TX16Wx not loading .sfz from Sonatina Symphonic Orchestra library correctly

[Archer6621]

Hey there,

The only thing that is preventing me from buying the extended version of this amazing plugin is the fact that it doesn't seem to be able to load/import all the sfz files from the Sonatina Symphonic Orchestra library 100% correctly, where other .sfz - supporting plugins are able to do this fine (such as sforzando).

For instance, when I load the "Strings - Celli Sustain.sfz", the sound quality seems to have reduced significantly and there even seem to be pitch issues at notes C4 and the one before compared to all other notes (it suddenly becomes an octave lower out of nowhere??).

Regarding the sound quality: all of the attack/release values are completely wrong and sound as if they have linear gain/falloff? I was hoping to be able to utilize these instrument presets by tweaking them from their initial state, and it's really the sole reason I want to use this plugin, because it is very useful for grouping/layering multiple of these orchestral instruments within the plugin itself, but I'm not keen on re-configuring each of them so they will sound like they should, it just takes too much time.

Will you consider making the .sfz import more accurate?

Thanks for reading!

[elcallio]

Usually, I'd probably say "version 3" to this. But ff I have a little time free this weekend, I can see if there are any low hanging fruit to be picked here. But I have a distinct memory of fixing a number of issues with this library already, so who knows.
Anyway, it probably helps if you list what you feel are the shortcomings in loading the library, as my ears might differ from yours.

[Archer6621]

Thanks for responding!

I'm happy that you want to tackle this one! When I have time (likely now/tomorrow) I will go through all of the instruments and make you a list of things that sound different from the plugins that simulate the sfz. playback (like Sforzando) via an engine (I encourage you to confirm them by playing back with Sforzando, the differences are sometimes quite notable).

I'll just edit this post then below

EDIT: I've also added recorded samples in some cases to show the difference, the first bit is always Tx16Wx in that case, and the second bit is always from Sforzando. Always velocity 64 unless I'm showing velocity-specific issues, although in some cases I normalized the volume to make anomalies more apparent (denoted with (N)).

The list:
Global issues:

• Velocity mapping seems incorrect, the scale seems smaller on Tx16Wx: https://dl.dropboxusercontent.com/u/518 ... locity.wav (N)
• Attack/Release fall-off is linear instead of exponential, makes them sound as if they're longer, I think the values are correct though. Especially notable on sustain instruments: https://dl.dropboxusercontent.com/u/518 ... ustain.wav (N)
• In Staccato instruments, it seems that the filter is too strong in some cases, it get muffled too much from the velocity: https://dl.dropboxusercontent.com/u/518 ... accato.wav (N)

Strings - Celli Sustain.sfz

• Pitch octave issue with note C4 and C#4, not sure why but it suddenly becomes 1 octave lower just for those two notes

Strings - Celli Staccato.sfz

• Seems to have an unintended low-pass filter going on when triggering lower velocities, or maybe it is way too strong

Strings - Basses Staccato.sfz

• Per note it alternates between two notes as intended, however one is significantly louder than the other, notable at C2 (could be related to filter bug though): https://dl.dropboxusercontent.com/u/518 ... accato.wav (N)