Level again
I did some reading on this site and found a way to change instruments via the mixer. Great I thought played around a few different voices, found one I liked for this project. But...when compared to what I had the audio level of the new voice, it was about 25% lower (quieter) All the level controls I could find within MuseScore are maxxed out. Is the nature of the beast? or is there a way to crank it more?
Comments
I believe that the standard SoundFont in MuseScore is a little on the quiet side.
Of course using reverse logic - if it's too quiet then you're too young - lol.
In reply to The TimGM6MB Soundfont is a bit quiet. by ChurchOrganist
Now THAT (can you hear me?) is funny.
Regards,
As far as I know, the mixer only lets you change sounds *within* a given soundfont. So changing sounds there shouldn't affect overall volume, unless it just so happens that the sound you changed to is quieter than the sound you started with in that soundfont. That's assuming you have the mixer levels set the same. As for overall sound, you don't say where you looked, but the master level for the soundfont is in the synthesizer window.
In reply to As far as I know, the mixer by Marc Sabatella
Don't forget that there is more than one way of controlling volume.
In GM there are three ways to affect it - Velocity, Expression and Volume.
I'm not sure how General MIDI compliant MuseScore is, but I'm guessing that the note you enter on the score is not at full velocity (127) which is why it sounds quieter on playback.
In order to adjust this you need to open the Note Properties dialogue by right-clicking selected note(s).
Just checked one of my scores and the default velocity setting is 80, which is about right for mf.
Not sure how you change the default but you can attach an expression marking to a stave to make it louder.
HTH
Michael
In reply to Could be the velocity values by ChurchOrganist
The velocity setting is probably the culprit, the folks that designed the software, try to keep the level so it doesn't "clip" and distort, so I will try the velocity setting.
In reply to The velocity setting is by earthshaker
It's nothing to do with clipping and distortion, it's just a (fairly reasonable) assumption that you want your voices played mezzoforte if you don't specify a dynamic, which as I said normaly equates to Velocity=80 in notation programs.
If you want everything triple forte then stick the instruction at the beginning of the score :)
In reply to They assume that you want mf by ChurchOrganist
Interesting observation! I hadn't thought along these lines before, but now it makes sense. You definitely don't want notes entered at maximum volume by default, because there would be nowhere to go up to from there. You might be thinking, but songs I play from MP3 files or CD or whatever don't start out at maximum volume either, so why should MuseScore be quieter than those? Well, most recorded music is heavily "compressed", by which I don't mean the thing MP3 does to save space on disk, but rather, something done to the audio to artificially pumped up the volume of qui. This is so the track doesn't seem softer than other tracks when played on the radio. Most people are not accustomed to haring uncompressed music coming out of their computer speakers, but that's pretty much exactly what MuseScore produces.
So an option to compress the output of MuseScore could be useful at times. I don't think just control over the mapping from dynamic markings to velocity values is enough, BTW - a compression facility should also affect velocity overrides entered via Note Properties, for instance. This doesn't strike me as anything all that important to provide, but it also doesn't strike me as all that difficult to implement.
In reply to Compression by Marc Sabatella
MuseScore audio export (not playback) is doing a normalization to avoid distortion. That's why there are two passes when you export to Wave, the scroll bar moves twice. The first one is for finding the higher level and the second one to normalize the audio.
When we talk about velocity, we are talking about MIDI values, there is no sound yet. When you change dynamics in the score or note properties -> velocity, you change a relative value between 0 and 127. So you don't input notes with a volume, just a midi velocity. The volume can be changed in the play panel. The velocity values are then transmitted to the synthesizer, and you can change the gain of the synth in Display -> Synthesizer. It's easy to get saturation here, especially if the soundfont samples are already high. That's why the audio export do a normalization.
In reply to MuseScore audio export (not by [DELETED] 5
Right, but I'm talking about perhaps doing a "MIDI compression" as a separate thing from the audio normalization done on export (didn't realize that was happening, but glad to hear it!). Basically, just a simple re-mapping of the MIDI values, that would affect both playback in the program as well as export. Before passing a note to to the synthesizer, its velocity value would be increased according to some simple scaling function, so that instead of a range of 1-127, the actual MIDI values passed to the synthesizer stay in a more restrict range. Even a simple linear compression like N / 2 + 64 might be useful this would map 1-127 into 65-127. Keep 0 as 0 so silent notes remain silent. Seems to me this successfully compress the dynamic range, handling both dynamics and note property overrides.
Anyhow, just a thought for a future "hack day" type of project.