))
mean_cut = math.fsum(cuts) / len(cuts)
print("mean cut-value discount of a pushed stone: {:.4f}".format(mean_cut))
print("expected discount of a random 1.00 stone: {:.4f}".format(
share * mean_cut))Single source and citation
Fan, Y. (2026). Diamond Magic-Size Price Jumps, the Lab-Grown Learning-Curve Slide and Supply-Cartel Signalling (Version 1.0.0), Chapter 3: The Cutter's Arithmetic: Bunching, Holes and Left-Digit Attention. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-209 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026209,
author = {Fan, Yafu},
title = {Diamond Magic-Size Price Jumps, the Lab-Grown Learning-Curve Slide and Supply-Cartel Signalling},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Magic Sizes: an Invisible Hundredth of a Carat and a Price Step
- Chapter 2 · Piecewise Power Laws and Breakpoint Scans: Finding the Steps in a Price List
- Chapter 4 · The Lab-Grown Learning Curve: How Fast Prices Fall and Where the Floor Is
- Chapter 5 · The Value Ledger: Consumer Good or Asset?
- Chapter 6 · Cartels, Cournot and Noisy Price Signals
- Chapter 7 · Sources East and West: Cao Zhao's Gegu yaolun and Tavernier's Six Voyages
- Chapter 8 · Closing the Loop: a Jewellery Quoting Tool, Twin-Track Copy and a Pre-Registered Test