🔢 mathematics

Variants of the Damascus inequality

This paper generalizes the 2016 Damascus inequality by characterizing all positive integers $m$ and $n$ for which the inequality $\sum_{j=1}^m\frac{x_j^n-1}{x_{j}^{n+1}+1}\leqslant 0$ holds under the constraint $\prod_{j=1}^mx_j=1$ , utilizing GA-convexity and Sturm's sequence to analyze both valid cases and the topological properties of non-solutions.

Original authors: Chanatip Sujsuntinukul, Christophe Chesneau

Published 2026-03-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Chanatip Sujsuntinukul, Christophe Chesneau

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are a chef trying to bake a perfect cake. You have a strict rule: the total amount of flour, sugar, and eggs you use must multiply to exactly 1. This is your "kitchen constraint."

Now, you have a special recipe (a mathematical formula) that tells you how "delicious" your cake is. If the result is zero or negative, the cake is a success. If the result is positive, the cake is a disaster (a "violation").

For a long time, mathematicians knew a specific recipe worked perfectly for cakes with 3 ingredients (variables) when the "power" of the recipe was 1 or 2. But they didn't know what happened if you changed the power (let's call it $n$ ) to 3, 4, 5, or higher. They also didn't know if this rule held true for cakes with 4, 5, or more ingredients.

This paper, written by Chanatip Sujsuntinukul and Christophe Chesneau, is like a master chef's guide that finally answers these questions. They map out exactly which recipes work and which ones fail.

Here is the breakdown of their findings, explained simply:

1. The "Damascus Inequality" (The Original Recipe)

In 2016, a mathematician named Dannan discovered a surprising rule for 3-ingredient cakes. No matter how you mix the ingredients (as long as they multiply to 1), the "disaster score" is always negative (or zero). It's like a magic trick that always works.

But then, a question arose: Does this magic trick still work if we change the "power" of the ingredients?

Power 1 & 2: Yes, it works.
Power 3: Surprisingly, yes! The authors proved this works too.
Power 4 and up: No. The magic breaks. If you use a high power, you can find a specific mix of ingredients that makes the cake explode (the score becomes positive).

2. The "Number of Ingredients" Rule ( $m$ )

The authors also asked: What if we have more than 3 ingredients?

1 or 2 Ingredients: The rule works for any power. It's foolproof.
3 Ingredients: Works for powers 1, 2, and 3. Fails for 4 and up.
4 Ingredients: Works only for power 1. If you try power 2 or higher, it fails.
5 Ingredients: Works only for power 1.
6 or more Ingredients: Fails even for power 1!

The Takeaway: The more ingredients you add to the mix, the stricter the rules become. The "magic" of the inequality disappears quickly as you add more variables or increase the power.

3. How They Solved It (The Tools)

To prove these things, the authors didn't just guess; they used two powerful mathematical "kitchen tools":

GA-Convexity (The "Shape" of the Curve): Imagine the recipe as a landscape. Sometimes the landscape is a smooth bowl (convex), meaning the middle is the lowest point. Sometimes it's a hill. The authors checked the "shape" of the recipe function to see if the ingredients naturally settle into a safe zone or if they can roll off a cliff into a disaster zone.
Sturm's Sequence (The "Root Finder"): This is like a super-precise metal detector. They used it to scan the landscape and find exactly where the "danger zones" (where the score becomes positive) begin and end. It helped them prove that for certain powers, the danger zones are real and reachable.

4. The "Forbidden Zone" (What Happens When It Fails)

When the inequality fails (for example, with 3 ingredients and power 4), where do the bad cakes hide?

They aren't near the center: The authors found that the "bad" mixes are not right next to the perfect mix (where all ingredients are 1). They stay a safe distance away.
They aren't infinite: The bad mixes don't require you to use an infinite amount of one ingredient. They are trapped in a specific, bounded box.
Visualizing it: Imagine a 3D map. The "safe" area is a green valley. The "bad" area is a red mountain range. The authors showed that the red mountains are floating islands that never touch the center of the map and never stretch to infinity.

5. The Big Mystery (The Conjecture)

The authors noticed something interesting: If a specific mix of ingredients fails for Power 4, it seems to also fail for Power 5, 6, and so on.

The Guess: Once a recipe is broken for a certain power, it stays broken for all higher powers.
The Status: They haven't proven this 100% yet, but the evidence is strong. They leave this as a puzzle for future mathematicians to solve.

Summary

Think of this paper as a safety manual for a complex mathematical machine.

Safe Zone: If you have 1 or 2 variables, you can crank the power dial as high as you want; the machine is safe.
Warning Zone: If you have 3 variables, you can go up to Power 3, but stop there.
Danger Zone: If you have 4+ variables, even Power 2 is too dangerous.

The authors didn't just say "it breaks"; they drew a detailed map of exactly where and why it breaks, using advanced geometry and algebra to ensure no corner of the mathematical universe was left unchecked.

