Abstract
This body of work has written its formulas with G = 4π and ħ = 1/(4π) (the Yamagishi patch; two of the three placings of Paper 39). The scope notes of Paper 22, Paper 26, Paper 28 and others say that G = 4π can be placed in the same sense as G = 1 (geometrized units) or SI; Paper 116 showed that under G = a and ħ = 1/a a product of ħ, G and c takes the form a to the power −μ (μ the mass dimension) times a power of c; and Paper 1130 showed that the sum ε + μ of the power ε of 4π standing in a value and the mass dimension μ does not change when G, ħ or c is multiplied in (the Yamagishi 4π-Mass Invariant). This paper asks which change of units links the two placings, and shows that the patch is obtained from the units with G = ħ = 1 (with the same number for c) by multiplying only the unit of mass by 4π, leaving the units of length and time as they are. A change of units that keeps the number of c and takes the numbers of G and ħ from 1 and 1 to 4π and 1/(4π) has only this one form. With c = 1 the reference is the true Planck units, and the unit of mass of the patch is 4π times the Planck mass (2.734988(30)×10⁻⁷ kg). What is shown is a theorem about the choice of units; no new law of physics is claimed and the measured value of G is not produced.
As corollaries, the number of a quantity of mass dimension μ in the patch is its number in Planck units times 4π to the power −μ. Since the masses in the formulas of this body of work are written as numbers in the patch, in Planck units a mass placed at 4π becomes 16π² and a mass M not placed becomes 4πM (M is counted as the letter for its number in the patch). From this, the weight ε + μ of Paper 1130 equals the power of π of the same quantity written as a number in Planck units. G, ħ and c do not move the weight because their numbers in Planck units carry no π (1, 1 and c). The ledger formula ε + μ = 2r + u + ρ reads as a count of the π in the number in Planck units, and splits into one 4π of the rescaling for each mass (r + u), one more 4π of its own number for each placed mass (r), and the 4π written directly in the formula (ρ). A constant not placed is a constant that did not receive the factor of the rescaling. Paper 1128 wrote that the four types are divided by 4π once at every step because the exponent of mass of every type is 1 (the ε = μ of Paper 116). That ε = μ comes from only the unit of mass having been moved. The ratio M(R)/kR of the counting scene of Paper 1129 stays 4π (∫dΩ), because the rescaling puts the same 4π on the numerator and the denominator, while the m = 4π of the carrying scene moves from 16π² in Planck units to 4π in the patch. The rescaling reproduces the distinction of Paper 1129 as it is. This paper does not rewrite Paper 1130; it gives one more reason, from the change of units, why its theorem holds.
Compared with other conventions, the only placings reached by the unit of mass alone, keeping the number of c, are the family in which the product of G and ħ is 1 (G = a and ħ = 1/a of Paper 116), and this family is exactly the change of units that multiplies the unit of mass by a. Rationalization (4πG = 1) and 8πG = 1 with ħ = 1 kept cannot be reached unless the units of length and time are also multiplied by √(4π) and √(8π) respectively. For a change of units that keeps c, a ledger of the form ε + sμ stands only when the factor for length and time carries no π, and under rationalization a ledger of that form does not stand (another form, with the dimensions of length and time, does). The coefficient of Gauss’s law is heavier by 1 in weight than the coefficient of the inverse-square law in every convention (five for gravity, seven for electromagnetism). This follows at once from the 4π in the ratio of the two coefficients being a number without dimension; a convention only decides in which law that 4π is seen.
A census was taken. The 501 places (85 papers) of conventions of gravity in this body of work are 237 for the Yamagishi patch, 187 for G kept as a letter, 57 for G = 1, 18 for 8πG = 1 (Paper 1092 and Paper 1093) and 0 for rationalization, so the gravity of this body of work has never been written rationalized. Reading the 257 places whose main label in the scan of powers of 2 was a power of 4 from the patch, 141 places (17 papers) are products of placings, and in every one the exponent of 4 equals the size of the power of the placing (the exponent of 2 is twice it), with 4 and π standing in pairs. Products of one placing and one 4π written in the formula are 20, places discussing the groupings in the Hawking formulas and others in Paper 43, Paper 44 and Paper 50 are 58, places that only name are 5, and powers of 4π that are not products of placings are 33. The Hawking temperature and ħ²/(2m), which carry the same 128π³ under the patch, both have the form of division by 32π² as numbers in Planck units, but the breakdown of 32π² differs: the 8π of the formula and the 4π of the mass not placed, against the 16π² of the placed mass and the 2 of motion. The statement of Paper 39 that spending the remaining degree of freedom on c = 1 makes the Planck length 1 and returns to the definition of Planck units holds for the units of length and time, and is worth reading with the condition that the unit of mass is 4π times the Planck mass. The question in Paper 74 of producing G = 4π by computation as an elastic constant, and the 32π² of Paper 63, can be read in the same way. The separator is whether only the unit of mass was moved, or the units of length and time as well.
Keywords: Yamagishi patch, Planck units, unit of mass, change of units, power of 4π, mass dimension, conventions of gravity, separator.
Scope (what is not claimed)
This paper claims no new theorem or law of physics and rewrites no sentence of any earlier paper. What is shown is a theorem about the choice of units, with its corollaries, obtained by symbolic computation from the rule of dimensional analysis for how numbers move when units are changed (Bridgman 1922), and a census applying it to the materials of the scans. Planck units, rationalized units and units with 8πG = 1 are all known. G = 4π and ħ = 1/(4π) are a way of placing numbers (what Paper 44 calls a convention), not a claim about measured values. This paper does not produce the measured value of G from geometry, and does not claim to compute G = 4π from the structure of the body of work. The theorem and the corollaries are elementary, so it would not be surprising if the same thing is written in textbooks or papers outside this body of work. For the 2nd edition, the public indexes of papers were searched for nearby work, and the differences from the closest one, Wutke (2023), are written in Section 10. Every number was produced by this paper’s check (symbols with sympy), and the value of the Planck mass is the CODATA 2022 value (NIST 2024), the same as in Paper 1128. The two unit systems compared are given the same number for c. With c = 1 the reference is the true Planck units; with c kept as a letter the reference is the units with G = ħ = 1 and the same number for c. "Numbers in Planck units" in this paper means numbers in this reference. The masses in the formulas of the body of work are taken to be written as numbers in the patch, and when moving to numbers in Planck units the number of each mass is multiplied by 4π (a placed m = 4π becomes 16π², and a mass M not placed becomes 4πM). A mass not placed is written as 4πM with the letter M for its number in the patch when π is counted; if a mass not placed is written with the letter M′ = 4πM for its number in Planck units, the power of π is that value minus u. Temperature is counted as energy with the Boltzmann constant set to 1, following Paper 1115 and Paper 1130. Formulas with electric charge (ε₀) are treated, as in Paper 1130, with ε₀ as a constant not placed, and changes of electromagnetic units are not treated. Whether the 4π of the patch is a solid angle or a convention is not decided (the definition of Paper 56 calls it the solid angle of the unit sphere, and Paper 35 and Paper 44 call it a convention). What this paper says goes only as far as that 4π working as the factor of the unit of mass. The materials of the census are the lists of two scans of 20 November 2026 (501 places of conventions of gravity and 3,801 places of powers of 2), used with their fingerprints fixed; the labels of places are as read when the scans were made. The classes this paper gave to the 257 places with the patch label in the scan of powers of 2 are this paper’s reading (the original sentences were read for each kind of quantity, and lines were decided one by one where a kind spans more than one reading; one reader), and the 58 places of Paper 43, Paper 44 and Paper 50 are not divided line by line. Paper 116 showed that under G = a and ħ = 1/a a product of ħ, G and c becomes a to the power −μ; Paper 44 wrote that the identity of ε is the mass dimension; and Paper 40 wrote about the 4π of the Planck mass. Paper 43 wrote the powers of a with the patch constant kept as a letter, and read those powers as the mass dimension. Paper 39 wrote that G and ħ use two of the budget of units and that with c = 1 the Planck length is 1. Paper 1083 showed in numbers that placing m = 4π is a change of the unit of mass. Paper 1130 wrote ε + μ = 2r + u + ρ as a theorem, and wrote that the only placing with weight 0 is the rule of the patch and that G and ħ are not seen in ε + μ because they were placed by the rule of the patch. Paper 1129 distinguished the counting scene and the carrying scene, and Paper 1128 wrote that the four types are divided by 4π once, with the reason (the exponent of mass is 1). Paper 1112 counted gravity written with G as units that are not rationalized, which put no 4π in the formula of a point. What is added is: naming the patch as a change of units from Planck units and showing that it is the only one; showing that the family in which the product of G and ħ is 1 is the family of changes of the unit of mass only (the 4π to the power −μ of the rule of the patch named by Paper 1130 is exactly the factor of that change); showing that the weight of Paper 1130 is the power of π of the number in Planck units, and how the ledger formula splits; tracing the reason of Paper 1128 and the powers of a of Paper 43 back to the change of units, and showing that for quantities with a mass not placed the power of a is the mass dimension, with sign reversed, of the part made of ħ and G; showing that the distinction of Paper 1129 is reproduced by the change of units; showing the factors for other conventions and the condition for a ledger of the form ε + sμ to stand; taking the census; and setting the separator.
