So this is the launcher script I have to trigger on resumption from sleep:

#!/bin/sh
# Restart Pianoteq after returning from suspend
# In systems using SYSTEMD, place in /usr/lib/systemd/system-sleep/pianoteq-restart.sh
# Otherwise, place in /etc/pm/sleep.d/20-pianoteq-restart

# Symlink exists to make a unique name to catch only the running process
#   eg: ln -s /usr/local/Pianoteq\ 5/i386/Pianoteq\ 5 /usr/local/Pianoteq/PT5
USER="piano"
PTBINARY="PT5"
RESTART="/usr/local/Pianoteq\ 5/restart_pianoteq"

if [ $(ps auxww | grep -v grep | grep -c $PTBINARY) -eq 0 ]; then
  exit 0
fi

case "${1}" in
  hibernate|suspend)
    ## nothing
  ;;
  post|resume|thaw)
    su $USER -c "$RESTART" 
  ;;
esac

Then I have the following script to do the actual restart:

## /usr/local/Pianoteq 5/restart_pianoteq

PTBINARY="PT5"

# w/o waiting for a few seconds, the script does not run successful after immediately returning from sleep
sleep 2
date >> /tmp/pt-sleep.log
echo "Stopping PT" >> /tmp/pt-sleep.log
ps -ef | grep -v grep | grep $PTBINARY | awk {'print $2'} | xargs kill

sleep 1
date >> /tmp/pt-sleep.log
echo "Starting PT" >> /tmp/pt-sleep.log
/usr/local/Pianoteq\ 5/PT5 &

Ran into some Linux annoyances today after setting a laptop dedicated solely for Pianoteq.

When doing Suspend -> Resume, Pianoteq loses the audio device and presents the audio device selection box.  Closing it and restarting works.

With JACK server, Pianoteq maintains the connection to JACK ... but unfortunately JACK loses it's connection to audio.  Restarting JACK does not work.

I'm scrounging the internet now to experiment with suspend/resume scripts to force redetection of audio devices.

Did you ever resolve this ?
Just using the square "Stop" button instead of the || "Pause" button does work, on Windows as well as on Linux, i.e. the same on Windows as on Linux.

It is a "shortcoming" (to be polite) or a minor bug at worst.

Then I suggest you start here;
http://cdimage.ubuntu.com/ubuntustudio/...4/release/
select the .iso distribution for your processor type, AMD or intel and save the file, about 2 gigs.
Burn it to DVD as an iso disc image, NOT as a file.

Despite the naming of the files it's not a choice between Intel or AMD but between 32 or 64 bits.

The filenames with "amd64" in it are for 64 bits systems, even when using an Intel processor.
The filenames with "i386" in it are for 32 bits systems, even when using an AMD processor.

just a little bit confusing :-)

Ran into some Linux annoyances today after setting a laptop dedicated solely for Pianoteq.

When doing Suspend -> Resume, Pianoteq loses the audio device and presents the audio device selection box.  Closing it and restarting works.

With JACK server, Pianoteq maintains the connection to JACK ... but unfortunately JACK loses it's connection to audio.  Restarting JACK does not work.

I'm scrounging the internet now to experiment with suspend/resume scripts to force redetection of audio devices.

I haven't used pause/play much in pianoteq, I normally do that in the midi player that feeds midi to pianoteq.
I tried it, there is no change in the Jack connections with pianoteq pause/play;
pause/play doesn't continue,
however pause/stop/play picks up and continues from where I paused.

This appears to be NOT a LINUX problem, i.e. it behaves the same way on Windoze.

I am still on 4.5.5 - - given that this happens on more than one OS my GUESS is that pianoteq is the one with the bug.

When doing Suspend -> Resume, Pianoteq loses the audio device and presents the audio device selection box.  Closing it and restarting works.

With JACK server, Pianoteq maintains the connection to JACK ... but unfortunately JACK loses it's connection to audio.  Restarting JACK does not work.

I'm scrounging the internet now to experiment with suspend/resume scripts to force redetection of audio devices.

Back up your windows system - just in case.
It is probably worth shrinking the windows partition first.
Control Panel -> Administrative Tools -> Computer Management -> Storage
Right click on your C drive to see the option to shrink.