1. Problem Statement

The paper investigates the generalized Damascus inequality, originally established in 2016 by Dannan and Sitnik. The classical inequality states that for positive real numbers $x, y, z$ with $xyz=1$:
$\frac{x-1}{x^2-x+1} + \frac{y-1}{y^2-y+1} + \frac{z-1}{z^2-z+1} \leq 0$
The authors extend this to a general setting involving $m$ variables and an exponent $n$ . They seek to characterize all pairs of positive integers $(m, n)$ such that the following inequality holds for any $x_1, \dots, x_m > 0$ satisfying $\prod_{j=1}^m x_j = 1$ :
$S_m^n(x_1, \dots, x_m) := \sum_{j=1}^m \frac{x_j^n - 1}{x_j^{n+1} + 1} \leq 0$
Additionally, for cases where the inequality fails, the authors analyze the topological properties of the set of non-solutions (where $S_m^n > 0$ ).

2. Methodology

The authors employ a combination of analytical and computational techniques:

Calculus and Derivative Tests: Used for small cases ( $m=1, 2$ ) to establish monotonicity and local extrema.
GA-Convexity (Geometric-Arithmetic Convexity): A central tool used to handle the multiplicative constraint. A function $f$ is GA-convex if $f(\alpha^\lambda \beta^{1-\lambda}) \leq \lambda f(\alpha) + (1-\lambda)f(\beta)$ . The authors utilize the criterion that $f$ is GA-convex iff $f(e^x)$ is convex, or via the condition $\Delta f(x) = f'(x) + xf''(x) \geq 0$ .
Sturm's Sequence: A computational algebraic method used to determine the number of real roots of polynomials within specific intervals. This is crucial for analyzing the critical points of the auxiliary functions derived from the inequality and proving sign changes in derivatives.
Case Analysis and Bounding: For cases where variables fall outside specific intervals (e.g., near 0 or $\infty$ ), the authors construct upper bounds and use continuity arguments to prove the inequality holds or fails.
Topological Analysis: For the failure cases ( $m=3, n \geq 4$ ), the authors use compactness arguments and the properties of the constraint hypersurface to prove boundedness and separation from specific planes.

3. Key Contributions and Results

The paper provides a complete characterization of the valid pairs $(m, n)$ , summarized in Table 1 of the paper:

$m$ (Variables)	Valid $n$ (Exponents)	Status
1	All positive integers	Holds
2	All positive integers	Holds
3	$1, 2, 3$	Holds for $n \leq 3$ ; Fails for $n \geq 4$
4	$1$	Holds for $n=1$ ; Fails for $n \geq 2$
5	$1$	Holds for $n=1$ ; Fails for $n \geq 2$
$\geq 6$	None	Fails for all $n \geq 1$

Specific Theoretical Results:

Theorem 2.1: Proves the inequality holds for $m=2$ for all $n$ .
Theorem 3.1: Demonstrates that for $m=3$ , the inequality fails for all $n \geq 4$ by providing explicit counterexamples (e.g., $(1/2, 3/2, 4/3)$ for $n \geq 5$ ).
Theorem 3.2: Rigorously proves the inequality holds for $m=3, n=3$ . This is the most technically demanding proof, utilizing Sturm's sequences to analyze the critical points of the function $f(x) = \frac{x^3-1}{x^4+1} - \frac{3}{2}\ln x$ .
Theorem 4.1: Establishes that for $m \geq 4$ , the inequality fails for $n \geq 2$ , and for $m \geq 6$ , it fails even for $n=1$ . The proofs rely on specific counterexamples where variables are set to $2$ and $1/2^{m-1}$ .
Theorem 4.3: Proves the inequality holds for the specific case $(m, n) = (5, 1)$ (and by extension $m=4, n=1$ ) using a similar GA-convexity and Sturm's sequence approach as the $m=3, n=3$ case.

4. Analysis of Non-Solutions (Violation Sets)

For the cases where the inequality fails (specifically $m=3, n \geq 4$ ), the authors define the violation set $I_3^n = \{ (x,y,z) \in H_3 : S_3^n > 0 \}$ . They prove two significant topological properties:

Separation from Unity (Theorem 5.1): The set $I_3^n$ is disjoint from a neighborhood of the planes $x=1$ , $y=1$ , and $z=1$ . This means the inequality holds strictly in the vicinity of the point $(1,1,1)$ .
Boundedness (Theorem 5.2): The set $I_3^n$ is bounded. The non-solutions cannot escape to infinity; they are confined within a finite box $[c, 1/c]^3$ .

Conjecture 5.5: The authors propose that the violation sets are nested: $I_3^n \subseteq I_3^{n+1}$ . This suggests that if a triple violates the inequality for a specific $n$ , it will also violate it for all larger exponents.

5. Significance

Resolution of Open Problems: The paper fully resolves the generalization question posed by Kalinin and Pankratova regarding the exponent $n$ for $m=3$ , and extends the analysis to arbitrary $m$ .
Methodological Advancement: The successful application of Sturm's sequences combined with GA-convexity provides a robust framework for tackling symmetric inequalities with multiplicative constraints where standard convexity (in the arithmetic sense) fails.
Structural Insight: The topological characterization of the violation sets reveals that the failure of the inequality is a "local" phenomenon away from the symmetric point $(1, \dots, 1)$ and is confined to a bounded region, offering deep geometric intuition into the behavior of these symmetric polynomials.
Future Directions: The paper identifies open questions regarding the behavior of violation sets for $m \geq 4$ (specifically boundedness and nesting), setting the stage for further research in inequality theory and optimization.