1. Introduction: what did the patch move?
This body of work has written its formulas with $G=4\pi$ and $\hbar=1/(4\pi)$ (the Yamagishi patch). That the three placings together with $m=4\pi$ use up the budget of mechanical units was written by Paper 39. The scope notes of Paper 22, Paper 26, Paper 28, Paper 29, Paper 30, Paper 32, Paper 35, Paper 38 and Paper 39 say in the same words that $G=4\pi$ can be placed in the same sense as $G=1$ (geometrized units) or $G=6.674\times10^{-11}$ (SI). Both placings are choices of units. Then, which units must be changed, and by how much, to go from the units with $G=1$ to the units of the patch?
Paper 116 relaxed the patch to the one-parameter $G=a$, $\hbar=1/a$, wrote $X=\hbar^pG^qc^N=a^{-\mu}c^N$ ($\mu$ the mass dimension), and wrote that $\hbar G=1$ holds identically on this family. Paper 44 wrote that the identity of $\epsilon$ is the mass dimension. Paper 1130 made into a theorem, as the ledger formula $\epsilon+\mu=2r+u+\rho$, that the sum of the power $\epsilon$ of $4\pi$ standing in a value and the mass dimension $\mu$ does not change when $G$, $\hbar$ or $c$ is multiplied in (the Yamagishi 4π-Mass Invariant).
The scan of conventions of gravity of 20 November 2026 found that for all 20 quantities of gravity, multiplying the number in the patch by $(4\pi)^\mu$ gives the number with $G=\hbar=1$. This paper writes this as a theorem. The patch is obtained from the units with $G=\hbar=1$ by multiplying only the unit of mass by $4\pi$, without moving the units of length and time. From this the reason for the weight and its invariance in Paper 1130 follows in one line, and the numbers of Paper 1128 and Paper 1129 can be read in the same light. Within the body of work, no paragraph writing the patch as a change of the unit of mass was found in what this paper’s check searched (the unit of mass, Planck and the patch appear in the same paragraph only once each in Paper 39 and Paper 1115, and neither speaks of a change of units).
2. How numbers move when units are changed
Multiply the unit of mass by $x$, the unit of length by $y$ and the unit of time by $z$ (new unit = factor times old unit). A quantity whose dimension is mass to the power $\mu$, length to the power $\lambda$ and time to the power $\tau$ is the same thing, and only its number is multiplied by
$$x^{-\mu}\,y^{-\lambda}\,z^{-\tau}$$(the dimensional bookkeeping of Bridgman 1922). $G$ is mass to the power $-1$, length cubed and time to the power $-2$, so its number is multiplied by $xz^2/y^3$. $\hbar$ is multiplied by $z/(xy^2)$, $c$ by $z/y$, and a mass by $1/x$. If two unit systems are given the same number for $c$, then $z=y$, and the units of length and time move by the same amount.
3. Theorem: the patch multiplies only the unit of mass by 4π
Theorem
A change of units that keeps the number of $c$ and takes the numbers of $G$ and $\hbar$ from $(1,1)$ to $(4\pi,1/(4\pi))$ is one and only one: it multiplies the unit of mass by $4\pi$ and does not move the units of length and time.
Proof
Since the number of $c$ is kept, $z=y$, and the number of $G$ is multiplied by $x/y$ and that of $\hbar$ by $1/(xy)$. Multiplying $x/y=4\pi$ and $1/(xy)=1/(4\pi)$ gives $1/y^2=1$, and dividing gives $x^2=16\pi^2$; since the factors are positive, $x=4\pi$ and $y=z=1$ only.
By the same computation, a change of units taking $(G,\hbar)$ from $(1,1)$ to a general $(a,b)$ multiplies the unit of mass by $\sqrt{a/b}$ and the units of length and time by $1/\sqrt{ab}$. The units of length and time do not move only when $ab=1$, and this is exactly the family $G=a$, $\hbar=1/a$ of Paper 116. The family on which Paper 116 found $\hbar G=1$ as an identity is the family of changes of units that multiply the unit of mass by $a$. $\hbar G$ is a quantity of mass dimension 0 (length to the fifth and time to the power $-3$), so its number does not move when the unit of mass is changed.
The reference is fixed for each number of $c$. Seen from the true Planck units ($G=\hbar=c=1$), the patch with the number $C$ for $c$ is the units with the unit of mass multiplied by $4\pi/\sqrt{C}$, length by $C^{3/2}$ and time by $C^{5/2}$. The reference with the number $C$ for $c$ ($G=\hbar=1$) has mass multiplied by $1/\sqrt{C}$ and length and time by the same factors as the patch. The difference between the two is, for every $C$, only the factor $4\pi$ in the unit of mass. This body of work often keeps $c$ as a letter ($l_p=c^{-3/2}$ of Paper 39). "Numbers in Planck units" in this paper means numbers with $G=\hbar=1$ and the same number for $c$, and they become numbers in the true Planck units when $c=1$.
In the patch with $c=1$, the numbers of the Planck length $\sqrt{\hbar G/c^3}$ and the Planck time $\sqrt{\hbar G/c^5}$ are both 1, and the number of the Planck mass $\sqrt{\hbar c/G}$ is $1/(4\pi)$. The units of length and time are the Planck units themselves, and the unit of mass is $4\pi$ times the Planck mass. Using $m_P=2.176434(24)\times10^{-8}$ kg, which Paper 1128 took from CODATA 2022, this unit of mass is $4\pi m_P=2.734988(30)\times10^{-7}$ kg, with the same relative uncertainty $1.1\times10^{-5}$ as $m_P$. When $m=4\pi$ is also placed, as in the formulas of this body of work, the third item of the budget is used, so $c=1$ cannot be placed (Paper 39) and the number $C$ of $c$ is fixed by the mass of the placed particle. The unit of mass of the patch is then $(4\pi/\sqrt{C}) m_P$, $1/(4\pi)$ of the mass of the placed particle (Paper 1083).
4. Corollaries: numbers are multiplied by 4π to the power −μ, and the weight is the power of π in Planck units
The masses entering the formulas of this body of work are written as numbers in the units of the patch. The number of a mass $m$ placed at $4\pi$ is $4\pi$, and a mass $M$ not placed is a number in the patch kept as a letter. Moving to numbers in Planck units multiplies the number of each mass by $4\pi$: a placed mass becomes $16\pi^2$ and a mass not placed becomes $4\pi M$. When this paper counts the $\pi$ in numbers in Planck units, a mass not placed is written as $4\pi M$ with the letter $M$ for its number in the patch. If it is written with the letter $M'=4\pi M$ for its number in Planck units, the power of $\pi$ is smaller by $u$ (the Hawking temperature is $c^3/(32\pi^2M)$ with $-2$, and $c^3/(8\pi M')$ with $-1$; check 2-9). "The power of $\pi$ in Planck units" below is the value under this convention.
