What does Collapsing Idea-to-Delivery Latency conclude, in numbers, about “Loop Bandwidth: With the Delay Unchanged, More Gain Only Buys Oscillation”?
Answer: Chapter 1 shows that delay sets the ceiling on force: multiplying the gain by 2.5 with the delay fixed stretches settling from 6.69 to 40.63 days.
Scope of the numbers: Parameters are illustrative and all calculations are generated by the book's programs; it offers general, not individual, business guidance, promises no launch date or income, and every production deployment is carried out only after a person confirms it. Chapter 2 splits latency into four drags, decision, start-up, feedback and hand-off, with an illustrative median of 99.9 hours and P90 of 199.8 hours, the tail driven mainly by feedback and decision.
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: 2.5, 40.63, 6.69
| case | Kc | Ti | overshoot | settle | IAE |
|---|---|---|---|---|---|
| delay 2.0, SIMC | 1.00 | 4.0 | 0.043 | 6.69 | 4.35 |
| delay 2.0, gain x1.5 | 1.50 | 4.0 | 0.259 | 13.48 | 4.50 |
| delay 2.0, gain x2.0 | 2.00 | 4.0 | 0.509 | 20.90 | 5.92 |
| delay 2.0, gain x2.5 | 2.50 | 4.0 | 0.761 | 40.63 | 10.19 |
| delay 1.0, SIMC | 2.00 | 4.0 | 0.045 | 3.33 | 2.19 |
| delay 0.5, SIMC | 4.00 | 4.0 | 0.051 | 2.42 | 1.11 |
Show the full program
# Chapter 1. p13: cut the delay or push the gain? PI control, same plant.
# Plant as in p11 (tau = 4, K = 1). PI tuned by the SIMC rule of
# Skogestad (2003): Kc = tau / (K (tc + theta)), Ti = min(tau,
# 4 (tc + theta)), with closed-loop time constant tc = theta.
# Scenarios: base delay 2 with SIMC; the same delay with the gain
# multiplied by 1.5, 2 and 2.5 ("push harder"); the delay halved to 1
# and quartered to 0.5, each re-tuned by SIMC. Settling is the last time
# the output is outside +/- 5 % of the set point; IAE is the integral
# of the absolute error, by math.fsum over Euler steps of dt = 0.01.
import math
TAU = 4.0
K = 1.0
DT = 0.01
T_END = 120.0
def simc(theta):
tc = theta
return TAU / (K * (tc + theta)), min(TAU, 4.0 * (tc + theta))
def run(kc, ti, theta):
n_delay = int(round(theta / DT))
buf = [0.0] * (n_delay + 1)
y, integ = 0.0, 0.0
errs, peak, last_out = [], 0.0, 0.0
for i in range(int(T_END / DT)):
e = 1.0 - y
integ += e * DT
u = kc * (e + integ / ti)
buf.append(u)
y += DT * (-y + K * buf.pop(0)) / TAU
if not math.isfinite(y) or abs(y) > 1e6:
return None
errs.append(abs(e) * DT)
peak = max(peak, y)
if abs(1.0 - y) > 0.05:
last_out = (i + 1) * DT
return max(0.0, peak - 1.0), last_out, math.fsum(errs)
if __name__ == "__main__":
print(" case Kc Ti overshoot settle IAE")
kc2, ti2 = simc(2.0)
rows = [("delay 2.0, SIMC", kc2, ti2, 2.0)]
for m in (1.5, 2.0, 2.5):
rows.append(("delay 2.0, gain x{:.1f}".format(m), kc2 * m, ti2,
2.0))
for th in (1.0, 0.5):
kc, ti = simc(th)
rows.append(("delay {:.1f}, SIMC".format(th), kc, ti, th))
for name, kc, ti, th in rows:
r = run(kc, ti, th)
if r is None:
print(" {:<22} {:>5.2f} {:>5.1f} diverges".format(
name, kc, ti))
continue
over, settle, iae = r
flag = " rings" if settle > T_END - 1.0 else ""
print(" {:<22} {:>5.2f} {:>5.1f} {:>7.3f} {:>6.2f} {:>5.2f}{}"
.format(name, kc, ti, over, settle, iae, flag))Single source and citation
Fan, Y. (2026). Collapsing Idea-to-Reality Latency: Feedback-Bandwidth Limits, Zero-Friction Toolchains and High-Velocity Execution Physics (Version 1.0.0), Chapter 1: Loop Bandwidth: With the Delay Unchanged, More Gain Only Buys Oscillation. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-420 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026420,
author = {Fan, Yafu},
title = {Collapsing Idea-to-Reality Latency: Feedback-Bandwidth Limits, Zero-Friction Toolchains and High-Velocity Execution Physics},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 2 · Decomposing End-to-End Latency: Four Drags and the Slowest Tenth
- Chapter 3 · Queues: Why Full Utilisation Is Not Efficiency
- Chapter 4 · Batches and Rehearsal: the U-Shaped Cost and the Smith Predictor
- Chapter 5 · Hand-offs, Switching and Hesitation: Three Losses Along the Chain
- Chapter 6 · Toolchain Friction and Gates: Why Speed Never Skips a Gate
- Chapter 7 · Classical Sources East and West: Clumsy Speed, the Ant Hole and Maxwell on Governors
- Chapter 8 · Closing the Loop: the Latency Ledger, Field Tools and Checker Specifications