For a reasonably complete Ubuntu Studio installation a 32 gig partition should be adequate.
Boot from the DVD.
Select the mundane things like time zone, language, user name, password, keyboard type, etc.
Eventually it will ask if you want to repartition your hard disc - this is why you should have backed-up the windows partition.
If you have shrunk the windows partition there should be free space available for you to create a new partition, if not you can PROBABLY do the re-sizing of windows now, but that has been risky in earlier versions so my habit is to only resize windows using windows.
   
Basically follow along, the install is quite automated.
It will install a boot loader, so when the install is complete you will be able to choose between windows and your new Ubuntu Studio OS - up arrow and down arrow keys.

When it comes up the first thing to notice is that the menus are all started from a button at top left of the screen.
The major categories should be fairly obvious, "Audio Production" will probably interest you first.
"Qjack ctl" is the jack control - this is what connects outputs of things to inputs of other things. 
The analogy is taken from telephone "Jack" plugs, the ancient days of telephone "Switchboards".

I appreciate that style of music, but I doubt that much would be lost from it with less polyphony, or if the sustain pedal was lifted a bit more often.

Lylo wrote:

Hello,
have you make a similar test with Pianoteq 4.5 ? I'm very surprise of the result of your test ? I play Pianoteq on a Core Duo 2 2,2ghz and I never have problem (and I play a lot of notes with another vsti in parallel in live performance).
If the Pianoteq 5 need more power perhaps I must not install it in my stage computer.

Hello,
Ton système est-il parfaitement stable sous windows en 48000Hz et 128 samples, pour 2,7 msec de latence? Chez moi ce n'était pas le cas. Alors que sous Tango Studio, pas de problème.

Je vais faire des tests, il faut simplement qu'en les faisant je n’abîme pas mon pc de scène, que si je rencontre un problème je puisse revenir en arrière.
Ça serait malheureux en tout cas que je ne puisse pas profiter de cette version 5.

------------------------------------------------

tracto_music, I must listen the opinion of stamkorg for my use, look at this video (Pianoteq 4.5) 

http://youtu.be/dB74AHJkqWU?t=4m40s  Clic on the link for watch from 4:40

@Lylo
With all due respect to stamkorg, such a stress test is WAY beyond any player's "NEEDS" for polyphony.
It is certainly interesting and does reveal the greater margin that is available with Linux, {specifically with the real time (or "low latency") kernel} than is available with the microsoft general purpose "office_ware" OS.
Basically, unless you throw incredibly dense midi files at pianoteq, e.g. midi files of player piano rolls, on a 2.2 gig Core duo I don't think you will come anywhere near any limits with MS_Windows.
=======================================================
In magazine tests supposedly my motorcycle is 300 Km/Hr capable, probably with a 70 Kg rider.
I don't have test track conditions, I weigh more than 70 Kg., I would like to live a lot longer etc., so while that limit is interesting, it is largely irrelevant.

Hello,
have you make a similar test with Pianoteq 4.5 ? I'm very surprise of the result of your test ? I play Pianoteq on a Core Duo 2 2,2ghz and I never have problem (and I play a lot of notes with another vsti in parallel in live performance).
If the Pianoteq 5 need more power perhaps I must not install it in my stage computer.

Hello,
Ton système est-il parfaitement stable sous windows en 48000Hz et 128 samples, pour 2,7 msec de latence? Chez moi ce n'était pas le cas. Alors que sous Tango Studio, pas de problème.

I may experiment with Tango Studio - just because I like messing with various strains of Linux

EDIT:
Ahh, I see it is a Debian distro - GOOD, I am already a fair way up that learning curve
I hope it continues to hang together when I update the kernel, I would like to use b-cache.
:END EDIT

So here we go:

I experienced a "stress test", playing a maximum of notes with the right pedal down, to increase the polyphony to 128 (limit fixed at 128 in Pianoteq), increasing the CPU charge too.
The test is made on the same laptop and with same strains, i5, 4Gb RAM, soundcard: Focusrite Scarlett 6i6, at 48000Hz for an "on the screen" latency of 2,7msec (total in/out).

Result:

With Windows 7: at 128 of polyphony, the CPU charge comes to +/- 90-95%, a lot of dropouts occur.

With Tango Studio: at 128 of polyphony, the CPU charge comes to max 60%, without dropouts (Xruns in the Linux world). I was able to maintain this strain during 1 entire minute without dropouts. Then I stopped the test.

I just wanted to share this because I have now the conviction that a Linux Tango Studio OS is far more efficient and stable for music than Win 7.

SK