Reconfigure Aeolus Pipe Organ Synth

• May 18, 2013 - 11:34
S5 - Suggestion

The existing implementation of Aeolus is unsatisfactory in it's allocation of stops to divisions for a number of reasons.

A new configuration would be beneficial to organists, working with Musescore.

I am going to take this job on, as I understand the tonal design of real organs and will be able to apply that knowledge to Aeolus.

Initially I will be working empirically using the Ubuntu Studio installation of Aeolus as documentation on how to configure Aeolus appears to be non-existent, although I have ascertained some things from the Ubuntu Studio installation.

Once I have a working version in Ubuntu Studio I will then transfer the data to MuseScore and submit a pull request.


Does anyone know where the stop list for the Aeolus GUI is read in from?

It isn't from instruments.xml

I need to know because one of the stop labels is currently wrong - reading Rohrflöte 8 instead of Rohrflöte 4

And also because some of these labels will need changing anyway.

The Linux installation reads them in from the .ae0 stop definition files, but from my experiments so far that is definitely not the case with the MuseScore version.

Incidentally experiments I have made so far have been successful in that changes to the stoplist have not crashed the Aeolus Synth in Ubuntu Studio.

EDIT: No worries I have just found the UI file - I was looking in Aeolus and in fact it is the StaffText UI file

Progress :)

Thanks to Miwarre's research I am now able to generate new stops for Aeolus, which is what I have been doing this afternoon.

I have now started on the design for a new organ for Aeolus, specifically designed to cross classical organ genres, and thus be suitable for use in MuseScore. There is still a little work to do on the design, but we are a world away from where we were at the beginning of this thread, and I am expecting to have the organ finished in the Linux version of Aelous towards the end of next week.

I shall then undertake the task of porting it across to MuseScore.

Once I have finished that it would be possible to design a second Aeolus organ specifically with theatre organ pipes. I would need to know whether there would be any interest in such an animal from the MuseScore community before I embarked upon this, however.

I am going to need testers shortly, so if you can help with that, please let me know.

Sorry Michael, I have overlooked this thread.

For testing, count on me for sure (I don't know if there are lots of peoples familiar with MuseScore-Aeolus usage...)

I have no clear idea about "theatre organ pipes" (I'm not even sure what they mean...). Generally speaking, I would assume that, once a 'general' enough instrument is available, any specific model would raise more criticisms for what is not supported ("I need a French Romantic organ!", "What about a regale for Monteverdi's Orfeo?" and so on) than applause for what is included.

Also, once we have two models, we need a way to select them in the score.

But, if nothing else, it would be a welcome addition to the standalone Aeolus version!


P.S.: I'm about to start working on an instrument editor for my oscAeolus fork: any specific suggestion, as you now have experience in creating Aeolus instruments?

I thought you guys would like to know I have just finished the first tweak to Aeolus, and am just about to see if it builds properly for MuseScore.

Ah, that branch has been superseded. Unfortunately I had to delete it to solve problems with merging a recent update.

Its replacement is in the process of more tweaks and is not in the state where you can build from it. I'm expecting to have a build ready later today - depends how the config behaves in the Linux Aelous version.

If you are on windows I have an install build you can try - let me know if you'd like it.

Guys, I would love to read these discussions and use them as reference in the future. Would be cool if you keep them public, maybe in the forum?

Status (old) active patch (ready to commit)

Pull request #484 sent 5 minutes ago.

Please note the problems I have mentioned with the Trombone - I have reduced the upper partials and also reduced the random harmonics in the upper register, which I believe were causing problems with the wave calculation, and causing digital clipping to occur.

My experience with this kind of programming is limited - if there is anyone else around with the experience to look at this please feel free :)

I didn't use the trombone stop frequently, so I haven't had issues in this respect. Was the clipping present at any volume level set in the audio section?