Spaces and the units of measurement for music notation

Tips

What is the distance between two lines on a staff?

If you’re thinking in millimeters, inches, or points, think again. The distance between two lines is always one space.

The most fundamental element of conventional Western music notation is the five-line staff, the quintet of equidistant lines that represent musical pitches upon which noteheads are placed. Filling the gaps between those lines are spaces, also representing pitches upon which other noteheads are placed. And with that, you’re off and running creating notated music.

But it’s the space as a unit of measurement that defines the appearance of all other notational items. Everything can be measured as it relates to the space. It’s so essential that it’s the very first item mentioned by Elaine Gould in her music notation reference Behind Bars, and it can be scaled regardless of the absolute size of the music.

Yet even if you’re an experienced musician, if you’ve never prepared notation music you’ve probably not given much thought to the idea. When you start working with music notation software it might be puzzling to poke around in your program’s settings and see “0.0918” or “1/8 spaces”. But these basic settings matter as much as anything to affect how your music appears.

Staff lines, stems, barlines

Staff lines

Indeed, the thickness of the aforementioned staff line itself is measured in spaces. You might think that its thickness would be a matter of settled notation law, but in the three major notation programs, the thickness of a staff line in a default is different in each:

Can’t we all just get along? These differences are what gives default Finale documents a delicate look and it’s what makes Dorico scores look robust.

It is often a good idea for the relative thickness (the value in spaces) of the staff line to increase as the absolute size (the value in millimeters or inches) of the staff decreases. This is so that staff lines don’t become invisible on smaller staves.

Stems

Although Gould says, “Stems should be thinner than the stave-line, but not so thin as to reproduce too faintly,” often the thickness of the stem and the staff line are identical. Here, again, the programs vary among each other, but hew closely or identically to the staff line thickness:

Barlines

When it comes to barline thickness, Gould takes a firmer stance. “The barline is thicker than a stave-line, and therefore conspicuously thicker than a stem. It is very important that barlines stand out from stems, especially when there are long, complicated bars in a single-stave instrumental part.”

The results from a Finale default document, therefore, are unsatisfactory, where barline thickness measures the same as staff lines and stems. Sibelius and Dorico fare better:

Practically speaking, these settings result in the appearance of the music notation in the following way:

Document settings

All of the above settings can be customized to your liking, of course. Here’s where they can be found:

  • Finale: Document > Document Options
  • Sibelius: Appearance > House Style > Engraving Rules
  • Dorico: Engrave > Engraving Options

Now that you know what a space is, you’ll notice these settings all across your document. The optimal settings for every item are a matter of taste and style, but legibility should always be paramount.

You might ask, though, what about units such as inches, millimeters, and points? These settings, which all refer to absolute measurements, also appear in documents.

Inches/millimeters

Let’s take the first two. Generally, you’d use inches or millimeters (depending upon your localization) to refer to page layout items, such as the page size, page margins, and staff margins.

Most crucially, of course, you’ll also use this to define the height of your staff.

In Finale, these settings are found in Document > Page Format:

In Sibelius, in Layout > Document Setup:

 

And as shown here in Dorico, in Setup > Layout Options > Page Setup.

Points

Points are a little different. Points generally refer to the size of text, defined as 1/72 of an inch (about 0.353 mm). You see this all the time in word processing programs and many other applications, including the notation programs.

Points are technically an absolute unit of measurement, but in all of the notation programs, you have the option to define text size as either an absolute (fixed) size, or relative to the size of the staff. Absolute sizes are generally better for titles, page headers, and page numbers, while relative sizes are generally better for staff text. System text (such as tempos and rehearsal marks) can be either absolute or relative, depending on your situation. You don’t want to scale those items to be so small in a conductor score as to be unreadable, but it’s often helpful for those items to bear some relationship to the musical elements.

The question is, though, if you choose relative sizes, to what setting is the size relative?

Behind the scenes, each of their programs have their baseline staff sizes. In Sibelius it’s 7.0 mm…

… but in Finale it’s 24 pt, which is exactly 1/3 inch (about 8.47 mm).

In Dorico it’s 20 pt, which is just ever so slightly greater than 7 mm (7.05556 mm).

What does this mean for you? It means that at those staff sizes, the absolute size of your text will equal the relative size. In other words, 10 pt relative text will appear legitimately as 10 points, if applied to a Finale staff at 8.47 mm, a Sibelius staff staff at 7.0 mm, or a Dorico staff at approximately 7.05 mm.

However, if you have a Finale staff at 7.0 mm, 10 pt relative text will appear smaller, at about 8.3 pt, or only about 83% of the size it would appear in Sibelius or Dorico, because Finale’s baseline staff is larger.

Be aware in Finale

If all of that seems a bit confusing, well, here’s a bit more.

Sibelius and Dorico do a good job of keeping spaces, inches, millimeters, and points in their proper places.

