Ties and Accents

• Sep 13, 2020 - 12:26

I had to type this: slurp.png
but in order for the whole duration to sound in the same level I had to add invisible accents to every other measure...
Any way to avoid that? Or I'll just have to... deal with it?

Attachment Size
slurp.png 14.58 KB


In 3.5 there is a bug. The accent is supposed to play for a moment then return to piano volume. The bug is that the accent applies to the entire first note of the tie. If you want an entire tied note louder, don't use an accent, use a louder dynamic (or increase the velocity on the p in the inspector).

In reply to by mike320

Wait you mean accent originally worked as decrescendo for notes individually? Well in a sense that is accent...

In the piece I was trying to transcribe, I think accents meant to give a bit bigger "force" (basically accent a little) but then do some sort of fade out... "fade out".

In reply to by Iothes

For example I was looking at this measure:
Only the first 3 "chords" are to be played "accented" while the rest have staccati/staccatisimi. Meanwhile the composer in his recording of the work seems to conduct all 5 chords in the same level...
https://youtu.be/bb8QrH1XVZQ?list=OLAK5uy_kOyWUO0cSNM4k3wN1L7bL3dT1Nkti… (o:o8)

So accents just act like decrescendos but for each note individually? ( I mean the symbol looks like a decrescendo anyway).

Attachment Size
accenti.png 56.89 KB

In reply to by Iothes

Playback is a computer representation of what is written in the score. It has an algorithm for a common playback of each symbol. What you describe is not what I would expect based upon what I see.

The sforzando accents are normally a harder or louder attack followed by an immediate subito change to the written dynamic. Every notation is interpreted by humans based on context that may vary from this basic meaning of the accent.

Many versions of Beethovens 6th sound like they put fermatas all over the first few measures but Beethoven didn't write them. His tempo mark is clearly dotted half note = 60 and not Rubato. Are the conductors wrong or do they simply interpret what's on the paper differently? Computers have to be programmed to make an interpretation that will probably never be heard in a live performance because each defined accent is played the same way. We do have the piano roll editor if you want to change this along with other methods to change playback.

In reply to by mike320

I wonder... after analyzing how Stravinsky uses it... I think "his accents" mean to NOT play clearly louder but instead they mean a decrescendo for the note that includes the articulation. Basically the accent is like a "mini decrescendo" after which it's like nothing happened. Could I reproduce that in Musescore somehow?

In reply to by Iothes

Accents on longer notes like this usually mean to attack the note louder then return to the prevailing dynamic. So I think this is what you are describing. I think there was a 3.5 but that had accents sustain the louder note when it had a tie and it will be fixed in 3.5.2 which should be released very soon.

For the decrescendo to total silence, you can either put a ppp dynamic (made invisible) after the decrescendo or set the velocity change to the same number as the last dynamic. Neither will actually ever get to total silence in MuseScore but you won't hear what's left playing unless you have your volume turned up extremely high.

In reply to by mike320

It's not THAT which I "exactly" describe: Everything is correct but in stravinsky they are not meant to attack the notes louder, but just do a sharp decrescendo. If it is an "attack" that would be a small one. Because the notes after the accent, the ones marked with staccatissimo, are meant to be played on the same level.. I think. At least that's what his recordings show me. But besides that, would be weird to play those chords, without accents, lower, in volume, than the ones with the accents on them.

Also the idea with the ppp marking wouldn't work in the above image... so maybe the roll editor could do(?)

What I mean is that accents aren't like marcatos but they are just rapid decrescendos, that apply only to the note we apply it to, and the next notes aren't affected by it.

In reply to by mike320

Ye I also try to not use piano roll editor... It's slow when you wanna change a lot of stuff, which might be a lot in my case. I believe Musescore should be made like so everything that the roll editor does can be done through the inspector, like when selecting similar elements for example. We could also have "custom items" that are basically items taken from the palette and have their settings tweaked from the inspector and then saved.

In reply to by Iothes

I find it difficult to define an accent as a mini decrescendo. To me, an accent has more to do with how the note is initiated rather than what happens to it after it starts. A staccato has more to do with what happens after the note starts. How to play the two is different for each class of instruments. In a large orchestra playing ff and fff, I'm not at all surprised that the two markings sound similar. I wonder if it is even possible to have a "one size fits all" approach to playback because of the difference in solo vs ensemble vs type of instrument matrix.

In reply to by bobjp

Good points Bob. MuseScore has to pick a reasonable common playback as the default. There are tools now, like the PRE and maybe BSG's plugin that will help and hopefully the integrated DAW in version 4 will make it even easier.

