wow, i didn't know about xpra, thanks for the input!!!

I also just have discovered Xpra, it seems to be really useful!

Hm, seems not to be a general problem, it is working here. Maybe you like to reproduce.

My test-scenario:

* A netbook with Debian, Pianoteq and xpra installed and my midi-keyboard connected.
* Netbooks Desktop-GUI is shutdown ('/etc/init.d/kdm stop' here)
* From another workstation with a Desktop-GUI and xpra installed I "fetch" the Pianoteq-window with:
* xpra start ssh:1.2.3.4:100 --start-child=/home/alice/Pianoteq
* result: I can play piano :-)
* I disrupt/detach the session on my workstation with Ctrl-c and keep on playing.
* I can re-attach the Pianoteq-window again from my workstation with: xpra attach ssh:1.2.3.4:100

PS: Audio-Output in Pianoteq is driven by ALSA direct to hardware.

I just like to mention, that something between your idea and simple X forwarding exists. With xpra you can attach and reattach a persistent running Pianoteq-Window, that is running on a GUI-less "server".

wow, i didn't know about xpra, thanks for the input!!!

...except (just after a rough tryout through xpra) pianoteq outputs sound only if tis window is showing on the actual screen. I'll have to look into more detail if i can get it to discard unneeded X11 events.

Thereby I had the idea that the up/down keys could be used to skip through the parameters of the respective panel once you clicked one parameter and the left/right keys to change the values. There had to be found a special solution for the spectrum profile.

The way i see it, the two straightforward approaches to this would be (1) something similar to VNC, where the --headless (ie. server) process sends pixel data to the --remote (client) process and the --remote process sends mouse and keyboard event data to the --headless session and (2) where only the actual settings would be exchanged between the two processes, in some newly invented format (which, again, is to be arbitrarily invented by Modartt if they implement this feature ) The (1) approach would be less efficient in terms of bandwidth use, but would be more tolerant with version mismatches.

Anyway, just a thought... This would also solve the problem (or improve/simplify the situation) of having pianoteq running on a PC or mac, with the interface showing up on a tablet, which a few people have mentioned recently.

Here's a feature i'd really like to see: being able to run pianoteq headless (no UI mode) and then "connect" to it by running another pianoteq process to control it, change presets and settings etc.

I was thinking about a "Essential Pianoteq" as a FXP reader with a GUI window made just with 16 "presets buttons" , MIDI setup and keyboard velocity calibration.
Stop.
Nothing else.

Here's a feature i'd really like to see: being able to run pianoteq headless (no UI mode) and then "connect" to it by running another pianoteq process to control it, change presets and settings etc.

Hi delt,

have you tried a normal xsession via SSH to your remote "Pianoteq-server"?
PTQ is started remote then and the GUI is rendered on your local PC. No remote desktop/GUI necessary for this.

I love the automatic MIDI history, something I've been wanting to create myself. Could you and an option for an infinite history length and a location for the MIDI files? MIDI data takes up so little space these days that there is no reason to ever delete any of it. And it's nice to know that everything that was ever played is in principle retrievable. I could sure use just having the last months worth....

Big history is good until you try to find something in it. 10-20 entries is a sufficient amount if you want save a recording you did accidentally and recently. In fact this is why we need this tool, it's not a way to eliminate file saving/loading at all.

But for the use  in a very "easy" live context, to think you can  use only a keyboard connected to the tablet without other programs from having to load and manage not mind ...

I could only use Pianoteq stand-alone.

It is not important for me to have a lot of sounds to use as pads ... being able to create 3 or 4 would be more than sufficient and I don't think this features make the program less professional in the eyes of purists ..;)

I think that even "Ivory" use this feature

Sorry for my bad english...

Ciao!

Among the features that I would like to see in the new release of Pianoteq there would be the implementation of a small section for adding a pad sound.

This section should provide a button on / off, a generation section also minimal for the timbre (some minimal  parameter like timbre n. 1-2-3 .. LP Filter, ADSR, Effect association, CC64 association on / off) .

the on/off status of this section should be saved along with the piano preset  ....

I forgot ... the sound generation could also be a "physical model type" (obviously a very simple version) so as not to overload the program size..

For all those who use Pianoteq live in areas not strictly classical, I think this is an remarkable improvement ...

What do you think about?

Ciao!!!

http://forum.renoise.com/index.php/topi...gin-linux/