Corollary 1 (how numbers change)
For a quantity $X$ of mass dimension $\mu$, the number in the patch $X_{\rm patch}$ and the number in Planck units $X_{\rm Pl}$ are related by
$$X_{\rm Pl}=(4\pi)^{\mu}\,X_{\rm patch}.$$Put $x=4\pi$ and $y=z=1$ in the rule of the previous section.
Corollary 2 (reading of the weight)
The weight $\epsilon+\mu$ of Paper 1130 ($\epsilon$ is the power of $4\pi$ standing in the number in the patch, counted as a power of $\pi$) equals the power $\epsilon_{\rm Pl}$ of $\pi$ standing in the number $X_{\rm Pl}$ in Planck units. Taking the power of $\pi$ on both sides of Corollary 1 gives $\epsilon_{\rm Pl}=\mu+\epsilon$.
Corollary 3 (why it is invariant)
The numbers of $G$, $\hbar$ and $c$ in Planck units are $1$, $1$ and $c$, and none carries $\pi$. Weights add under products of quantities, so multiplying or dividing by $G$, $\hbar$ or $c$ any number of times does not move $\epsilon+\mu$. Part (ii) of the theorem of Paper 1130 can be said in this one line.
Corollary 4 (how the ledger formula splits)
The $\pi$ standing in $X_{\rm Pl}$ comes only from the $4\pi$ written directly in the formula ($\rho$), the $16\pi^2$ of placed masses (2 each) and the $4\pi M$ of masses not placed (1 each). Therefore
$$\epsilon+\mu=\epsilon_{\rm Pl}=2r+u+\rho=(r+u)+r+\rho ,$$and part (i) of the ledger formula of Paper 1130 follows as it is. In the split on the right, each mass carries one $4\pi$ of the rescaling ($r+u$), each placed mass carries one more $4\pi$ of its own number ($r$), and the $4\pi$ written directly in the formula ($\rho$) stays as it is. Paper 1130 wrote that reading a placed mass as a pair of a direct $4\pi$ and a mass not placed does not change the sum (weight $2=1+1$). The 1 of the direct $4\pi$ corresponds to the $4\pi$ of the placed number, and the 1 of the mass not placed to the $4\pi$ of the rescaling. The parity of part (iii) follows in the same way from $2r$ being even.
Corollary 5 (constants not placed)
In formulas that keep $G=1$, or keep $\hbar$ at the number 1, that constant enters with its number in Planck units and has not received the factor $(4\pi)^{-\mu_Y}$ of the rescaling. Because of what it did not receive, $(4\pi)^{\mu_Y}$ remains in the value, and the weight becomes $\epsilon_{\rm Pl}+U$ ($U$ the sum of the mass dimensions of the constants not placed). This is the extension in Section 5 of Paper 1130, which also counts in the same form formulas that keep $\hbar$ as a letter and $\varepsilon_0$. For $\hbar^2/(2m)=1/(8\pi)$ with $\hbar=1$ kept (Paper 17), $0=-2+2$; for $(3/5)GM^2/R$ with $G=1$ kept ($M=Nm$, Paper 1106), $3=4-1$.
In the check, Corollaries 1, 2 and 4 were confirmed by symbolic substitution on 32 quantities (including the formulas of the ledger table of Paper 1130: formulas within the three placings, Hawking formulas with a mass not placed, and the counting scene $M(R)=4\pi\cdot kR$) and on 300 products of the ledger factors raised to random integer and half-integer powers. That Corollary 1 fails when a mass not placed is not converted to its number in Planck units was also confirmed as a failing example.
5. Links with Paper 1128, Paper 1129 and Paper 1130
Paper 1130
Paper 1130 proved $\epsilon+\mu=2r+u+\rho$ from the way the books are kept, by bookkeeping factor by factor. This paper does not rewrite that formula; it shows that the same formula is the way of counting $\pi$ when the number in the patch is taken back to the number in Planck units. $G$, $\hbar$ and $c$ do not move the weight because none of them carries $\pi$ in Planck units, and a placed mass carries 2 while a mass not placed and a direct $4\pi$ carry 1 because in Planck units they become $16\pi^2$, $4\pi M$ and $4\pi$. A quantity of weight 0 is a quantity whose number in Planck units carries no $\pi$. In Section 5 Paper 1130 wrote that $G$ and $\hbar$ are not seen in $\epsilon+\mu$ because they were placed by the rule of the patch, and $c$ is not seen because its mass dimension is 0. That rule of the patch (placing a constant of mass dimension $\mu_Y$ at $(4\pi)^{-\mu_Y}$) is exactly the factor of the change of units that multiplies the unit of mass by $4\pi$. In this light the theorem of Paper 1130 becomes a consequence of the change of units. The form of the formula does not change, and there is one more reason why it holds.
Paper 1128
Paper 1128 itself gives the reason: every type is divided by 4π once because the exponent of mass is 1 (the ε = μ of Paper 116). The four types $\sqrt{\hbar/G} c^{k+1/2}$ ($k=0,1,2,3$) are $c^{k+1/2}$ in Planck units and carry no $\pi$, and by Corollary 1 their numbers in the patch are $c^{k+1/2}/(4\pi)$. That the ratio $c$ between steps has mass dimension 0 and carries no $4\pi$ is also written by Paper 1128. What this paper adds is the reason one step further back: that $\epsilon=\mu$ comes from only the unit of mass having been moved.
Paper 1129
Paper 1129 wrote that in a ball of density $k/r^2$, while the weight is counted with $kR$ written beside it ($M(R)=4\pi\cdot kR$, $\rho=1$, $u=1$), the $4\pi$ is $\int d\Omega$, and that when $kR$ is folded into the unit of mass and carried ($m=4\pi$, $r=1$), the $4\pi$ becomes a number of mass, and that writing $kR$ back returns it to the counting scene. As numbers in Planck units the two scenes are both $16\pi^2$ (the counting scene $16\pi^2 kR$), and two $4\pi$ stand. One is the $4\pi$ of the rescaling (one per mass), and the other is the $4\pi$ of Paper 1129: $\int d\Omega$ in the counting scene and the placed number in the carrying scene. The ratio $M(R)/kR$ of the counting scene stays $4\pi$ both in the patch and in Planck units, because the rescaling puts the same $4\pi$ on the numerator and the denominator. The number of $m$ in the carrying scene moves from $16\pi^2$ in Planck units to $4\pi$ in the patch. The rescaling thus reproduces the distinction of Paper 1129 as it is: the $4\pi$ of the counting scene does not move with the units, and the $4\pi$ of the carrying scene does.
As for the $4\pi$ of the patch constants themselves (the $4\pi$ standing in $G$ and $\hbar$), Paper 1129 states in its scope note that it does not treat the roots of a 4π that is not that of m = 4π, including the patch constants of G and ħ. In this paper’s light this $4\pi$ is a $4\pi$ of placing that does not enter the $\rho$ of Paper 1130, and differs from the $4\pi$ of the carrying scene of Paper 1129 (which returns to $\int d\Omega$ when $kR$ is written back). It stands in $G$ and $\hbar$ as numbers in the patch, and when written back to Planck units it does not return to $\int d\Omega$ but moves to the side of the numbers of masses as the factor of the unit of mass.