In reply to by Iothes

Exactly. Look at your score. He has given each group of instruments a different way to play those 5 chords based on the final effect he is after. The strings have all ff down bows. The brass have f to ff with staccato on the loudest three. WWs have the shortest notes of all. I imagine that in a live concert this is amazing. A recording can't do it justice. I think I would try to go more by what the score is saying, rather than what a recording suggests.
But, this is why a sound library for a DAW can cost a thousand dollars. It has all the articulations at different volume levels. As well as recording at different distances from the mic to simulate depth. As well as hundreds of other quality aspects.

In reply to by bobjp

I think that basically everyone plays the same stuff, in those 2 measures, except for the strings which don't have crescendo for some reason. BASICALLY (x2) everyone plays the first 2 chords with accent and the next 3 staccato. But I really think there that the accents may mean a louder attack, but also mean sharp/rapid decrescendo, based on the way he usually conducts on areas which contain accents. I think I have linked his recording of this specific part above.

Sorry, I just haven't looked at what DAW is and does. It just lets you play stuff with many different ways?

Oh, just noticed what you meany by "not what the recording suggests". Yes in the recording you cannot hear any kind of crescendo. But you can hear how the accents are performed. The basic idea anyway.

In reply to by Iothes

The strings are already at ff. Everyone else gets there later.
An accent does not imply a louder attack followed by an abrupt decrescendo. An accent is a strong, crisp attack followed by a resumption of whatever dynamic is going on. There is a big difference. An accent denotes presence, not just volume. How you end the note before the accent matters also. In this case the strings are marked all down bows. It takes time to lift the bow off the strings move it to the frog, then dig in for the accent. Followed by a slight relaxing of pressure on the bow. Wind players do it with a hard tonguing technique along with additional breath support. It's more about power than volume alone.
A DAW allows you to use far better sounds and control them far more minutely than notation software.

In reply to by bobjp

Maybe the way I describe it is just my imagination... Or maybe the idea that accents= attack followed by decrescendo is an illusion if that's what you mean... But for some reasonI hear it that way in many recordings... You mean that if I go to a live concert it will sound different? We shouldnt replicate this.. decrease in volume?

By the way I'm not sure why he wrote those bars like so... In all the recordings the recordings, recent of not, all 5 chords, seem to be played in the same level. Well, this revision has only be recorded 5 times, as we know so far.

Notice though, that wws start with ff and brass start with f because they are considered more loud than the rest... Kinda bugs me that strings don't, also, have crescendo lines... :/

Okay this might be an exception about accents (although I do think the symbol symbolises a drop in volume, like decrescendo). If I'm wrong guess ill just leave it to the experts.

In reply to by Iothes

Of course, a live concert will sound different than a recording. And it will sound different depending on where you are sitting.
Software should be able to replicate an accent. The punch of the attack. But not a decrease in volume. Because it's not a decrease, but a resumption.
Indeed, there seems to be many odd things about the markings for these two measures. We wonder if he did this or some engraver. Yes, the WW's are blasting away at ff. Then they crescendo playing extra short notes. My guess is that they have to play louder just so some sound can be heard at all. The brass are at f. Some, but not all, crescendo. But to what level? Something in the lower brass is muted. And the strings just blast away the whole time. I suspect that anything louder than ff is just a suggestion. Sure, you can write fff. But if you are already at ff....that's already really loud. No wonder everything sounds alike.
So we have to ask ourselves this: did Stravinsky follow his own markings? Is that even possible? Or as he heard it, did he do something he liked better? To me, recordings are not the end all and be all. It's too easy to fixate on them and forget that this kind of music was meant to be in the concert hall.

Should I make a new post about using inspector as an alternative to roll editor and the feature about "custom items"?

In reply to by Iothes

If I understand correctly, you're wanting the PRE to make it easier to adjust several accents at once as an alternative to using the inspector. I would consider this a suggestions, accents work the way they normally do in the current releases. One thing I didn't think about is the plugin at https://musescore.org/en/project/articulation-and-ornamentation-control. I've never used it but it might do what you want.

In reply to by Iothes

I thought of one more thing. If you really think the accents should act like a decrescendo, then just use a decrescendo and adjust it so it looks like an accent. You will need to do this one at a time since it's impossible to adjust the display several decrescendos at once and the display doesn't survive being put in a palette.

What would survive is copy and paste and note/chord. You can then change the pitch of the notes/chords after you paste.

Do you still have an unanswered question? Please log in first to post your question.