In Finale, though, you might have noticed that you can set your default measurement units to any one of these, in the Finale menu:

(Let’s not even get into what an EVPU is — you can look it up.)

The problem is, saying that a stem has a thickness of 0.00765 inches is meaningless, unless you only use a staff size of 1/3 inch (Finale’s baseline). You should always think in terms of spaces for notational elements.

Fortunately, you can choose a different unit of measurement in certain dialogs, like here, in Document Options, as seen at the bottom (Units:)

Power tip: You don’t even have to change the unit of measurement to enter spaces. Simply type s after the number and Finale will process your entry as spaces. (The same works for i for inches or m for millimeters.)

With all of this information now at your disposal, we’ll meter our thoughts and inch away to give you space to create some beautiful-looking music.

Comments

  1. Ben

    Dorico uses an absolute staff of 20pt (7.05556 mm). Finale’s 0.3333 (third of an inch) is 72 / 3 = 24pt. The default music fonts for each program are designed with these point sizes in mind.

    1. Philip Rothman

      Thanks, Ben. I tried to round it off but we may as well be precise, so I’ve updated the post.

    2. mirabilos

      MuseScore uses a relative staff by default, scaled by the spatium (space unit). The default spatium in MuseScore (2) is an amazingly odd 1.764 mm, which makes for a default total staff size of (4 * 1 sp) + (5 * 0.08 sp) = 4.4 sp = 7.761 mm.

      In imperialistic units, the spatium is 0.069 in (just as amazingly odd), for a default total staff size of 0.303 in (which makes 7.696 mm, so rounding errors are guaranteed when switching back and forth).

      In the source code, I find this…

      libmscore/mscore.h:static constexpr qreal SPATIUM20 = 5.0;

      … which appears to be in dpi, which would make for a _real_ default spatium of 5/72″ or 0.0694̅ in or 1.7638̅ mm which accounts for those odd numbers (but still isn’t fun with the accumulating rounding errors).

      I usually set my spatium to 1.675 mm (small), 1.75 mm (medium / normal), 1.8 mm (large), since I mostly do vocal scores, and the elderly ladies are always happy because they can read my editions better than the tinier commercial ones. (These make for 7.37/7.7/7.9 mm staves.)

      1. Philip Rothman

        Terrific, thanks for this additional information. Your resulting staff sizes seem very sensible.

        1. mirabilos

          As I wrote in the other comment, they assumed interior model, whereas I found out from the comments here that MuseScore uses midpoint model.

          That means we have 6.7 mm (small), 7 mm (medium / normal), 7.2 mm (large) staff sizes here, and MuseScore’s default staff size is 7.05̅ mm (0.27̅″ or 20 pt) modulo rounding errors.

  2. Waldbaer

    Interesting to read such basic stuff, especially including the comparison how the major scoring apps handle this. I’m sure I’m not the only one who does not care about this very often and found something to learn here…

    Of course I had to look up EVPUs (since you asked ;-) ). “Finally” I found this: http://www.finalemusic.fr/Finale2k14_Help/win/Content/Finale/ID_MENU_OPTS_UNITS_EVPUS.htm
    I still wonder what this might be used for. Is it some kind of historical standard? I could not find any further information about it on a quick search…

    1. Philip Rothman

      Thanks, Waldbaer. Perhaps one day we will delve into this historical esoterica…

    2. Ben

      When Finale was first conceived, an EVPU was defined as the small unit of distance that the program could calculate. If memory serves, it may even tie in to pixel-matching on early 300dpi laser printers.

      Now, Finale can cope with fractions of an EVPU, rendering them somewhat pointless to the user, though Finale’s spacing engine may still think in terms of them.

      1. Philip Rothman

        Thanks for that info. Quite an enigma.

  3. mirabilos

    I’ve checked the values for MuseScore, and both MuseScore 2 and 3 use the same:

    Staff lines: 0.08 sp
    Stems: 0.13 sp
    Barlines: 0.16 sp

    The spatium is set in 「Layout → Page Settings…」 in 2.x and in 「Format → Page Settings…」 in 3.x.

    The baseline for staff size scaling is not documented, in https://musescore.org/en/node/288947 I’ve asked upstream for the value and will report back one I know them.

    If you wish, I can also provide corresponding screenshots. I’d love for information like this about MuseScore to be included in your articles, so they don’t smell like the commercial softwares have paid for the articles.

    1. Philip Rothman

      Thanks, Mirabilos. I really appreciate it. It’s enough work writing up info for three programs, so I can’t guarantee that I’ll update the post :-) But please do report back in the comments what you find out about the baseline staff size for MuseScore.

      In the meantime, I can assure you that this article and all other posts on Scoring Notes are not paid for by any company. If we ever do run a sponsored post, it will be thoroughly labeled as such.

      1. mirabilos

        OK, sure, and thanks for the extra information.

        In the meantime, we’re relatively sure that the default spatium of 5 pt (see my other comment) is the scaling baseline.

