Anyone using the Soundtower Librarian

[Shaw]

Just wondering if there are any users here who can comment on the Soundtower Librarian for the Prophet 5 (on a Mac, if possible)... Just wondering about bugs and reliability.


Thanks in advance,


Shaw

[S_A_P]

Hey Shaw-

Im using it. Generally like it, but its got some bugs:

1) 3 way switches (Filter tracking, Velocity, aftertouch) dont track correctly in the UI.(the off state is never represented when you click it on the synth, just red led or orange led) Velocity and aftertouch light up the incorrect led on the UI( for instance the patch Im looking at right now has Filter(red led) selected for velocity but in the soundtower ui it shows Amp(orange led)
2) I believe the volume and glide rate pots do not import with the patch when you click "Receive current patch"
3) Frequency of Osc A and B is incorrectly represented (on the synth its set to A0 and Soundtower indicates E1) However the knob is set to the correct position in the editor( for instance if the synth says its on 2 the UI will indicate 2 as well, just associate the wrong note with that value)
4) some general instability on occasion(Im on MacOS Catalina 15.7.x)


In spite of the bugs(Ive reported to the developer, who was initially responsive but became much less so as I reported more) Im glad I have it. I hope that maybe the bugs will be worked out at some point but its plenty usable as is. I find the patch librarian, morphing and virtual patch memory a handy addition(you can have multiple banks in the editor that you can switch into the user memory without much fuss, effectively giving you infinite patch memory)

[Shaw]

Hey Shaw-

Im using it. Generally like it, but its got some bugs:

1) 3 way switches (Filter tracking, Velocity, aftertouch) dont track correctly in the UI.(the off state is never represented when you click it on the synth, just red led or orange led) Velocity and aftertouch light up the incorrect led on the UI( for instance the patch Im looking at right now has Filter(red led) selected for velocity but in the soundtower ui it shows Amp(orange led)
2) I believe the volume and glide rate pots do not import with the patch when you click "Receive current patch"
3) Frequency of Osc A and B is incorrectly represented (on the synth its set to A0 and Soundtower indicates E1) However the knob is set to the correct position in the editor( for instance if the synth says its on 2 the UI will indicate 2 as well, just associate the wrong note with that value)
4) some general instability on occasion(Im on MacOS Catalina 15.7.x)


In spite of the bugs(Ive reported to the developer, who was initially responsive but became much less so as I reported more) Im glad I have it. I hope that maybe the bugs will be worked out at some point but its plenty usable as is. I find the patch librarian, morphing and virtual patch memory a handy addition(you can have multiple banks in the editor that you can switch into the user memory without much fuss, effectively giving you infinite patch memory)

Yeah... I’ve noticed some bugs too... the LFO initial amount and frequency knobs are switched (if I turn the initial amount knob on the synth, it effects the frequency knob on the editor.  And vice versa)


Essentially it is not very useful as an editor, but as you stated, golden as a librarian.

[S_A_P]

You know I haven't really even tested that the patches are exported correctly, have you? I could download your patches from the site and not know they were "incorrect". Im going to load a few sysex and p5ps I've created and compare with the originals...

[S_A_P]

Can confirm subtle differences in some exports. Not sure why(and the reason could be as simple as I changed/improved the preset after export). I am going to resubmit a few patches to you Shaw. I exported and reloaded the few I noticed issues with and they seemed to match. Weird...

[Shaw]

Can confirm subtle differences in some exports. Not sure why(and the reason could be as simple as I changed/improved the preset after export). I am going to resubmit a few patches to you Shaw. I exported and reloaded the few I noticed issues with and they seemed to match. Weird...

I will try to test this later, but essentially, download both the .sysex patch and Soundtower patch. Use Soundtower to move the preset to an adjacent slot.  Load the sysex file directly into the P5 (not using Soundtower), then use Soundtower to load the “same” into the adjacent program slot... then compare.

That should give you a good comparison.

... or export a patch from Soundtower to sysex format, then look into the files to compare them bit by bit... comparing the original sysex from the P5 to the exported sysex from Soundtower.

