View Single Post
Old 11-07-2017, 02:53 PM
Sherry C's Avatar
Sherry C Sherry C is offline
Product Manager
Join Date: Jul 2009
Location: Bad Axe, MI, USA (The Tip of the Thumb of Michigan)
Posts: 2,355
Default Re: Audio export 3.0.6-130

Hi David,

Thanks for your (always) thoughtful testing and feedback. I'll respond below.

Originally Posted by dj View Post
1: Why does it default to .mp3 export each time I call up the export feature -- even if I've just exported to .wav format a minute before? It should remember the last format.
The mp3 format was chosen as the most-used format and thus the default, but you have a good point about remembering the last-used audio filetype. That is now Task 2980, to make Composer remember the last-used audio export filetype.

2: I see that there are separate settings for playback and export gain in the Soundfont configuration box. Should they be separate or should it be a "what you hear is what you get"? I know the idea is to make adjustment for different Soundfonts, but I can see surprises with export gain being 'way too low (or too high) in comparison to what the user has been hearing.
The following response is directly from Reinhold:
I believe the principle "what you hear is what you get" is simply not feasible because it strongly depends on the volume of the playing devices. A MP3 ear phone playe may have different volume that a player on a PC. The principle which I used was to keep the volume as high as possible but w/o any clipping. That is how the values were determined based on our standard soundfonts. The reason why the gain for the internal player and the export file is different is, the software is different. Internally we use our well approved internal player. For export I use fluidsynth which creates different volumes. Further, each different audio format has different parameters from different software libraries and need to be adjusted differently.

[Sherry] I agree with Reinhold because I get quite a variety of volume levels on my different devices, and so it is really going to be up to the user to determine what levels work for them for the device(s) and filetype(s) they want to use.

3. Even with export gain at the default 54%, I am getting a lot of distortion in test exports. That suggests to me that the user will have to do a lot of back and forth between gain and export in order to get it right. I've had to reduce .wav export gain to 32% (with my system volume at about 40%) before I get rid of the distortion.
Again, this can vary by the user's device(s), and we also offer the option to load various soundfonts which can differ in their inherent volume levels. So really all we can do is offer the option to be able to adjust the soundfont's volume properly on the user's particular system. Perhaps we can set up a page here in the User Forum where folks can share the gains/volume for different soundfonts that they're using on their systems.

It would be more work for you, I know, but may I suggest, rather than manual gain adjustment, an "auto-optimization" on export? Have the export function go through the playback (silently), find the highest level and automatically adjust the export gain to account for that. That would make it MUCH more user-friendly. If the user did not want to wait the extra few seconds, a check-box to switch the auto-optimization off before export could be provided.
Yes, that would indeed prove to be quite an effort for our development team We are not aware of anyone else in our genre of software that offers such advanced capability AND with the flexibility of loading different soundfonts. If all we offered were a single soundfont for the export capability then it would be a much easier task, and this is indeed what happens with other software that we are aware of. However, by offering the capability to do audio export with whatever soundfont the user selects, it becomes a much more difficult task. In order to get these features out to our users sooner rather than later we plan to leave the adjustments as a manual user task for now, with hopefully safe default levels to start with. If we find that users are having too much trouble with these settings then of course we will address that in a future release.

Thanks so much for your consistently excellent feedback!

Music is to the soul like water is to green growing things.
Reply With Quote