imaginary bar line
Is there an automatic feature so that notes are split at the imaginary bar line that divides a measure in half so that the third beat of 4/4 is clearly visible?
For example, dotted quarter followed by two tied eighth notes rather than a quarter note.
If I have not expressed myself clearly I found an explanation of the rule at
http://www.musicarrangerspage.com/tag/imaginary-bar-line/
Thanks
Comments
This request goes back many years.
Apparently, in European classical music, the midpoint of the bar is not automatically noted.
I have, for years, had to "fix" this. Meh.
:)
In reply to This request goes back many by xavierjazz
You are right about that. I grew up in Europe, have played the violin since I was a boy and I had never even heard of this rule before reading this thread.
Reading at the link I find some of what is said convincing. But quite a few of the examples don't look any easier to read when following the rule. Sometimes the opposite. Syncopated notes in particular signal "syncopated" much stronger when they are one note rather than two tied half size notes. But it depends on context which version is more readable and maybe one ought to do this flexibly.
I am not sure I understand. I mean, I am of course very familiar with the idea of splitting a note to not cross that imaginary bar line, but MuseScore in no way prevents you from doing this. That is, if you want a dotted quarter followed by two tied eights, simply enter it that way - don't enter it incorrectly and then hope MsueScore will correct your mistake for you.
I guess you could try entering your music in 2/4 time then changing the time signature to 4/4, but you won't be able to get whole notes or dotted halves.
I see what you are saying. Is it possible to make this automatic? Especially for a user who does not know the rule or does not want to remember to enter it manually each time, it would be convenient to have the program take care of this detail.
In reply to I see what you are saying. Is by musescoreuser25
No, there is no way to make this automatic. It's not out of the question that a mode like this would be added some day - feel free to submit an official feature request to the issue tracker (see link in menu at right). But realistically, this but there are tons of rules in writing music and MuseScore can't be expected to correct them all for you.
I suppose you could try exporting a score to MIDI then re-importing it - the MIDI import does try to make reaosnable choices. But you'll probably run against other issues, like the fact that MIDI has no way of knowing the differences between C# and Db - another musical rule you pretty much have to know for yourself (when it is appropriate to use one versus the other) because the software cannot always guess right.
In reply to No, there is no way to make by Marc Sabatella
I understand. I'll make the request.
I do not have much experience with Sibelius or Finale, so I don't know if they make those corrections. But I just tried it in Band in a Box, which is not primarily a notation program, and it made the correction seamlessly. I think it also has options for how to decide whether a note is a C# or Db, such as "use chord scale for enharmonics".
Thanks for your reply.
In reply to I understand. I'll make the by musescoreuser25
Yes, MuseScore will guess based on context too. But it's just a guess, and is pretty much guaranteed to be wrong much of the time. There is no substitute for actually knowing a thing or two about music, to know when a chromatic alteration based primarily on the key versus on the chord versus on the direction of the line versus the contour of the line versus readability etc.
I'm surprised BIAB automatically fixes rhythmic errors for you. Neither Sibelius nor Finale do. But it's not a particularly *difficult* thing for a computer to do - just would be highly annoying to most people because sometimes people *want* to break the rules for various reasons. So it would have to be an option, not the default. Maybe Sibelius has such an option buried somewhere, but I am pretty sure Finale didn't last time I used it.
In reply to Yes, MuseScore will guess by Marc Sabatella
Marc, I think you are underselling the quality of accidental-guessing that Musescore offers. I find it correct much more often than wrong, at least in 19th century classical music which is the center of my interest.
In reply to No, there is no way to make by Marc Sabatella
Actually it's already the case in master de8ea53700 (the future version 3 of MuseScore). Entering a dotted quarter + a quarter in 4/4 will result in a dotted quarter and a a 8th note tied to a 8th note. It's a byproduct of shoogle his work on MIDI input.
In reply to Actually it's already the by [DELETED] 5
That's pretty cool, but - I wonder if maybe this should *only* be for MIDI input, or at least be possible to disable this? There are any number of situations where one wants/needs to break this. Eg, highly syncopated genres typically allow quarter notes across that imaginary barline if the whole measure consists of offbeat quarter notes.