Paper 116, Paper 44, Paper 40 and Paper 43
The $X=a^{-\mu}c^N$ of Paper 116 is, for quantities made only of $\hbar$, $G$ and $c$, exactly how numbers change under the change of units that multiplies the unit of mass by $a$ (confirmed in the check with 60 sets of random powers). The statement of Paper 44 that the identity of $\epsilon$ is the mass dimension can be restated as: because only the unit of mass was moved. The same holds for Paper 40 writing that one $4\pi$ attaches to the Planck mass and the Planck energy. Paper 43 kept the patch constant as a letter and wrote that the Hawking temperature $T_H$ carries $a^{-2}$, the area of the horizon $A$ carries $a^2$, the Stefan-Boltzmann $\sigma$ carries $a^3$ and the power $P$ carries $a^{-3}$, and in edition 3 wrote that the power of $a$ is the mass dimension, not a geometric capacity. In this paper’s light, for $T_H$, $A$ and $P$, which contain the mass $M$ not placed, that power is not the mass dimension $\mu$ of the whole quantity with sign reversed but $-\mu+u$ (with the power $u$ of $M$ added), the mass dimension, with sign reversed, of the part made of $\hbar$ and $G$ without $M$ ($T_H$ has $\mu=1$ and carries $a^{-2}$). The $a^3$ of $\sigma$, which contains no $M$, is the mass dimension $-3$ of the whole quantity with sign reversed. Paper 43 derives each power from the factors of $\hbar$ and $G$. What this paper adds is writing that power as the single expression $-\mu+u$ and making explicit that, for quantities containing $M$, it departs from the $\mu$ of the whole quantity.
6. Comparison with conventions
The factors that move from the units with $G=\hbar=1$ to each placing, keeping the number of $c$, are listed.
| placing | $G$ | $\hbar$ | unit of mass | units of length and time | only the unit of mass? |
|---|---|---|---|---|---|
| Planck units (reference) | $1$ | $1$ | $1$ times | $1$ times | (reference) |
| Yamagishi patch | $4\pi$ | $1/(4\pi)$ | $4\pi$ times | $1$ times | yes |
| placing $G$ only | $4\pi$ | $1$ | $2\sqrt{\pi}$ times | $1/(2\sqrt{\pi})$ times | no |
| rationalized ($\hbar=1$ kept) | $1/(4\pi)$ | $1$ | $1/(2\sqrt{\pi})$ times | $2\sqrt{\pi}$ times | no |
| $8\pi G=1$ ($\hbar=1$ kept) | $1/(8\pi)$ | $1$ | $1/\sqrt{8\pi}$ times | $\sqrt{8\pi}$ times | no |
| rationalized ($\hbar=4\pi$) | $1/(4\pi)$ | $4\pi$ | $1/(4\pi)$ times | $1$ times | yes |
Rationalization and $8\pi G=1$ with $\hbar=1$ kept cannot be reached by changing only the unit of mass; the units of length and time also move, by $2\sqrt{\pi}=\sqrt{4\pi}$ and $\sqrt{8\pi}$ respectively. If $\hbar$ may be placed at $4\pi$, then $4\pi G=1$ is $a=1/(4\pi)$ in the family with $G\hbar=1$ and is reached by multiplying the unit of mass by $1/(4\pi)$ alone. Placing only $G$ at $4\pi$ without moving $\hbar$ also moves the units of length and time. In formulas of gravity alone, without $\hbar$, only the factor $x/y$ of $G$ matters, and the number of $G$ can be set by the unit of mass alone.
In placings where the units of length and time also move, the shift of the power of $\pi$ in a number involves the dimensions of length and time. Under rationalization with $\hbar=1$ kept, the power of $\pi$ in the number of a quantity with dimensions mass $\mu$, length $\lambda$ and time $\tau$ shifts by $(\mu-\lambda-\tau)/2$ (confirmed in the check for $G$, $\hbar$, $c$, $4\pi G$, $8\pi G/c^4$ and the three Planck quantities). For $G$, $\epsilon$ shifts by $-1$ with $\mu=-1$, and $\hbar$ does not shift, with $\mu=1$, so no sum of the form $\epsilon+s\mu$ hides both. On the other hand, within the family $G=(4\pi)^s$, $\hbar=(4\pi)^{-s}$ with $G\hbar=1$, numbers are multiplied by $(4\pi)^{-s\mu}$, and $\epsilon+s\mu$ becomes the power of $\pi$ in Planck units. Counting with $4\pi$, the patch corresponds to $s=1$, and the ledger of the parity law is a ledger that stands within placings that moved only the unit of mass.
The converse also holds. For a change of units that keeps $c$, let $\alpha$ be the power of $\pi$ in the factor for mass and $\beta$ that in the factor for length and time; the shift of the power of $\pi$ in a number is $-\alpha\mu-\beta(\lambda+\tau)$, and a ledger of the form $\epsilon+s\mu$ that hides both $G$ and $\hbar$ stands only when $\beta=0$, in which case $s=\alpha$. Even under rationalization ($\hbar=1$), $\epsilon-(\mu-\lambda-\tau)/2$, which includes the dimensions of length and time, does not move when $G$, $\hbar$ or $c$ is multiplied in. What does not stand is only a ledger of the form $\epsilon+s\mu$.
The coefficients of Gauss’s law and of the inverse-square law are listed with the weight of Paper 1130 (five placings of gravity).
| placing | weight of the inverse-square coefficient $G$ | weight of the Gauss coefficient $4\pi G$ |
|---|---|---|
| Planck units | $-1$ | $0$ |
| Yamagishi patch | $0$ | $1$ |
| placing $G$ only | $0$ | $1$ |
| rationalized ($\hbar=1$ kept) | $-2$ | $-1$ |
| $8\pi G=1$ ($\hbar=1$ kept) | $-2$ | $-1$ |
In every placing, the Gauss coefficient is heavier by 1 in weight than the inverse-square coefficient (this follows at once from the ratio $4\pi$ of the two coefficients having no dimension). The same held for seven cases of electromagnetism (five reading the mass dimension of charge as 0 and two reading it as 1/2). The number of the ratio $4\pi$ does not move under any change of units. This 1 is the solid angle $4\pi$ that surrounds a point with a sphere. A convention only decides whether that $4\pi$ is seen in the inverse-square law (rationalization) or in Gauss’s law (gravity written with $G$, Gaussian units). This restates in terms of the difference of weights the division Paper 1112 made between units that are rationalized and units that are not rationalized (which put no 4π in the formula of a point: electric Gaussian units and gravity written with G).
7. Census
Conventions of gravity in this body of work
The scan of conventions of gravity of 20 November 2026 picked up candidates for conventions of the $G$ of gravity from the 1,130 Japanese editions of the papers, and four readers decided the convention place by place.
| convention | places | papers |
|---|---|---|
| Yamagishi patch ($G=4\pi$, $G=a$, and values with it) | 237 | 48 |
| $G$ kept as a letter (including SI numbers put in) | 187 | 40 |
| $G=1$ (including $G=c=1$ and Planck units) | 57 | 30 |
| $8\pi G=1$ | 18 | 2 |
| rationalized $4\pi G=1$ | 0 | 0 |
| not the $G$ of gravity | 2 | 2 |
| total | 501 | 85 |
The gravity of this body of work has never been written rationalized. $8\pi G=1$ appears only in the 18 places of Paper 1092 and Paper 1093, and Paper 1092 lists the coefficient $1/(8\pi G)$ under the three placings $G=1$, $8\pi G=1$ and $G=4\pi$. In the language of this paper’s theorem, if $\hbar$ is also placed, the patch and $G=1$ ($\hbar=1$) are two that can be passed between by the unit of mass alone, and $8\pi G=1$ with $\hbar=1$ kept is on the side that also moves the units of length and time. Formulas that keep $G$ as a letter have the same form in any units.
