Pianoteq on Linux netbook

Seems as though the netbooks need to have just a slightly faster cpu for audio. Maybe the next generation will move up to 2 ghz processors?

Hey, that works great! I overlooked that, because under options>Devices I only have the Sample rates 44100, 48000, 96000 and 192000 Hz. But under options>Perf I can choose an internal sample rate of 11025, 14700, 22050 and 44100 Hz (as you wrote). With 22050 I can play all Acoustic-Pianos without any glitch, even the performance-hungry preset Grand C3 Solo Recording! I'm a happy camper now. Bought! It is a little pricy (for me), but I want to honour Modartt's engagement in Linux thereby.

After playing a while it shows Performance index 8. But I don't understand, what's the meaning of this index - can you shed a little light on this?

Aha. For comparison I started the eee-pc with a stock kernel (linux-image-2.6.26-1-686) and multicore rendering is greyed/uncheckable there also.
Here the answer to your question with the realtime-kernel:

~# cat /proc/cpuinfo                                                                             
processor       : 0                                                                                   
vendor_id       : GenuineIntel                                                                         
cpu family      : 6                                                                                   
model           : 28                                                                                   
model name      : Intel(R) Atom(TM) CPU N270   @ 1.60GHz                                               
stepping        : 2                                                                                   
cpu MHz         : 1600.000                                                                             
cache size      : 512 KB                                                                               
physical id     : 0                                                                                   
siblings        : 2                                                                                   
core id         : 0                                                                                   
cpu cores       : 1                                                                                   
apicid          : 0                                                                                   
initial apicid  : 0                                                                                   
fdiv_bug        : no                                                                                   
hlt_bug         : no                                                                                   
f00f_bug        : no                                                                                   
coma_bug        : no                                                                                   
fpu             : yes                                                                                 
fpu_exception   : yes                                                                                 
cpuid level     : 10                                                                                   
wp              : yes                                                                                 
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx constant_tsc arch_perfmon pebs bts pni dtes64 monitor ds_cpl est tm2 ssse3 xtpr pdcm lahf_lm                                                                                 
bogomips        : 3199.91                                                                             
clflush size    : 64                                                                                   
power management:                                                                                     

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6           
model           : 28         
model name      : Intel(R) Atom(TM) CPU N270   @ 1.60GHz
stepping        : 2                                     
cpu MHz         : 1600.000                             
cache size      : 512 KB                               
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
apicid          : 1
initial apicid  : 1
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx constant_tsc arch_perfmon pebs bts pni dtes64 monitor ds_cpl est tm2 ssse3 xtpr pdcm lahf_lm
bogomips        : 3199.89
clflush size    : 64
power management:

- groovy

groovy: thanks, we made a stupid mistake that prevents the linux version from considering your configuration as eligible for multicore rendering. We will update version 3.0.4 very soon on our site. It will be interesting to see if there is a performance improvement or not !

jake: the difference is generally barely noticeable when the rate is set above 22kHz (except for the harpsichords). This is the rate at which the internal engine is generating sound, which is either the soundcard frequency, and half its freq, etc.

They are virtual cores instead of real physical ones, so they are not counted by the "cpu cores" line. Apparently, one should instead take into account the line "siblings" which is the product of the number of cores per cpu with the number of hardware thread per core.

I connected my eee-pc with USB-interfaces now and something happend, I didn't expect (?!)

First I changend audio-out from Intel HDA to the external USB-Soundcard Edirol UA-1X. The response is not as fast as before. Does the USB-SC add latency???

The same happens, when I connect my Kawai ES3 with its USB-Midi-Out to the eee. I am sure I hear more latency by doing this, than by using the Kawai's standard DIN-Midi-Out with my midi-to-usb-adapter Edirol UM-1SX. Has the USB-Out more latency than the standard DIN-Out? Or can it be, that the Edirol Midi-Interface converts the incoming midi faster to USB, than the Kawai's internal USB-interface?

The whole time I had set Pianoteq to 44kHz,64 samples so this latency remains unchanged.

The Kawai and the Edirol UA-1X are a few years old - do they have too slow chips? Or is USB in general a bottleneck for latency?

The result is, that I only have a good latency using the onboard SC and an external USB-to-MIDI-Konverter. Now I'm curious, if there are even faster converters, than the UM-1SX ... every millisecond less counts!

- groovy

The generic snd_usb_audio from kernel 2.6.29 (or is it from alsa?).

