How long did it take to upload the files? Do the sounds show up right away? Or do I need to do more? I assumed Zynthian will automatically deflate the compressed file and sort things out. Any thoughts? Thanks
FYI: It took more than an hour to upload the compressed file. The uncompressed file was 24 GB. So perhaps it will take a long time (perhaps several hours) to decompress the files and add them to the banks?
i copied the files uncompressed and It worked well. You should delete de compresdrd file from the zynthian and upload the files uncompressed to this fĆ³lder:
/zynthian/zynthian -my-data/soundfonts/sf2
Regards
1 Like
How do I get to that folder?
Also did you upload only one file at a time to check it or did you load all of the sounds? I am not familiar with Zynthian, The only option it gave me to upload was as a compressed file.
Did you try the file browser? You can access from the webconf:
user: zynthian
passwd: opensynth
You can also access using SFTP:
user: root
passwd: opensynth
I copied the full collection.
Regards
1 Like
Thank you @hannesmenzel, for this nice addition to the Zynthian soundfont weaponry 
.
I downloaded the first version of the library and copied it to the Zynth with FileZilla: lengthy process, no errors, quite heavy once deflated.
I have played around with this collection this morning, on a custom Pi5-8Gb-Oram machine with nvme, at medium overclock. Loading of the presets in FluidSynth is fast on this setup, and I couldnāt detect any issues in performance, with no x-runs at all and quick response (very playable latency) at 48000/256.
The transfer of the Fantom X sounds has been accurate. I am pretty familiar with the time-honoured classic Roland patch base, dating back to the distinguished sound-design work of Eric Persing in the late 80s and early 90s.
The presets in this library mandatorily require FX treatment, which I appropriately imparted through delay, reverb and occasionally chorus.
Generally, I find the sonic quality to be not completely consistent: some of the patches obviously sound way more static and less dynamic than what the original and complex rompler synth engine did, with approximately the same samples base.
I noticed that strings in particular suffer, from being somehow photocopied and crystallised in soundfont presets - sometimes lacking dynamic response, realistic attack and high end -, but there are nevertheless some convincing and pretty beautiful sounds even in this category.
Pianos are a bit on the brighty side but playable and expressive, and some of the orchestral brass are solid and imposing. The sonic areas where I discovered the most pleasant surprises are Bright Synth Pads, with some excellent and complex evolving soundscapes, and the guitars department, with many remarkable acoustic and electric presets.
Overall, I find this soundfont library to be very useful, for both live playing and multi-track arrangements. I must say that Iām very impressed, by the arguably demanding adaptation and translation task from Fantom X.
1 Like
Iām glad it is partly useful. I originally stated this just as an empowering showcase for the close relation of āIām searching for sound Xā and āI googled for sound Xā 
1 Like
Good morning:
Can any one help me how to call up the sound banks and the sound patches using MIDI? In other words what MIDI ācodeā I need to send to call up e.g., the Pianoteq sound group and the first patch of that group?
Also can some ātranslateā the various commands to save as a new set of 16 track presets? I am so confused. What happened to the terms such as āSaveā āSave Asā I am used to in PC language?
Thanks
Hi,
Zynthian, as just stated elsewhere in another thread, employs a system of Midi-learning based on fast recall of the whole preset state, of a complete snapshot up to 16 Midi channels. The classic bank selection + program change Midi packet would be theoretically usable, and mostly recognised, but would be slow and actually impractical.
I suggest you search and study the wiki looking into the ZS3 sub-snapshot topic, andā¦ happy learning! 
1 Like