The patch label in the scan of powers of 2
The scan of powers of 2 of the same day picked up, from the formulas of this body of work, coefficients that are multiples of 16 from 16 upward and power forms such as $(4\pi)^n$, and divided them into labels by root of the powers of 2. The places whose main label was a power of 4 from the patch (the powers of 4 of $G=4\pi$, $\hbar=1/(4\pi)$ and $m=4\pi$) are 257. This paper read these 257 places against the original sentences for each kind of quantity and divided them into five classes.
| class | places | papers | papers (numbers) | relation to the rescaling |
|---|---|---|---|---|
| products of placings | 141 | 17 | Paper 13, Paper 17, Paper 26, Paper 28, Paper 30, Paper 34, Paper 35, Paper 38, Paper 42, Paper 43, Paper 44, Paper 1086, Paper 1106, Paper 1115, Paper 1128, Paper 1129, Paper 1130 | the exponent of 4 equals the size of the placing power $|r+q-p|$ (the exponent of 2 is twice it; all 141) |
| one placing and one $4\pi$ written in the formula | 20 | 4 | Paper 15, Paper 22, Paper 56, Paper 67 | the $4\pi$ of the placed mass is the placed number ($r$ of Corollary 4); the $4\pi$ of $4\pi\varepsilon_0$ is $\rho$ |
| groupings in the Hawking formulas and others (editions 2 and 3 of Paper 43, the $32\pi^2$ of Paper 44) | 58 | 3 | Paper 43, Paper 44, Paper 50 | the pure powers of $a$ agree with $-\mu+u$; the regroupings of edition 2 are not factors of a rescaling |
| naming only, generic | 5 | 3 | Paper 13, Paper 35, Paper 46 | not a value of a formula |
| powers of $4\pi$ that are not products of placings | 33 | 8 | Paper 05, Paper 06, Paper 08, Paper 09, Paper 11, Paper 15, Paper 57, Paper 1056 | outside the theorem of this paper |
| total | 257 |
The 141 places that are products of placings include $GMm=64\pi^3$ in Paper 35 and Paper 38, $\hbar^2/(2m)=1/(128\pi^3)$ in Paper 17, Paper 1129 and Paper 1130, $U=-16\pi^2M/r$ in Paper 26, $E^2=(4\pi c^2)^2+\cdots$ in Paper 13 and Paper 34, and the $(4\pi)^3$ of $\sigma$ in Paper 43. Computing symbolically the placing power $j=r+q-p$ ($q$ and $p$ the powers of $G$ and $\hbar$) for each formula, the exponent of 2 agreed with $2|j|$ in every place (the exponent of 4 is $|j|$), and the size of the power of $\pi$ also agreed with $|j|$ (the Hawking temperature and the area of the horizon agree after adding the one $\pi$ of the $8\pi$ and $16\pi$ on the side of the formula). They have the form $(4\pi)^j$ with 4 and $\pi$ standing in pairs, the product of the factor of the rescaling of Corollary 1 and the $4\pi$ of placed masses.
Of the 217 places carrying only the patch label, 8 have a power of $\pi$ that is not equal to the exponent of 4 (not half the exponent of 2). They are 5 places of the "bare 64" of edition 2 of Paper 43 (the number left when $\pi^3$ is stripped from $(4\pi)^3$) and 3 places of $32\pi^{5/2}$ in the staircase of Paper 57 (the scan’s list recorded the half-integer power as 1; the $32\pi^{5/2}$ of the original sentence is $(4\pi)^{5/2}$, with 4 and $\pi$ in pairs). A change of units never produces 4 apart from $\pi$, so the bare 64 is not a factor of a rescaling. This is the same conclusion as the place where edition 3 of Paper 43 kept the withdrawal as an operation of stripping $\pi^3$ from $(4\pi)^3$.
The 20 places that are products of one placing and one direct $4\pi$ are $16\pi^2v^2r^2=e^2$, where Paper 15 applied $m=4\pi$ and $\varepsilon_0=r$ to the balance of the Bohr model, the plasma formula of Paper 22 that comes from it, and the $16\pi^2$ of the atom in Paper 56 and Paper 67 (section titles pointing to that quantity included). One $4\pi$ of the $16\pi^2$ is the placed number ($r$ of Corollary 4) and the other is the $4\pi$ of $4\pi\varepsilon_0$, which is a $4\pi$ written directly in the formula (the $\rho$ of Paper 1130). Paper 22 itself treats the $(2\pi)\times(8\pi)=16\pi^2$ it once applied to the structure of a tokamak in its first edition as an arithmetic decomposition; the reading here is a different one. The 7 places of the first-edition ratio $16\pi^2/16\pi^2$ of Paper 15 and the first-edition ratio of Paper 09 listed beside it are not the balance formula and were put in the class of non-products of placings.
In the 58 places of groupings in the Hawking formulas and others, the pure powers of $a$ that Paper 43 wrote with the patch constant kept as a letter, and the places discussing the groupings $2(4\pi)^3$, $4(4\pi)^3$, $(4\pi)^4$, $(16\pi^2)^2$, the bare 64 and $16384\pi^4$ listed by edition 2, stand in the same sentences. This paper does not divide them line by line. The pure powers of $a$ agree with the product of the $(4\pi)^{-\mu}$ of Corollary 1 and the $(4\pi)^u$ of the mass not placed. The regroupings of edition 2 redivide numbers already multiplied by numbers of fact ($8\pi$, $16\pi$, $15360\pi$, $60$), and are not factors of a rescaling. This says from the side of the change of units the conclusion of edition 3 of Paper 43, that the decomposition is not determined because fact and convention were multiplied first. The $8\pi\times4\pi=32\pi^2=2(4\pi)^2$ of Paper 44 (which Paper 44 itself notes is a number multiplying fact and convention) was also put in this class.
The 5 places that only name are sentences such as the one in Paper 46 not to treat $16\pi^2$ and the patch’s $(4\pi)^2$ as the same geometric quantity, a list of names in Paper 13, and a revision record of Paper 35 speaking of the four addresses of $16\pi^2$. The 33 places that are not products of placings include the volume $4\pi\times4\pi$ of the three-sphere of radius 2 (Paper 08), the first-edition ratio $16\pi^2/(4\pi)^2$ (written as an identity by Paper 09 and cited by Paper 11 and Paper 15), the first-edition ratio $16\pi^2/16\pi^2$ of Paper 15, the $64\pi^2$ of the first edition of Paper 05 (a withdrawn construction, also cited by Paper 09), a first-edition name in Paper 06, the dimensional staircase $(4\pi)^{d/2}$ of Paper 57, and a table of numbers in Paper 1056; they lie outside this paper’s theorem.
The breakdown of 16π², 64π³ and 1/(128π³)
Numbers in the patch and numbers in Planck units are listed side by side.
| quantity | number in Planck units | $\mu$ | number in the patch | weight |
|---|---|---|---|---|
| $GM$ ($M$ placed) | $16\pi^2$ | $0$ | $16\pi^2$ | $2$ |
| $GMm/r$ ($M$ and $m$ placed) | $256\pi^4/r$ | $1$ | $64\pi^3/r$ | $4$ |
| $\hbar^2/(2m)$ ($m$ placed) | $1/(32\pi^2)$ | $1$ | $1/(128\pi^3)$ | $-2$ |
| Hawking temperature $\hbar c^3/(8\pi GM)$ ($M$ not placed) | $c^3/(32\pi^2M)$ | $1$ | $c^3/(128\pi^3M)$ | $-2$ |
| Poisson coefficient $4\pi G$ | $4\pi$ | $-1$ | $16\pi^2$ | $1$ |
| Einstein coefficient $8\pi G/c^4$ | $8\pi/c^4$ | $-1$ | $32\pi^2/c^4$ | $1$ |
| Planck mass $\sqrt{\hbar c/G}$ | $\sqrt{c}$ | $1$ | $\sqrt{c}/(4\pi)$ | $0$ |
| $E=mc^2$ ($m$ placed) | $16\pi^2c^2$ | $1$ | $4\pi c^2$ | $2$ |
The $16\pi^2$, $64\pi^3$ and $1/(128\pi^3)$ of the patch are numbers in Planck units multiplied or divided by $4\pi$ as many times as the mass dimension. $GM$ has mass dimension 0, so its number does not move, and the $16\pi^2$ in Planck units (the number of the placed mass $M$) becomes the $16\pi^2$ of the patch as it is. The Hawking temperature and $\hbar^2/(2m)$, which carry the same $128\pi^3$ under the patch, both have the form of division by $32\pi^2$ as numbers in Planck units, and both have mass dimension 1, so each receives one $4\pi$ when moved to the patch. But the breakdown of $32\pi^2$ differs even in Planck units: for $\hbar^2/(2m)$ it is the $16\pi^2$ of the placed mass and the 2 of motion, and for the Hawking temperature the $8\pi$ of the formula and the $4\pi$ of the mass not placed. That the sum alone does not fix the allocation (Paper 1130) does not change when seen in numbers in Planck units. The $16\pi^2$ of the Poisson coefficient and the $32\pi^2$ of the Einstein coefficient are the $4\pi$ and $8\pi$ in Planck units (numbers written directly in the formulas) multiplied by one $4\pi$ for mass dimension $-1$.