The usb-audio Edirol UA-1X is USB 1.1. It ist connected with the USB-host (UHCI, USB 1.1) of the eee-pc. As I understand, the transfer is isochronous with 12 Mbit/s from the USB-host to the USB-client and I expect no latency between those peers. I guess a latency is located between the eee-pc and its USB-Interface or between the USB-interface of the UA-1X and its DAC-conversion. Alas I could not find any specs about the latency of this USB-soundcard (or those types in general).

I examined USB and found the eee has a second USB-2.0-controller (EHCI), which is only connected with the internal webcam (480 MBit/s) and not with the 3 USB-connectors.

No other USB devices than maximal the both mentioned: UA-1X (audio) and UM-1SX (midi).

This is an interesting idea! First I unloaded the module ehci_hcd for the USB-2.0-controller. But it had not much effect (if any).  Then I disabled the (USB-)webcam in the bios and that speeds up the responsivness significantly! (but keep all this with care, it's not impossible that autosuggestion   plays a role, though I guess not).

Btw, just for interest I disabled hyperthreading in the bios, which results in that the OS just founds 1 CPU (and not cpu0 and cpu1 as before). No acceleration effect and I could drive my system to freeze playing the full-blown Grand C3.

Thanks for your answer, Jake!

Hi,

why should USB1.1 not be able to provide low latency? Is this documented somewhere?

Thanks, but it seems that link is outdated (2004/2005). I'm not aware of any other drivers/kernelmodules for USB audio-devices than snd_usb_audio under Linux. Every USB1.1 audio device should work. USB2.0 is AFAIK more problematic, because this standard is more open and sometimes firmwares have to be loaded. I never heard about latencies of snd_usb_audio, but maybe I haven't googled enough ;-)   

Troughput of USB1.1 should not be the problem, if the following is approximately right: 44,1 kHz * 16 bit = 0,7 Mbit/s. But USB1.1 provides 12 Mbit/s, which is more than enough.

My impression is that usb 1.1 is just slower for any data through-put, whether one is moving files over a usb cable or sending out midi data. That's the reason that the standard was upgraded to 2.0.  And I suspect that PianoTeq, particularly, with its high polyphony abilities, might increase the latency.

But I hope I'm wrong. I want you to be able to get this up and running. (So we can all copy your settings and have a portable PianoTeq system without having to build a rack system.)

Have you tried, by the way, using other music programs with the soundcard in the loop? Do monophonic instruments fare better in terms of latency?

I have very low latency with usb-audio now! The kernel-module("driver") snd-usb-audio has an option to process USB in shorter chunks. The default number of packs is 8 which I now set to 1:
options snd-usb-audio nrpacks=1
(in /etc/modprobe.d/alsa-base.conf to make it bootfix)

The smaller this number is, the better the latency. - Now I feel "connected" with my instrument, which is only possible with low latencies IMHO.

The only other audio-app I tested a few weeks ago was the organ-sim beatrix. Though I only had a standard debian-kernel at that time, it had good performance (and an amazing good sound). With my new RT-environment it will be pleasure, I guess.

cheers

Since we are talking netbooks/mini-pc's.... has anyone got Pianoteq to work on one using their standard config with Windows XP ?

Hmh, I'm using Alsa directly without jackd, so I can't choose the number of periods. I guess, 2 periods is ALSA's default then, when 2,9ms is the right value. Thanks for clearification, the GUI is a little confusing, but the README-LINUX.txt nails it.

Ready to play?? No, no, no. Forget all of this playing business. Show us how to set this up.

hello all of you... wow... i am very impressed and very interested in what is going on here...
i have been looking for a solution to travel with a netbook and a midi controller... just to practice a bit, do some writing, and possibly even use for some less important gigs...
so, i have a few questions... i am a TOTAL BEGINNER here... happy mac user for about 1 year only, and very frustrated PC user for the 15 years prior! never used LINUX before.

1. what is the hardware that you are using - eeepc? which model? does it matter, or does it matter more which Linux kernel? have you tried this configuration on other hardware - i.e. I AM CURIOUS TO KNOW WHICH NETBOOK I SHOULD BUY ...

2. Linux Build - which kernal are you using...? and HOW ON EARTH can i set it up so that it works as well as it does for you? i am a total beginner with this, so please be gentle!

3. what other software is required, and how can i install? i read about JACK and other patches to linux...

4. is it possible that you can compile a boot file to obtain the same settings? in the same machine? on a different machine?

you have really done alot of work here, and something that i'm sure ALL pianists would LOVE to get their hands on... you should be rewarded for your hard work!

please post a reply.
johnroney