Page 1 of 1

Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 10:57 am
by stubbsonic
I'm advising the developer of a popular and incredibly capabile music app for Android. He has enthusiastically agreed to add some new step-rate/beat-divisions to his range of apps. They will now divide beats by 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 16 steps per beat. Incidentally, his apps let you select individual steps per beat to apply swing to, and in another section, you can delay any steps by "micro steps". It's very flexible!

He has insisted on keeping with some traditional time-signatures and note-values. His software takes a user selected time-signature and creates pattern-length and step-rate settings that match that intent. The software also displays a time-signature and does some calculations based on it.

There are a couple hazy areas needing clarification. The main controversy is around relating the beats (as defined by the time-signature and note values) to the beats as defined by the tempo in BPM. I wanted your opinions.

1. If a time-signature has an 8 in the denominator, there are multiple "rules/customs" about what is defined as a beat. What do you think about these statements?
-- a. with 3/8, 6/8, 9/8, 12/8, 15/8 the dotted-quarter customarily gets the beat, yes?
would you consider dividing the beat into 3 steps to be standard? or 6 steps?
-- b. with 5/8, 7/8 the quarter-note probably defines the beat w.r.t. tempo, yes?
-- c. with 2/8, 4/8, 8/8 is the beat really an 8th note? Or does the quarter-note still define the beat for tempo purposes?

2. With 2/2, 3/2, 4/2, we consider a half-note to be a beat in most circumstances. But for the purposes of a step sequencer, we could just substitute quarter-time equivalents, yes?
For example, 2/2 = 2/4 (4 steps per beat), 3/2 = 3/4 (4 steps per beat), etc.

I'd really appreciate your opinions on this. Thanks.

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 1:39 pm
by MIDI Life Crisis
If I'm understanding your question, it depends on the feel of the music. For example, 6/8 could be either two dotted quarters, 6 individual eighths, or even three quarter notes. I would suggest your friend/collaborator/??? snag a demo copy of Finale and take a look at the meter options. If, for example, I specify two dotted quarters, the phrasing will default to two sets of three eighths (IE - dividing the measure in half with three eights as the unit in each half - more of a "2" feeling). If I specify three quarters, it will default to thee sets of two either, giving more of a 3/4 feeling.

This is probably more important for print and how the metronome ticks accents than simple playback, which only really needs the number of pulses per measure. The bottom line is the pulse and feeling desired. In the above examples of 6/8, are we thinking in two, three, or six?

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 1:59 pm
by mikehalloran
I agree on Finale.

The two main users boards have people who are more than happy to split hairs and give explanations on these matters, sometime citing Elaine Gould who has pretty much become the Last Word on issues like these.
https://www.amazon.com/Behind-Bars-Defi ... 0571514561

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 2:32 pm
by stubbsonic
Thanks, fellers.

What the developer is trying to do is find the "most common case" for each of these time-signatures. So, if a person selects one of these with a 2 or an 8 in the denominator, is there a consensus about what might most commonly be expected, both in terms of what beat is felt, and also, how the beat might be divided?

In any case, once the software makes settings, the user can override them. But he is trying to make these initial decisions in a manner that is typical and intuitive. I also think he'd like to fix this once and not have to keep revisiting it.

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 2:35 pm
by MIDI Life Crisis
You might also post on Facebook. There are a few groups there. Music engraving tips group will get you a few dozen answers (all conflicting, of course).

As far as most common, that’s a fool’s errand. Then again, they are using Androids so it may not even matter... lol

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 3:23 pm
by stubbsonic
MIDI Life Crisis wrote: As far as most common, that’s a fool’s errand. Then again, they are using Androids so it may not even matter... lol
Hee hee! Yea, I have an iPad, but use an Android for my phone.

This would all be so much easier if he avoided traditional time-signature and note-values, and just went with:
step-rate (steps-per-beat),
track-length (steps-per-track-pattern)

But he's definitely going with the traditional terms. So, I think I'll just settle on this:

If there is an 8 in the denominator:
1. if the numerator is a multiple of 3, the dotted quarter is the beat, and there are three steps per beat.
2. If the numerator is 2, 4, 5, 7, 8, treat the quarter note as the beat, with four steps per beat.

If there is a 2 in the denominator, change it to a 4, and keep the numerator (with four steps per beat) so:
1/2 = 1/4
2/2 = 2/4
3/2 = 3/4
etc.

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 3:34 pm
by MIDI Life Crisis
stubbsonic wrote:...he's definitely going with the traditional terms.
I guess that then depends on what you consider traditional. It's difficult for me, as a composer and initially a drummer, to think about meter as being so narrowly defined. But I guess he's thinking about the "average" Joe and Jane end user who doesn't't know or care, in which case it really doesn't matter, does it? :rofl: .

Re: Music Theory: Time-Signature questions

Posted: Thu Aug 29, 2019 8:17 pm
by stubbsonic
As we discussed in another thread, the time-signature is often "abused" in modern tech.

Developers with half an understanding about the terminology (not that I claim much expertise) often make things MUCH more confusing than necessary.

Fortunately, this developer has shown sincere curiosity, AND perhaps even more admirable, his software provides much more flexibility than the conventions he uses would suggest.

