Inconsistent attack in higher pitches of viola
Musescore 2.0.2. When playing a fragment violin and cello give a normal playback, but viola plays froggy notes - see attachment. Oops, I can't download an mp3 file, so they are there:
//yadi.sk/d/aMMjy4QQpLqsc
//yadi.sk/d/_Ezsq1KIpLqqU
//yadi.sk/d/n8Q-521ZpLqsp
And I attach a .mscz fragment. Switch to viola to see the viola bug.
Attachment | Size |
---|---|
test_.mscz | 3.83 KB |
Comments
I wouln't call is corrupted, but it seems not as legato as it should be.
Here's a score for direct comparison.
Violin and Viola (and Violoncello) use a different GM Sound (Channel/Patch or whatever the correct name is), 40 vs. 41 (vs. 42), but the same for pizzicato and tremolo, 45 and 45 respectively
And the top 2 notes are out of the amateur range for Viola
In Mixer just change Viola to use the Violin sound, and they do sound identically.
Yes, I change Viola to Violin in the mixer to obtain a correct sound, but it should be fixed in the future, I suppose. Just to remind. )
I don't understand the problem. The viola samples used int he default soundfont do have a more crisp attack than the other strings; this is sometimes good, not as good other times. Such are the limiations with only having a single sample to represent all sorts of different types of notes an instrument can play. But I don't see a bug here. Am I missing something?
Well, seems viola has this issue only/mainly in the high pitches, those above amateur range
Yes, slurred notes should't be played with a crisp attack, there are other signs for that. May be it was a feature in 20 century, but now I consider it as a bug. An audible bug. Thrust a violin player. )
To be clear: again, there is only one sample provided by the soundfont per note (well, perhaps multiple ones for different dynamics), and that same sample has to be used whether slurred or not. So the fact that some instruments have too much attack for slurs is not a bug, it's just a current limitation of the playback system. Conversely, other instruments have too *little* attack for non-slurred notes. You're welcome to try a different soundfont that makes other choices.
Meanwhile there is already a feature request in the tracker regarding eventual support for slurred notes: see #9345: Playback of slurred notes. But it's not trivial to implement.
So again, aside from the fact that the particular samples used in the particular soundfont you are using happen to have a bit ore attack than one might like on slurred notes, is there an actual bug demonstrated here? I am thinking not, and it's really just a matter of wanting us to support playback of slurs.
Really, I don't undestand why "current limitation of playback system" concerns only viola voice, not violin (higher) neither cello (much lower). Bug or not, for a user that expects correct playback it is anormal, as well as to substitute viola voice by a violin one to have correct playback. I wouldn't discuss this philosophic question further, I posted my comment just to attract attention of MuseScore team. And as I expect they are clever enoughs to undestand the problem and to solve it, I hope that I did my best.
You're missunderstanding, playback of slured notes is currently not possible due to limitations of the playback system, this is a different, but related problem.
The problem at hand here is that the Viola sound in the default soundfont has some inconsistencies in the samples used for the various pithces (ort pitch ranges) in regards to attack
I am not sure if we are understanding each other. As I hear it, there is no "problem" here except that slurs have no effect, which is the subject of the feature request I referenced . For violins, it basically *always* sounds slurred on these notes, for viola it basically *never* does, but that' not a problem with the viola or with the violin.. There is simply no support for different sounds depending on slurs - whatever the soundfont provides is what we play, period. So unless someone can explain clearly to me exactly what the "problem" is perceived to be other than "we don't currently support different playback for slurred versus non-slurred notes", I would like to close this as a duplicate of the existing feature request.
No, to me it seems that the just in high pitches, abobe amateur range, viola sound significantly different (less slured if you want to call it that) than in lower pitches.
OK, but that's not a bug - that's just a question of how the samples in the soundfont were recorded? The person they sampled happened to use more attack on some notes than others. Human musicians do that. I don't hear anything that doesn't sound like normal human variation.
In any case, if there is any sort of issue here aside form the request to support slur playback better, then it is just particualr to the soundfont you are using, not anything about MuseScore code. Could be worth discussing in the Soundfonts forum, to see if there is any sort of consensus on ways that particular sound could be improved.
Automatically closed -- issue fixed for 2 weeks with no activity.
Automatically closed -- issue fixed for 2 weeks with no activity.
Automatically closed -- issue fixed for 2 weeks with no activity.
Automatically closed -- issue fixed for 2 weeks with no activity.
Automatically closed -- issue fixed for 2 weeks with no activity.
Automatically closed -- issue fixed for 2 weeks with no activity.
Automatically closed -- issue fixed for 2 weeks with no activity.