String section dynamics

[thorrild]

Try this:
- Create a partial string section score: Violin I, Violin II, Viola and Cello.
- Enter staccato eighth notes, middle C, to fill 10 measures in Violin I
- Assign 10 different dynamics, one in each measure: e.g., pppp, ppp, pp, p, mp, mf, f, ff, fff, ffff
- Play it back. It should sound as expected.
- Cut the 10 bars of music from Violin I and copy it into Violin II.
- Play it back. All bars sound the same: A strong ff dynamic.
- Cut the music and paste it into the Viola staff.
- Play it back. Everything sounds good, but the sound changes drastically when going from f to ff.
- Cut the music and paste it into the Cello staff.
- Play it back. Same odd timbral change as in the Viola sample.

Am I doing something wrong, or are these problems that others can confirm:
- Violin II Section does not respond to dynamic markings when staccato is in effect.
- Viola and Cello Section have a drastic timbral shift between f and ff (also staccato).

If somebody can confirm this, I'll write it up as a bug. If not, then please tell me what I am missing.

Best wishes,
Thorrild

[pcartwright]

- Violin II Section does not respond to dynamic markings when staccato is in effect.

All the string parts seem to have a fairly consistent change in dynamics from what I could hear (at least from part to part). However, the change in dynamics for each part doesn't feel linear to me, more like a cosine curve or 'S' curve.

- Viola and Cello Section have a drastic timbral shift between f and ff (also staccato).

Referring to my above comment, I think the curve that I hear would explain this. If velocity is responding along a cosine-like curve, there will be dynamic markings in the middle of the dynamic spectrum that change more drastically than dynamic markings near the ends of the spectrum.

I'm not sure if this is really a bug or not. I should also say that my observations could entirely be a side effect of my listening environment or my ears as opposed to Notion's actual output.

Try uploading a file that has the issues you describe so we can look at it more carefully.

[thorrild]

As usual, this was mostly pilot error.

Today I recreated my own test file as outlined in my original post, and the dynamics in Violin II played back just fine. I also went back to the original file, where the string section dynamics had first bothered me, copied and pasted the offending string parts into a String Orchestra template file, and discovered that everything played back fine there as well.

I then remembered that I had had problems with unresponsive dynamics from time to time in previous files, and it turned out that if I added additional dynamics on the opposite side of the staff, and then hid them, the file began to play back as intended.

So:

One problem was that dynamics sometimes are not read by Notion unless they are above or below the staff, as the case may be.

The other problem — the annoying shift in timbre between f and ff (for Viola, Cello and Bass sections) and between mf and f (for Violin section) remains, and is, I assume, just a really bad joint between two samples. I've read through pcartwright's elaborate explanation of linear, cosine and s-curves, but that's just a little to technical for me. All I know is that a staccato note played by a Violin section at mf sounds like a completely different instrument if played at f, and that similar ugliness exists in the rest of the String Section sound set, as explained above.

[pcartwright]

if I added additional dynamics on the opposite side of the staff, and then hid them, the file began to play back as intended.

What happens when you add a dynamic directly onto the staff (not really above or below, but actually click on the note where the dynamic is to begin) and then manually adjust the position of the dynamic to taste (above or below the staff)? Try that and see if you get better results.

The other problem — the annoying shift in timbre between f and ff (for Viola, Cello and Bass sections) and between mf and f (for Violin section) remains, and is, I assume, just a really bad joint between two samples. I've read through pcartwright's elaborate explanation of linear, cosine and s-curves, but that's just a little to technical for me. All I know is that a staccato note played by a Violin section at mf sounds like a completely different instrument if played at f, and that similar ugliness exists in the rest of the String Section sound set, as explained above.

Sorry to be too technical. You may be correct about the samples; the 'S' curve was just as I heard it (I didn't really hear a change in timbre, but that could be a difference in listening environments). You can't really change the timbre, but you might add a compressor to these tracks in order to make the volume changes a little more consistent.

Personally, I like the sampled string sections that come with Notion3 (especially the sound of the pizz.), but I’ve been able to get a little more depth and continuity out of them by adding strings (solo and ensemble) from other libraries “underneath” the main strings. By underneath, I mean to copy your given string part and paste it into a new staff using a different sample library.

I just copy and paste one part to another, but you may be able to send the MIDI information from one track to another (like assigning a sequencer staff to an instrument), but I’m not sure if the original staff would still playback or if the new staff would correctly interpret the various dynamics and articulations. I’ll have to try this out later and see if it works.

Anyway, adjust the volume of the new tracks in the mixer window to taste. You can easily hide the extra tracks when you print the score or the parts. I know this isn’t the easiest or simplest way to resolve the issue, but it may be the most convincing way to get a more consistent sound.