[S_A_P]

good idea- I will

Can confirm subtle differences in some exports. Not sure why(and the reason could be as simple as I changed/improved the preset after export). I am going to resubmit a few patches to you Shaw. I exported and reloaded the few I noticed issues with and they seemed to match. Weird...

I will try to test this later, but essentially, download both the .sysex patch and Soundtower patch. Use Soundtower to move the preset to an adjacent slot.  Load the sysex file directly into the P5 (not using Soundtower), then use Soundtower to load the “same” into the adjacent program slot... then compare.

That should give you a good comparison.

... or export a patch from Soundtower to sysex format, then look into the files to compare them bit by bit... comparing the original sysex from the P5 to the exported sysex from Soundtower.

Can confirm that they are much different. I sent a few patches your way a few minutes ago hopefully correcting the issue.

[Analog Prophet]

I have bad experience of lack of support from Sound Tower, can not recommend them.

[Shaw]

Damn it...  Not that I didn't trust SAP's results, but I did my own test and the syses values are very different between the syses transmitted from the P5 and the same preset exported from Soundtower.  (see attached)


... so this appears to not work very well as an editor OR a librarian.


Does anyone know of another librarian solution that is Mac compatible?

[S_A_P]

Shaw and I have corresponded via email over this- from what I see is the non soundtower file has 2 additional bytes of data after the sequence F0 01 32 at the beginning of the file. In what I assume is the P5 export there is 2 bytes of 02 02, and then the rest of the file appears the same at a quick glance(did not check every value, but it looks like there is just a 2 byte offset due to the 02 02) I thought the sysex table was in the manual but dont see it. Do you know where I could find it to take a look? Hoping the 2 bytes are not relevant to the sound, but concerned that it may be the case especially considering the bugs in 3 way button handling...

[Shaw]

There are also bytes at the end that are different.

[Analog Prophet]

Does anyone know of another librarian solution that is Mac compatible?

I have used Sysex Librarian for a number of years to my synths and happy with it: https://www.snoize.com/SysExLibrarian/

[Qwave]

I have bad experience of lack of support from Sound Tower, can not recommend them.

Same here.

[Qwave]

Does anyone know of another librarian solution that is Mac compatible?

I have used Sysex Librarian for a number of years to my synths and happy with it: https://www.snoize.com/SysExLibrarian/

The SysEx Librarian is a wonderful software. It never failed me. And I did huge SysEx files to many, many different synths. But using it to shovel patches in your synth, it is just like a record player used to do a playing list. There is no way of moving the files. You have to manually receive and send them to the needed places. I would not recommend it as Librarian to sort patches. But it is great for backups, sending SysEx to synths and to do updates via SysEx.

[markwilkins]

Hey Shaw-

...
3) Frequency of Osc A and B is incorrectly represented (on the synth its set to A0 and Soundtower indicates E1) However the knob is set to the correct position in the editor( for instance if the synth says its on 2 the UI will indicate 2 as well, just associate the wrong note with that value)
...

Some of the behavior you describe sounds very similar to what I see when using the SoundTower software with a Prophet 6. It seems to differ by patch, but quite often the software simply does not represent the actual settings on the synth. I change patches on the synth, and the settings in SoundTower update ... but only partially.

What I have found, though, is that if I open some dialog box in the SoundTower app (the about box for example), the settings “correct” themselves. I wonder if the behavior might be the same with the P5 software. I have to imagine it is largely the same code base and may share the same bugs.

[S_A_P]

Hey Shaw-

...
3) Frequency of Osc A and B is incorrectly represented (on the synth its set to A0 and Soundtower indicates E1) However the knob is set to the correct position in the editor( for instance if the synth says its on 2 the UI will indicate 2 as well, just associate the wrong note with that value)
...

Some of the behavior you describe sounds very similar to what I see when using the SoundTower software with a Prophet 6. It seems to differ by patch, but quite often the software simply does not represent the actual settings on the synth. I change patches on the synth, and the settings in SoundTower update ... but only partially.

What I have found, though, is that if I open some dialog box in the SoundTower app (the about box for example), the settings “correct” themselves. I wonder if the behavior might be the same with the P5 software. I have to imagine it is largely the same code base and may share the same bugs.