In reply to That's pretty cool, but - I by Marc Sabatella
An option is not an option...
Of course, it's possible to override it.
The previous behavior was a pain if the user wanted the split to happen automatically but it has the advantage to be consistent with what the user choose (you want a quarter, here is a quarter).
This behavior, the one in master, is a pain if the user doesn't want the split (I feel it will be less often but I can be wrong) and it's not consistent.
I don't have a strong preference for any approach. The current behavior does make the code simpler... A different behavior for MIDI would make the code a bit more complex.
In reply to An option is not an by [DELETED] 5
I can live with it since as you mentioned you can override it after the fact, but I can imagine some complaints down the road. The rules for time signatures other than 4/4 are nowhere near so clear, and I think I will find this automatic correction to need correcting as often than not. I would still advocate this be something that can be turned off.
In reply to I can live with it since as by Marc Sabatella
I programmed the rhythmic groupings for notes and rests according to the rules set out in Elaine Gould's "Behind Bars". If anyone wants to see what that means for each time signature then have a look at the examples in mtest/libmscore/rhythmicGrouping (see the
*-ref.mscxfiles). I've attached the 4/4 reference as a PDF, along with a file that shows the cases where the rules conflict.When deciding whether to split a note or rest the algorithm currently only looks at the note or rest in question. Solving the conflicting cases requires looking at the surrounding content within the bar. (For example, the rule "always tie across the middle beat, except for syncopation" requires looking elsewhere in the bar and not just at the note being tied.) Where the rules conflict I tried to satisfy the most common case.
As @Marc pointed out, rules for time signatures other than 4/4 can be quite complicated. However, the rules I programmed are completely general, because they decide whether to tie or cross the beat depending on whether the beat is stressed (important) or non-stressed (ordinary), rather than hard-coded rules like "Thou shalt tie notes across the middle beat in 4/4". All that's missing is a way for the user to define which beats are stressed and which are not. This could be represented internally by an array within the TimeSigFrac class. 4 would be stored as [2,2] by default. 5 could be [2,3] or [3,2]. Calling TimeSigFrac::numerator() would simply add up the values in the array.
In the future, the beaming and grouping rules should both be made to use the BeatType mechanism. There could even be an option to specify global rules for beaming and grouping that would affect all time signatures, not only the current one as is the case at present.
In reply to I programmed the rhythmic by shoogle
What wonderful work!
My only question would be about N = 4, Bar 4, 2nd stave.
I would think it should be 2 tied quarters.
Thanks so much.
In reply to What wonderful work! My only by xavierjazz
Thanks for the kind words!
Your suggestion seems like a perfectly valid form of notation, but I would need to see a quote from a reliable source before I would consider changing the current behaviour. As I said, my groupings follow the styles set out in "Behind Bars". Other sources might give different recommendations because groupings are largely a matter of taste and style.
For anyone wishing to make changes, the rules are all below line 491 in libmscore/durationtype.cpp. The rule @xavierjazz refers to is here, and it basically says "notes that begin and end on a beat (stressed or unstressed) are allowed to cross a stressed beat without a tie". It would be relatively simple to make that rule (and all the other rules) configurable via style options if this was a common request. However, for the sake of consistency I think the "Behind Bars" groupings should remain the default.
In reply to Thanks for the kind by shoogle
The rules are pretty subtle and largely subjective even for 4/4, but indeed, I don't think any authoritative source would recommend a half note not be allowed on beat 2 of a 4/4 measure, particulartly if beats 1 & 3 are not subdivided. I've never seen published music notate this using tied quarter notes. However, the same cannot be said for other time signatures, where often such ties *are* used even when not strictly necessary, on the theory that showing the divisions is that much more important in less familiar time signatures.
In reply to The rules are pretty subtle by Marc Sabatella
@Marc, indeed! See line 666 ;)
In reply to I programmed the rhythmic by shoogle
Oooh, I do like the idea of allowing customization of the beat groupings. Still a bit uncomfortable with MuseScore not simply doing what I tell it to, but it's a nice thing for some people to be sure.