Math & Science ⇒ What should be the natural spacement between equations?

[Cham]

I wrote a large document full of equations that I vertically separated using commands like \\[12pt], \\[18pt] (see the example below). The document was written with a 12pts font size.

Then, I decided that 12pts font was too big and reduced it to 11pts. The spacements between equations need to be adjusted, and I guess that I shouldn't have used commands like \\[12pt] and \\[18pt].

Here's a MWE example showing the issue:

Code: Select all

Code, edit and compile here:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
\documentclass[11pt,letterpaper,twoside]{book}
\usepackage{lmodern}
\usepackage{amsmath}
\begin{document}
 
Blablabla:
    \begin{gather*}
        x^2 - 3 y = 2, \\[12pt]
        \frac{1}{2} \, x y - z^3 = 2, \\[12pt]
        \frac{3}{4} \, x - y = z^2. \\[12pt]
        z + x - y = 1. \\[12pt]
        z^2 - y^2 = 0.
    \end{gather*}
Blabla
    \begin{align*}
        x+ 3 y &= 2, \\[12pt]
        \frac{1}{2} \, x^2 - z &= 2, \\[12pt]
        \frac{2}{5} \, z + z^2 &= y. \\[12pt]
        z + x - y &= 1. \\[12pt]
        x^2 - y^2 &= -1.
    \end{align*}
\end{document}
 
 
הההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההה
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Run LaTeX here

I want all the equations to be equally spaced, no variable vertical spacings.

What should be the proper commands to vertically dispose the equations, in a font size independant way, and so that the spacings stay constant for any set of equations in any environment (array, align, aligned, gather, gathered, ...)?

[Johannes_B]

Well, the simple way would be to not use manual spacing and let TeX do its job. Knuth designed it to prepare beautiful books.

[Cham]

Well, that's not very specific. I'm already asking how to do this properly!

Do you have a suggestion code? I tried using \\, but the spacing is way to short (and variable). Using \\ \\ gives some spacing troubles too.

[Johannes_B]

I would suggest to use a single \\ at the end of an equation line. The spacing will be different, as very equation needs a bit of variable space.

Your constriction of exact space, isn't what I would enforce.

[Stefan Kottwitz]

Hi Cham,

take a look at this topic. Instead of 0pt as there, you can choose any you want for all equations with multiple lines at one. And for interline space in multiple eqautions, as above, you can globally adjust \jot, such as

\settolength{\jot}{12pt}

Stefan

[Cham]

I guess you mean \setlength{\jot}{12pt} ?

[Stefan Kottwitz]

Yes, mixed it with \addtolength,

Stefan

[Cham]

Well, I now have another point of view that I would like to try.

Before, I was using fixed spaces like \\[6pt], \\[12pt], \\[18pt] and even \\[24pt] (depending on the size and "type" of equations), for equations in a 12pts font document.

I hate it when LaTeX change the spacements to adapt vertical spaces to the equations. I prefer to get nice looking constant vertical spacings, even if I have to do it myself.

Now I'm writing my docs with 11pts font. I want the vertical spacings of equations to be constant and stay nice with this font size (and even if I change the font size!), without having to use some multiple of 11pts (which is a prime number, so can't use \\[5.5pts], for example)!

So are there vertical spacing commands for math mode that would do the equivalent of what I was doing before (i.e half and multiples of 12pts), instead of using something like \\[11pts], \\[16.5pts] or \\[22pts] ? For example, something like \medskip, \bigskip made for the math mode and the align, gather and other math environments?

[rais]

0.5\baselineskip, perhaps?

KR
Rainer

[Cham]

Here's a MWE showing my troubles:

Code: Select all

Code, edit and compile here:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
\documentclass[11pt,letterpaper,twoside]{book}
\usepackage{lmodern}
\usepackage{amsmath}
%\setlength{\jot}{\baselineskip}
\begin{document}
 
Blabla:
    \begin{gather}
        x^2 - \frac{1}{2} \, x = 3, \\
        \frac{a^2}{b^3} \, y + 2 x \, y = y^2, \\
        a - b = c.      
    \end{gather}
Bla:
    \begin{gather}
        x^2 - \frac{1}{2} \, x = 3, \\[11pt]
        \frac{a^2}{b^3} \, y + 2 x \, y = y^2, \\[11pt]
        a - b = c.      
    \end{gather}
Blabla:
    \begin{gather}
        \setlength{\jot}{12ex}
        x^2 - \frac{1}{2} \, x = 3, \\[1ex]
        \frac{a^2}{b^3} \, y + 2 x \, y = y^2, \\[1ex]
        a - b = c.      
    \end{gather}
\end{document}
 
 
הההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההה
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Run LaTeX here

Here's a preview:

vspacements.jpg (26.38 KiB) Viewed 9380 times

As you can see from the equation numbers on this preview, the vertical spacing is variable, and I hate that!

I would like to define some small macros to replace \\ and \\[11pt] and the likes to something that would adjust with the font size (11pts here), and that makes the vertical spacing between "=" to be the same. I don't want LaTeX to make variable vertical spacings between equations. Is that possible?**

** To solve the issue, I was using an ugly hack: smash the fractions and summation signs in equations and use \\[11pt], or \\[22pts] to adjust the spacing according to the equations. This is certainly not a good way of doing things!