8. Places in earlier papers worth checking
In the light of this paper’s theorem, three places are listed where it is worth adding a condition when reading. None of them denies the numbers or conclusions of the paper.
Paper 39
Paper 39 writes of $l_p=c^{-3/2}$ that spending the remaining one on $c=1$ gives $l_p=1$, and continues that this means the Planck length becomes the unit of length, which is the very definition of Planck units. By the theorem, setting $c=1$ in the patch makes the units of length and time the Planck units themselves, and this part holds. The unit of mass is $4\pi$ times the Planck mass ($m_P=1/(4\pi)$), so it is worth reading "the definition of Planck units" as a statement about the units of length and time, with the condition that the unit of mass is $4\pi m_P$. Paper 39 also writes, in another place, that adopting $(\hbar,G,c)$ gives Planck units and the unit of mass is fixed to be the Planck mass. The sentence before it is in the context of the values of the patch, so if $(\hbar,G,c)$ is adopted with the values of the patch, the unit of mass there too is read as $4\pi$ times the Planck mass (the Planck mass if $\hbar=G=c=1$ is adopted). The bookkeeping of the budget in the same place, that either may be adopted but not both ($c=1$ and $m=4\pi$ cannot both be placed), agrees with the theorem.
Paper 74
Paper 74 reads the Einstein-Hilbert coupling $1/(16\pi G)$ as an elastic constant of spacetime, writes that the patch value $G=4\pi$ (Paper 35) fixes it in the solid-angle units of the body of work, and, as an open problem, asks whether the structure of the body of work supplies a candidate that would produce that elastic constant as a computed output rather than an assigned input. In the language of the theorem, $G=4\pi$ is the number $G$ takes when $\hbar$ is placed at $1/(4\pi)$ and the unit of mass is taken to be $4\pi$ times that of the reference with the same number for $c$ (the unit of mass $4\pi m_P$ when $c=1$). Computing $G=4\pi$ from the structure of the body of work would be producing a choice of units, not producing the measured value of $G$. Paper 1115 also writes that it has not solved computing G = 4π or S = A/4 from the structure of the body of work (the open problem of Paper 74). This question is worth reading as a question about the choice of units.
Paper 63
Paper 63, noting that it works under $G=4\pi$ (the Yamagishi patch, Paper 35), writes the Einstein equation as $G_{\mu\nu}=(32\pi^2/c^4)T_{\mu\nu}$. As in the table of breakdowns, this $32\pi^2$ is the $8\pi$ of $8\pi/c^4$ in Planck units (a number written directly in the formula) multiplied by one $4\pi$ for the mass dimension $-1$ of the Einstein coefficient. Paper 44 notes that the same number is a number multiplying fact and convention. Paper 63 does not add this note, so it is worth adding this breakdown when reading.
9. Separator
There is one thing to divide: whether only the unit of mass was moved, or the units of length and time as well.
Keeping the number of $c$, there are two ways to tell. One is whether the shift of the power of $\pi$ in numbers is fixed by the mass dimension alone: with only the unit of mass it is a shift of $(4\pi)^{-s\mu}$, and if the units of length and time also move, the dimensions of length and time enter, as in the $(\mu-\lambda-\tau)/2$ of rationalization. The other is whether the numbers of $\hbar G$ and of the Planck length stay as they are: $\hbar G$ has mass dimension 0, so if only the unit of mass was moved its number does not move. In the check, the second test agreed with the units of length and time not moving for all six placings. The first follows from the formula of the factors, and its converse (if a ledger of the form $\epsilon+s\mu$ stands, the factor for length and time carries no $\pi$) also holds.
Among the placings of this body of work, the patch and $G=1$ ($\hbar=1$) are linked by the unit of mass alone, and rationalization and $8\pi G=1$ with $\hbar=1$ kept are on the side that also moves the units of length and time. The weight of the parity law is unmoved by $G$, $\hbar$ and $c$ when one is on the former side.
This separator is on a different axis from the separator of Paper 1130 (whether that $4\pi$ was placed in a constant, placed in a mass, or written directly in the formula) and from that of Paper 1129 (whether that $4\pi$ is now coming from $\int d\Omega$ or has been carried as a number placed earlier): it looks not at which $4\pi$ one is looking at, but at which of the units was moved.
10. Nearby work outside this body of work
The set of values $\hbar=4\pi$, the number $4\pi$ for the Planck mass, and lengths and times left at their Planck values is the same as in the last row of the table in Section 6 (rationalization with $\hbar=4\pi$), and Wutke (2023) built this set by another road. Wutke carried the rationalization that Heaviside applied to the equations of electromagnetism over to Newton’s law, built a system of units in which the Planck length and time stay as they are and only the Planck mass becomes $4\pi$ ($c=1$, $h=8\pi^2$, that is $\hbar=4\pi$), and wrote that the values of the Planck quantities other than the Planck mass do not change. Re-choosing only the mass side of Planck units by $4\pi$ has the same shape as the theorem of this paper.
There are three differences. The first is the direction: Wutke’s set of values corresponds to $a=1/(4\pi)$ in the family $G=a$, $\hbar=1/a$ of this paper (the unit of mass taken as $m_P/(4\pi)$), while the Yamagishi patch corresponds to $a=4\pi$ (the unit of mass $4\pi m_P$). This correspondence is this paper’s reading, taken from the values in the table of Section 6. The second is the procedure: Wutke changes the dimension of mass itself to length cubed over time squared and removes the kilogram from the base units, which differs from the change of units in this paper, which keeps dimensions and changes only numbers. Third, the weight $\epsilon+\mu$ of Paper 1130 and the count of powers of $\pi$ in numbers do not appear in Wutke’s paper. Since Wutke’s system does not keep $\hbar=1$, it does not conflict with the statement of Section 6 that rationalization with $\hbar=1$ kept also moves the units of length and time.
The relation between the $4\pi$ and $8\pi$ conventions of gravity and the definition of the reduced Planck mass is written by Robinson (2006) in arbitrary spacetime dimension. Section 6 of this paper lines such conventions up again by the factors of a change of units that keeps $c$ (the unit of mass, and the units of length and time).
In the public indexes of papers searched on 21 November 2026 (OpenAlex, Crossref, DataCite and arXiv), no work was found that writes the Yamagishi patch as Planck units with only the unit of mass multiplied by $4\pi$, or the weight $\epsilon+\mu$ as the power of $\pi$ in Planck-unit numbers. The theorem and the corollaries are elementary, so it would not be surprising if the same thing is written in textbooks.