        The two corresponding screenshots for MuseScore are http://fish.mirbsd.org/~tg/tmp/a1.png and http://fish.mirbsd.org/~tg/tmp/a2.png (I would appreciate copying these, not deeplinking, as that’s a temporary location in my user home space on a low-spec server), and MuseScore also “do[es] a good job of keeping spaces, inches, millimeters, and points in their proper places”: almost everything (except things like page size and margins, where it does not make sense) is in spatium, and those which aren’t often have a checkbox whether to scale them with spatium or to take them as absolute value (and except the score title/subtitle/composer/lyricist texts, all default to scaling).

        1. mirabilos

  4. Charles Gaskell

    Can you clarify whether “one space” is the internal distance between two lines (i.e. the bottom of the higher line to the top of the lower line) or the mid distance between the two lines (i.e. top of the higher line to the lower line, or mid-point to mid-point)

    Fascinating post by the way!

    1. Waldbaer

      Should be the mid-distance. The Dorico Screenshot shows this, too, if you look closely. If not, every other measurement would not make sense since e.g. the staff height would have to take into account the line thickness and so on…

      1. Philip Rothman

        Yes. Midpoint to midpoint. Imagine a hypothetical staff line of thickness 0.

        1. mirabilos

          Hm, that throws off my calculations, I was assuming internal. (But that makes for a difference of only 0.08 so…)

          Gould’s picture is not clear on whether it’s midpoint or interior.

          “If not, every other measurement would not make sense” is not true, because you can easily calculate based on interior (if needed at all, because you set the staff size instead of the spatium directly like MuseScore does): 4*1sp + 5*(line height in sp) = x mm is trivially solved.

          That being said, temporarily raising the staff line size to 0.3 sp shows that MuseScore follows the midpoint model. Ouch, and good to know (so *4 instead of *4.4 in the other calculations I did).

  5. francois grillot

    What is the distance in between 2 lines of a staff is slightly smaller the size of a note, so by definition determine the size of the note!

    1. Waldbaer

      I hope that I understood your questions correctly:
      Distance between 2 lines = 1space – line thickness

      The size of the note depends on the font and is mostly a little bigger than 1space. But even the noteheads of quarter, half and whole note may differ quite considerably.

  6. Bob Zawalich

    Very nicely done, Philip. It is no mean feat to describe such intricate details without inducing confusion or sleepiness, especially across different notation programs.

    I have had to deal with these issues quite a bit when writing Sibelius plugins. Here are a few more details specific to Sibelius.

    1. The height of a space is defined exactly as the staff size divided by 4. Changing staff line widths does not affect the space size. Try setting the staff line width to .5 spaces in Engraving Rules for a demonstration of this. For a 7 mm staff, a space is 1.75 mm (7/4), but for a 10 mm staff, a space is 2.5 mm (10/4).

    2. Sibelius converts all “fixed size” measurements, such as point sizes for fonts, to units of 1/32 of a space when storing measurements internally. It converts them back to the current unit chosen in Document Setup when it displays these to you, and some rounding can take place. This is why you will sometimes enter a font as 12 points and see it displayed as 11.9 points. Doing this allows Sibelius to easily display different values when you change the unit from mm to inches, for example.

    3. I wrote a plugin, Convert Spaces To Units, that will convert inches, millimeters, or points to and from spaces, based on the current staff size. I discussed it in a thread in the Sibelius Technical Support forum: http://www.sibelius.com/cgi-bin/helpcenter/chat/chat.pl?com=thread&start=470184&groupid=3&&guest=1, and there is a Scoring Notes tutorial that describes a use for it at https://www.scoringnotes.com/tutorials/using-font-sizes-in-wildcards/.

    4. The physical width of a bar can change when you drag the barline or change the number of bars in a system, and the positioning of notes, lines, text, and other objects uses a different measurement system altogether, based on the rhythmic position within a bar, so the spacing adjusts automatically when the bar width changes. In Sibelius there are 256 rhythmic units in a quarter note, so a 4/4 bar is 1024 units long. If you have 4 quarter notes in a bar, they will be at rhythmic positions 0, 256, 512, and 768 respectively, all multiples of 256. If you adjust the X offset for a note in the Inspector/Property Window, the rhythmic position does not change, but the note is shifted by a number of spaces. That distance remains the same even when the width of a bar changes, and the distance between the notes changes.

    1. Philip Rothman

      Thanks, Bob, for this additional information, which is very useful. I was almost induced to confusion myself while writing this :-)

    2. Tiago

      Nice article Philip,

      Bob Zawalich’s point #2 might explain why Sibelius skips from 0.1 to 0.06 in the staff line width. This always intrigued me. In some situations 0.1 seems a bit thick but 0.06 is dangerously thin.

  7. Don Stewart

    PROGRESS.
    (All things that Leland addressed in his earlier versions of Score

Leave a Comment

Your email address will not be published. Required fields are marked *