Re: Music Theory: Time-Signature questions

Posted: Fri Aug 30, 2019 7:43 am
by stubbsonic
BTW, the apps are G-Stomper Producer, G-Stomper Studio, and G-Stomper Rhythm, by planet-h.com

https://www.planet-h.com

Re: Music Theory: Time-Signature questions

Posted: Fri Aug 30, 2019 4:34 pm
by Prime Mover
stubbsonic wrote:What the developer is trying to do is find the "most common case" for each of these time-signatures. So, if a person selects one of these with a 2 or an 8 in the denominator, is there a consensus about what might most commonly be expected, both in terms of what beat is felt, and also, how the beat might be divided?
There is typically a "most common case" for specific time signatures, but definitely not for the DENOMINATOR as a whole. 6/8 and 7/8, for instance, are completely different animals, the former most likely having a dotted-quarter pulse, and the latter having a standard quarter-note pulse. I think the developer would be best served by:

A) Either allowing for another option that allows you to select "what note gets the beat"

B) At the very least, evaluate the signature as a whole on a case-by-case basis, and not just the numerator. In fact, almost all commonly-used signatures trend toward quarter-note beats with the exception of 6/8, 9/8, and 12/8. He could put in a simple line that says:

if the denominator = 8, AND the numerator is divisible by 3, then make the pulse a dotted-quarter.

It wouldn't be perfect, but that would cover the most bases. 6/8 is SOMETIMES a quarter-note pulse, but realistically, not that often. Unless you give the user an option to change the pulse on their own, this is really the only option.

Re: Music Theory: Time-Signature questions

Posted: Fri Aug 30, 2019 7:27 pm
by stubbsonic
Prime Mover wrote: There is typically a "most common case" for specific time signatures, but definitely not for the DENOMINATOR as a whole.
Yes, we're on the same page, Prime Mover.

I was really trying to discourage him from using time-signatures at all, but apparently his user base is pretty large and runs the gamut of experience, backgrounds and expectations. He is convinced that getting rid of time-signatures would cost him in some ways.

At first, he wanted it to follow a mathematical formula, but I was able to convince him that this wasn't possible. So now he has to account for lots of factors.

On one hand, there's a Time-Signature Wizard, where you pick a time-sig and it pre-selects various step-rate and step-number functions. Here's where it gets messy: if the user picks a step-rate, and track/pattern length, it will assign a time-sig from that. The good news is you can set it however you want. The only downside is sometimes the time-sig is unexpected.

He seems open and capable of fixing it so it is at least intuitive.

Re: Music Theory: Time-Signature questions

Posted: Fri Aug 30, 2019 10:37 pm
by mikehalloran
MIDI Life Crisis wrote:I guess that then depends on what you consider traditional. It's difficult for me, as a composer and initially a drummer, to think about meter as being so narrowly defined. But I guess he's thinking about the "average" Joe and Jane end user who doesn't't know or care, in which case it really doesn't matter, does it? :rofl: .
As someone who hung backstage with folk musicians and dancers in Europe during the summer then played in a Persian wedding band on Sundays, getting back to the rigid concept of time signature and proper rhythmic division takes some doing. Having had Lou Harrison as a teacher at San Jose State doesn't help.

Even arranging for the church ensembles that I do so much of these days, I have to keep thinking, Well, I can read it but can they?

My daughter worked with the great Alice Parker (composer and arranger for the Robert Shaw Chorale—still working in her mid-'90s) a few years ago. Ms Parker and Marjorie each gave a talk on writing and arranging for singers. Alice's was on teaching composition majors that their works have to be singable to get performed. My daughter's talk was on how difficult a lesson that can be to learn after one gets the sheepskin.

Image

Re: Music Theory: Time-Signature questions

Posted: Sat Aug 31, 2019 5:39 pm
by MIDI Life Crisis
My mom sang with Robert Shaw. I think she said they only sang in 3, 4, and 6. Lol Of course the solution to the problem at hand is simple. Just allow any number of beats with whole to 16th notes and selectable accents to account for phrasing. Is it all that difficult to program? If so, has the guy tried flipping burgers for a living? :rofl:

Re: Music Theory: Time-Signature questions

Posted: Sun Sep 01, 2019 9:55 am
by stubbsonic
MIDI Life Crisis wrote:Of course the solution to the problem at hand is simple. Just allow any number of beats with whole to 16th notes and selectable accents to account for phrasing. Is it all that difficult to program? If so, has the guy tried flipping burgers for a living? :rofl:
First of all, it may be simple to you, but for us non-geniuses.... :dance:

There are two separate aspects.

First, is the actual functionality (without regard to applied terminology)- which involves step-rate, steps per bar, and number of bars. He provides 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 16 steps per beat; 1-16 steps per bar, and 1-8 bars in length. Select individual sub-divisions for adjustable swing, and per step micro-step-shift. And, all tracks can have their own settings (they share tempo). Tempos from 20-999 BPM. It isn't limitless, but it combines more options than any mobile step-editor I've used.

The second aspect is just how the terminology is thrown around; i.e., time-signatures, note/step values. The ambiguities of the time-signature have to be taken into account. This is a tricky aspect, and very few developers get this right, much less give it a second thought. This guy cares enough to try to get it right (as right as possible/practical).

Re: Music Theory: Time-Signature questions

Posted: Sun Sep 01, 2019 11:27 am
by MIDI Life Crisis
So maybe our technology is not up to the task of defining music as well as a paper and pen can. Lol