velocity - Inspector

Your ears are playing tricks on you Frans

Here is the velocity split map for the Trumpet in FluidR3Mono

You can see that there is a different filter cutoff point set for each velocity level sent to the patch, and also a different Envelope attack.

As currently the only means of changing volume with MuseScore is by changing velocity, so whether you are using dynamic markings, or controlling velocity from the inspector the Trumpet will respond in exactly the same way.

The only difference being that you have a finer degree of control over velocity from the Inspector, but if you need to you can control the velocity played by the dynamic marking by using the Offset value in the Inspector to fine-tune it to exactly the dynamic level you want.

This method enables your human players to understand your dynamic intentions, and you to be able to enter the velocities which you prefer when playing back the score.

I've altered the dynamic level of mp and p notes in this way in your test score and attached so you can see how its done.

HTH :)

PS With the velocities you are using there would be no discernable change in timbre as the first velocity split change happens at Velocity 81

Sorry, but you are are simply wrong that real instruments do not have different tone qualities as they are played louder. Almost evety single acoustic instruments changes quality as it is played louder - it's an unavoidable fact of how acoustic instruments produce sound. It's a very well-known phenomenon, and that is why many soundpnfts have "velocity splits" to simulate this - using different samples for different volume levels.

Now, ideally there would be a different sample at each and every velocity level level to capture the gradual transitions well, but that would make the soundfonts enormous. So indpstead they pick certain splits points where sound changes. And depending in the specific soundfont and the specific instrument, the sudden change might sound a little abrupt or artificial.

But if you are expecting no change at all, that is is simply *not* how real instruments work, and trying to circumvent this by not using dynamics at all, you are making a mistake - you are making your music sound *less*, not *more*, realistic to anyone familiar with the real instruments, and you are guaranteeing it will sound wrong when printed and played by real musicians.

And again, as I stated, the velocity splits happen at *exactly* the same place whether you you use dynamics or the Inspector to set the velocity. A trumpet note with velocity 25 sounds *exactly* the same whether that 25 camefrom a dynamic marking or from the Inspector. There is no possible way they could be different.

In your sample, the first line is played at velocity 80, because that is the default velocity for "mf" which is the default dynamic.

The second line is played at velocity 64, which is the default velocity for "mp" (you can see that by right clicking the "mp" and looking at the Inspector). I think it probably does sound slightly different than "mf" - what you descrbed as "muffled" but more accurately described as "less bright". That *is* what happens to real trumpets as they get softer. Maybe sometimes this effect is a little too exaggerated at these particular velocities, but that's the issue with having only a limited number of split points instead of different samples for every velocity - the transitions are not as gradual as they should be.

The third line is played at velocity 49, because that's the default velocity for "p" (you can see that by right clicking the "p"). This sounds different from "mp", and I would say, it sounds *wrong* - it has become brighter again. Perhaps the trumpet sample used at velocity 49 comes from a different source than the sample used at 64, or the sample settings need to be tweaked which is a shame, because it makes that transition sound more artificial. The sound should have have gotten darker still, but instead it got brighter.

Anyhow, in the Inspector section of your test file, you chose different velocities for some reaosn, so of course there is not an exact match. The second line that you set to 70 probably *does* use the same sample as the "mp" line that was at 64; it just sounds louder because for some reason you chose 70 rather than 64. Change it to 64 and you'll hear it sounds *exactly* the same as the "mp" line. Or, changing the dynamic of the "mp" itself to 70 using the Inspector and you'll hear that line now sounds the same as the other 70 line. Again, MuseScore doesn't care *how* you set the velcoity, but I *absolutely positively guarantee* that a trumpet at velocity 64 sounds the same regardless of whether you used a dynamic marking or the Inspector to set it to 64. Thjere is no possible way it could be different - the MIDI event in MuseScore gets constructed *exactly* the same same way - the *exact* same sequence of bytes - either way.

So, the bottom line remains - using the Inspector to set velocity for notes individually in *no way* defeatsa the normal and correct velocity splits built in to a soundfont. if a soundfont switches to a less bright sound as the velocity gets less - which most soundfonts do, because that is how real isntruments work - then it will do this whether you use dynamic markings or the Inspector.

If you prefer the sound of 70 for your "mp" on trumept over "64", you can simply enter the "mp" then change the velocity *of the marking itself* to 70 - that way, the marking still appears in the score as it should.

And I would agree the transitions on this particular sound do sound a little awkward, but I think it's because "p" is too birght, not because "mp" is too dark. Still, a small quirk of this one patch on this one soundfont, not a reason to abandon the use of dynamics.