11. The check
All numbers and formulas were produced by the check (v1132.py, Python 3 and sympy). From the rule of changing units the changes of the numbers of $G$, $\hbar$ and $c$ were derived, and the theorem and the factors for a general $(a,b)$ were confirmed by solving the simultaneous equations symbolically. The corollaries were confirmed by computing numbers in the patch and in Planck units by symbolic substitution for 32 quantities and 300 random products and comparing them. The condition for a ledger of the form $\epsilon+s\mu$ was solved with the powers of $\pi$ in the factors kept as letters. For the census, the fingerprints (SHA256) of the lists of the two scans were fixed, the counts of conventions of gravity were recounted, and the check confirmed that the reading table for the 257 places with the patch label covers each line of the list once, the counts of classes, and, for each formula key, the exponents of 4 and the powers of $\pi$ of the 141 places that are products of placings. Quoted sentences were confirmed to be present word for word in the Japanese .tex of the earlier papers by plain string search. There are 44 checks, each written so that it fails if the claim is wrong.
| section of the check | what it confirms | what falls if it fails |
|---|---|---|
| 1 | rule of rescaling, uniqueness in the theorem, general $(a,b)$, reference for each number of $c$, Planck quantities at $c=1$, $4\pi m_P$ | Sections 2 and 3 |
| 2 | Corollaries 1, 2 and 4 on 32 quantities and 300 random products, a failing example, weights of $G$, $\hbar$, $c$, constants not placed | Section 4 |
| 3 | four types of Paper 1128, two scenes and the ratio of Paper 1129, family of Paper 116, powers of $a$ in Paper 43 | Section 5 |
| 4 | factors for six placings, shift of the power of $\pi$ under rationalization and the books that stand there, books within the family, the converse, the test, difference 1 in 12 cases | Sections 6 and 9 |
| 5 | fingerprints of the two scan lists, counts of conventions of gravity, classes and formula keys of the 257 places of $P$, table of breakdowns | Section 7 |
| 6 | quoted sentences present word for word, zero paragraphs that write the patch as a rescaling of the unit of mass | Sections 1, 5 and 8 |
12. Scope of this paper
The items of this paper are sorted by what they are.
| item | what it is |
|---|---|
| numbers changing by $x^{-\mu}y^{-\lambda}z^{-\tau}$ under rescaling | known rule (Bridgman) |
| the patch being a rescaling of the unit of mass only, and the only one | theorem (settled by symbolic computation) |
| numbers changing by $(4\pi)^{-\mu}$, the weight as the power of $\pi$ in Planck units | corollaries of the theorem |
| the split $2r+u+\rho=(r+u)+r+\rho$ | reading of a corollary |
| readings of numbers in Paper 1128, Paper 1129, Paper 116 and Paper 43 | computations applying the corollaries |
| factors for rationalization and $8\pi G=1$, and the condition for a book of the form $\epsilon+s\mu$ | computations with the same rule |
| counts of the 501 places of conventions of gravity | counts as read by the scan (census) |
| classes of the 257 places of $P$ | reading of this paper (one reader; Paper 43, Paper 44 and Paper 50 not divided line by line) |
| places in Paper 39, Paper 74 and Paper 63 | places worth checking (not called errors) |
| whether the $4\pi$ of the patch is a solid angle or a convention | not decided |
| the measured value of $G$ | not produced (a convention is a way of placing numbers) |
| a new law of physics | not claimed |
Closing
The Yamagishi patch G = 4π, ħ = 1/(4π) is obtained from the units with G = ħ = 1 (and the same number for c) by multiplying only the unit of mass by 4π, without moving the units of length and time. Keeping the number of c, the change of units linking the two has only this one form. With c = 1 the reference is the true Planck units, and the unit of mass of the patch is 4π times the Planck mass. The number of a quantity of mass dimension μ is multiplied by 4π to the power −μ, so the weight ε + μ of the parity law is the power of π of the same quantity written as a number in Planck units, and G and ħ do not move the weight because both are 1, carrying no π, in Planck units. The ledger formula 2r + u + ρ splits into one 4π of the rescaling for each mass, one more 4π of its own number for each placed mass, and the 4π written directly in the formula. The reason Paper 1128 gave, that the 4π of the types appears once because the exponent of mass is 1, goes back to only the unit of mass having been moved, and the distinction between the two scenes of Paper 1129 is reproduced by the change of units as it is. Rationalization and 8πG = 1 with ħ = 1 kept also move the units of length and time, and there a ledger of the form ε + sμ does not stand. Because the 4π in their ratio has no dimension, the Gauss coefficient is heavier by 1 in weight than the inverse-square coefficient in every convention. There is one thing to divide: whether only the unit of mass was moved, or the units of length and time as well.
Revision Record
2nd Edition (November 2026): Section 10, “Nearby work outside this body of work”, was added, and the sentence of the scope note “This paper has not searched literature outside the body of work (no web was used)” was replaced by a sentence pointing to it. It cites the rationalized Planck units of Wutke (2023), in which only the Planck mass becomes 4π, and states that they have the same set of values as the row with ħ = 4π in the table of Section 6, that their direction is opposite to the Yamagishi patch, and that the procedure changes the dimension of mass; Robinson (2006) was also added. Two references were added. The theorem, the corollaries, the numbers and the census are unchanged.
References
- Bridgman, P. W. (1922). Dimensional Analysis. Yale University Press.
- National Institute of Standards and Technology (2024). CODATA Recommended Values of the Fundamental Physical Constants: 2022. NIST.
- Robinson, S. P. (2006). Normalization conventions for Newton’s constant and the Planck scale in arbitrary spacetime dimension. arXiv:gr-qc/0609060.
- Wutke, A. (2023). From Newton to universal Planck natural units – disentangling the constants of nature. Journal of Physics Communications. DOI: 10.1088/2399-6528/ad0090.