It is definitely quite frustrating. I get that its early yet in the rev4 lifespan, but sounds like this is an ongoing problem with SoundTower...

[composerjk]

Damn it...  Not that I didn't trust SAP's results, but I did my own test and the syses values are very different between the syses transmitted from the P5 and the same preset exported from Soundtower.  (see attached)


... so this appears to not work very well as an editor OR a librarian.


Does anyone know of another librarian solution that is Mac compatible?

In this example, the SoundTower SysEx is a Program Edit Buffer Data Dump (byte 4, value 03), whereas the one from the Prophet-5 is a Program Data Dump (byte 4, value 02), which has two extra bytes for the patch number.

If you want to request the Program Edit Buffer Data Dump, you can send this to the Prophet-5 to receive a dump of the edit buffer (which should hopefully more closely match the SoundTower SysEx):

Code: [Select]

sendmidi dev Prophet syx hex 01 32 06
where Prophet is the name of the MIDI device to use to send data to the Prophet-5. Note I haven't tested that, as I don't have a Prophet-5 Rev4, at the moment.

The byte before the EOX (0xF7) at the end is a checksum and will be different. I believe those other last bytes are currently undefined and possibly not used, so that may be okay. If there are differences you're concerned about, let me know and I can take a look at the SysEx file.

The MIDI Implementation Chart for the Prophet-5/10 is in the Sequential support sections for those synths. Note that there are some typos, as of rev. 1.3 of the document, e.g., the device number listed should be 0x32 (0011 0010) instead of 0x31. I already reported that.

[composerjk]

Shaw and I have corresponded via email over this- from what I see is the non soundtower file has 2 additional bytes of data after the sequence F0 01 32 at the beginning of the file. In what I assume is the P5 export there is 2 bytes of 02 02, and then the rest of the file appears the same at a quick glance(did not check every value, but it looks like there is just a 2 byte offset due to the 02 02) I thought the sysex table was in the manual but dont see it. Do you know where I could find it to take a look? Hoping the 2 bytes are not relevant to the sound, but concerned that it may be the case especially considering the bugs in 3 way button handling...

The Prophet-5 dump is a Program Data Dump (byte 4, value 02). The following two bytes (02 02) indicate the patch number. Byte 4 in the SoundTower dump is 03 indicating a Program Edit Buffer Dump, which doesn't include patch numbers, hence the 2-byte difference. I responded to Shaw's post with his images and also included mention that the MIDI Implementation Chart is in the Prophet-5/10 support section.

[Shaw]

Shaw and I have corresponded via email over this- from what I see is the non soundtower file has 2 additional bytes of data after the sequence F0 01 32 at the beginning of the file. In what I assume is the P5 export there is 2 bytes of 02 02, and then the rest of the file appears the same at a quick glance(did not check every value, but it looks like there is just a 2 byte offset due to the 02 02) I thought the sysex table was in the manual but dont see it. Do you know where I could find it to take a look? Hoping the 2 bytes are not relevant to the sound, but concerned that it may be the case especially considering the bugs in 3 way button handling...

The Prophet-5 dump is a Program Data Dump (byte 4, value 02). The following two bytes (02 02) indicate the patch number. Byte 4 in the SoundTower dump is 03 indicating a Program Edit Buffer Dump, which doesn't include patch numbers, hence the 2-byte difference. I responded to Shaw's post with his images and also included mention that the MIDI Implementation Chart is in the Prophet-5/10 support section.

So, in dumb lay-man, non Midi speak, are you saying that the two sysex files contain the same program parameters, and the only bytes that are different are not related to the formation of the sound.

[composerjk]

So, in dumb lay-man, non Midi speak, are you saying that the two sysex files contain the same program parameters, and the only bytes that are different are not related to the formation of the sound.

Possibly. I didn't look at every byte. The two byte difference is the patch number included in one and not the other. The Program Edit Buffer Data Dumps never include the patch number. That's likely the key difference in your example.

If you have SysEx files to compare, I can see what the underlying differences may be.