- Yamagishi, Y. (2026). The Yang--Mills Mass Gap and Dimensional Transmutation [Paper 05]. YOI: https://www.yamagishiyuuki.com/yoi/5; DOI: 10.5281/zenodo.21404236
- Yamagishi, Y. (2026). The Critical Line and the Fixed Point of the Functional Equation [Paper 06]. YOI: https://www.yamagishiyuuki.com/yoi/6; DOI: 10.5281/zenodo.21404262
- Yamagishi, Y. (2026). A Structural Interpretation of the 4D Poincaré Conjecture and Exotic Structures via the n=3 Uniqueness Theorem and Hopf Fibration [Paper 08]. YOI: https://www.yamagishiyuuki.com/yoi/8; DOI: 10.5281/zenodo.21404419
- Yamagishi, Y. (2026). The Geometric Origin of 16π² in the Higgs Quadratic Divergence [Paper 09]. YOI: https://www.yamagishiyuuki.com/yoi/9; DOI: 10.5281/zenodo.21404525
- Yamagishi, Y. (2026). Factors That Cancel in Normalisation, and the Exponent 2 of Born’s Rule [Paper 11]. YOI: https://www.yamagishiyuuki.com/yoi/11; DOI: 10.5281/zenodo.21404636
- Yamagishi, Y. (2026). Physical Law as a Description Language of the Universe OS [Paper 13]. YOI: https://www.yamagishiyuuki.com/yoi/13; DOI: 10.5281/zenodo.21404917
- Yamagishi, Y. (2026). A Convention Is a Number, Not a Function of the Configuration [Paper 15]. YOI: https://www.yamagishiyuuki.com/yoi/15; DOI: 10.5281/zenodo.21600352
- Yamagishi, Y. (2026). The 8π Is an Artefact of the Units [Paper 17]. YOI: https://www.yamagishiyuuki.com/yoi/17; DOI: 10.5281/zenodo.21600813
- Yamagishi, Y. (2026). The Yamagishi Zero-Tension Plasma Equation [Paper 22]. YOI: https://www.yamagishiyuuki.com/yoi/22; DOI: 10.5281/zenodo.21601236
- Yamagishi, Y. (2026). The Yamagishi Classical Mechanics Reformulation [Paper 26]. YOI: https://www.yamagishiyuuki.com/yoi/26; DOI: 10.5281/zenodo.21601339
- Yamagishi, Y. (2026). The Yamagishi-Kepler Equation [Paper 28]. YOI: https://www.yamagishiyuuki.com/yoi/28; DOI: 10.5281/zenodo.21430531
- Yamagishi, Y. (2026). The Yamagishi--Adler--Bardeen Note [Paper 29]. YOI: https://www.yamagishiyuuki.com/yoi/29; DOI: 10.5281/zenodo.21601581
- Yamagishi, Y. (2026). The Yamagishi Double-Torus Stellar Fortress Equation [Paper 30]. YOI: https://www.yamagishiyuuki.com/yoi/30; DOI: 10.5281/zenodo.21601732
- Yamagishi, Y. (2026). The Yamagishi Unified Budget Canvas Theory [Paper 32]. YOI: https://www.yamagishiyuuki.com/yoi/32; DOI: 10.5281/zenodo.21601849
- Yamagishi, Y. (2026). A Census of the c⁵s [Paper 34]. YOI: https://www.yamagishiyuuki.com/yoi/34; DOI: 10.5281/zenodo.21430340
- Yamagishi, Y. (2026). The Yamagishi Gravitational Constant Unpacking Theory [Paper 35]. YOI: https://www.yamagishiyuuki.com/yoi/35; DOI: 10.5281/zenodo.21430118
- Yamagishi, Y. (2026). The Yamagishi Master Table of Classical-Physics Geometric Unpacking [Paper 38]. YOI: https://www.yamagishiyuuki.com/yoi/38; DOI: 10.5281/zenodo.21430234
- Yamagishi, Y. (2026). The Yamagishi Planck Length Unpacking Theory (l_p=c^-3/2) [Paper 39]. YOI: https://www.yamagishiyuuki.com/yoi/39; DOI: 10.5281/zenodo.21430490
- Yamagishi, Y. (2026). The Four Planck Quantities Are Four Instances of One Rule [Paper 40]. YOI: https://www.yamagishiyuuki.com/yoi/40; DOI: 10.5281/zenodo.21435739
- Yamagishi, Y. (2026). The G and the c⁵ of the Quadrupole Formula Cannot Be Separated [Paper 42]. YOI: https://www.yamagishiyuuki.com/yoi/42; DOI: 10.5281/zenodo.21601988
- Yamagishi, Y. (2026). Keep the Patch Constant as a Letter and the Ambiguity Dissolves [Paper 43]. YOI: https://www.yamagishiyuuki.com/yoi/43; DOI: 10.5281/zenodo.21602132
- Yamagishi, Y. (2026). The Power of 4π Is the Mass Dimension [Paper 44]. YOI: https://www.yamagishiyuuki.com/yoi/44; DOI: 10.5281/zenodo.21602231
- Yamagishi, Y. (2026). Of 16π², Only Ω₃=2π² Is Geometry [Paper 46]. YOI: https://www.yamagishiyuuki.com/yoi/46; DOI: 10.5281/zenodo.21602422
- Yamagishi, Y. (2026). The Internal Structure of the Exponent 7 Is Not Unique [Paper 50]. YOI: https://www.yamagishiyuuki.com/yoi/50; DOI: 10.5281/zenodo.21602796
- Yamagishi, Y. (2026). A Structural Study of the 4π Patch [Paper 56]. YOI: https://www.yamagishiyuuki.com/yoi/56; DOI: 10.5281/zenodo.21597500
- Yamagishi, Y. (2026). A Kaluza--Klein Re-entry [Paper 57]. YOI: https://www.yamagishiyuuki.com/yoi/57; DOI: 10.5281/zenodo.21597608
- Yamagishi, Y. (2026). The Velocity--Spacetime Curvature Duality [Paper 63]. YOI: https://www.yamagishiyuuki.com/yoi/63; DOI: 10.5281/zenodo.21606250
- Yamagishi, Y. (2026). The Euler-Characteristic Origin of the 4π Patch [Paper 67]. YOI: https://www.yamagishiyuuki.com/yoi/67; DOI: 10.5281/zenodo.21642170
- Yamagishi, Y. (2026). The Yamagishi Patch within Emergent Gravity [Paper 74]. YOI: https://www.yamagishiyuuki.com/yoi/74; DOI: 10.5281/zenodo.21670102
- Yamagishi, Y. (2026). What Does the Exponent of c Count? [Paper 116]. YOI: https://www.yamagishiyuuki.com/yoi/116; DOI: 10.5281/zenodo.22003794
- Yamagishi, Y. (2026). The Fine-Structure Constant Does Not Move with the Units, but It Moves with Energy [Paper 1056]. YOI: https://www.yamagishiyuuki.com/yoi/1056; DOI: 10.5281/zenodo.22905436
- Yamagishi, Y. (2026). Setting m = 4π Turns h/m into ħ/2, and Its 2 Has a Different Root from the 2 of the Uncertainty Bound [Paper 1083]. YOI: https://www.yamagishiyuuki.com/yoi/1083
- Yamagishi, Y. (2026). The π in the Formulas of Gravity Appears with Density and Vanishes with Mass; What Remains Is the Angle of a Revolution [Paper 1086]. YOI: https://www.yamagishiyuuki.com/yoi/1086
- Yamagishi, Y. (2026). The “2” of the Inverse-Square Density Is Two 2s, and the 2 of the Shell Mass Meets the 2 of the Horizon Balance Only in Three Dimensions [Paper 1092]. YOI: https://www.yamagishiyuuki.com/yoi/1092
- Yamagishi, Y. (2026). In Three Dimensions the Entropy of a Black Hole Is the Same Number as the Area of the Horizon’s Cross-Sectional Disk, and the Shadow Disk in the Photographs Is 27/4 Times Larger [Paper 1093]. YOI: https://www.yamagishiyuuki.com/yoi/1093
- Yamagishi, Y. (2026). A Physical Infinity Is Set by How Fast Things Thin Out against the Dimension, Not by the Mass [Paper 1106]. YOI: https://www.yamagishiyuuki.com/yoi/1106
- Yamagishi, Y. (2026). A Point’s 4π Gives Way to a Line’s or a Sheet’s Law at About the Spacing, and the Two 2π Trade Places with the Units [Paper 1112]. YOI: https://www.yamagishiyuuki.com/yoi/1112
- Yamagishi, Y. (2026). E = 4πc² Reads as Energy Shared Out over the Cells of the Light Sphere, and Its Total Is the Same on Every Skin Only for a Massless Field [Paper 1115]. YOI: https://www.yamagishiyuuki.com/yoi/1115
- Yamagishi, Y. (2026). Multiplying by a Speed Once Raises at Once the Ladder of Point, Line, Skin and Flesh Types and the Geometric Sequence of Planck Quantities with Ratio c [Paper 1128]. YOI: https://www.yamagishiyuuki.com/yoi/1128
- Yamagishi, Y. (2026). The 4π of m = 4π Is a Solid Angle Only While the Mass Inside the Inverse-Square Ball Is Being Counted [Paper 1129]. YOI: https://www.yamagishiyuuki.com/yoi/1129
- Yamagishi, Y. (2026). Under the Yamagishi Patch, the Power of 4π Plus the Mass Dimension Is Unchanged by Multiplying In G, ħ and c, and Only Masses and an Explicit 4π Move It [Paper 1130]. YOI: https://www.yamagishiyuuki.com/yoi/1130
Reference
Yamagishi, Y. (2026). The Yamagishi Patch Is Planck Units with Only the Unit of Mass Rescaled by 4π [Paper 1132]. YOI: https://www.yamagishiyuuki.com/yoi/1132
ORCID: 0009-0006-8587-3450
PDF (English) · PDF (Japanese) · Date proof of the English PDF (OpenTimestamps .ots) · SHA256SUMS
The record page for this paper (in Japanese, with the PDFs in each language and the date proof): YOI 1132. This paper is a preprint and has not been peer reviewed.