[LAU] Linuxsampler/qsampler usability and session portability

Philipp Überbacher hollunder at lavabit.com
Fri Oct 14 15:59:46 UTC 2011


Excerpts from Dan MacDonald's message of 2011-10-14 17:45:41 +0200:
> On Fri, Oct 14, 2011 at 1:19 PM, Rui Nuno Capela <rncbc at rncbc.org> wrote:
> 
> > On 10/14/2011 12:40 PM, Dan MacDonald wrote:
> >
> >> I addressed the list yesterday regarding my dissatisfaction wrt the
> >> current state of soft samplers under Linux. Since then I have checked
> >> out the latest svn versions of Linuxsampler and qsampler just to make
> >> sure nothing had dramatically changed since I last tried them. I'd much
> >> rather we fix up LS and qsampler to be usable in all cases rather than
> >> anyone, such as myself, go starting a whole new project. I was thinking
> >> about posting this to the LS mailing list (if there is one) but the
> >> problems I've encountered today involves a wider range of Linux audio
> >> projects than just LS (such as LV2 and Ardour) so I thought it best to
> >> discuss these issues here.
> >>
> >> I have two goals I'd like to see achieved:
> >>
> >> 1 - You should be able to import wavs into qsampler and create/use
> >> simple .sfz files without opening/editing/manually creating any text
> >> (.sfz) files. This should be easily achieved by adding a new menu option
> >> to qsampler called something like 'Import audio files' or 'Create sfz'.
> >> In most cases you just need to specify the soundfile and the note or
> >> range that will trigger it. Comments Rui?
> >>
> >>
> > in case you didn't notice, qsampler is a _front-end_ to linuxsampler. that
> > definition also applies to the other cases, like jsampler (aka. fantasia)
> > and gsampler.
> >
> > the actual sampler engine is linuxsampler which also acts like a server to
> > any of its front-ends (clients) via lscp (linuxsampler control protocol) a
> > simple line text protocol over plain tcp/ip.
> >
> > linuxsampler front-ends like qsampler only "talks" lscp, they don't know
> > nothing about jack, midi nor sound-files, whatsoever. it is linuxsampler
> > that does all the dirty work, even if its being redirected via its plugin
> > interface.
> >
> > however, linuxsampler as an engine does not provide any interface to edit
> > sample files be that .gig or .sfz . you edit those files through an
> > appropriate stand-alone application (eg. gigedit) and then tell linuxsampler
> > to load the product into its engine. that's it.
> >
> 
> Yes, I know what a front end and what LS does and I really don't think my
> request/suggestion was out of place. Far as I know gigedit just does that -
> edits gig files and has nothing to do with sfz. You need to realise I wasn't
> requesting anything like a full gigedit clone but for sfz, all I wanted was
> a most basic sfz creation feature that would simply let you pick files,
> assign them to notes or ranges and create a matching .sfz - thats all!
> 
> > you can now guess the answer is no ;)
> >
> 
> Yep! Well at least I tried.
> 
> Mr Prokoudine:
> 
> Ah! I didn't know about gsampler- thanks for highlighting that. It compiles
> but I'm not sure it actually works yet. I think I may get in touch with its
> author and see if he is willing to integrate a very simple sfz creation
> function into his app.
> 
> 
> Mr Dyer:
> 
> Thanks for your example of a simple sfz but that doesn't work for me under
> either qsampler or fantasia, specifying the full path to the sample or not,
> putting all the commands on one line or separate lines like you did- I just
> cannot get my basic sfz to load. I'm running very latest svn versions of all
> the LS apps btw.
> 
> 
> Everybody:
> 
> Paul says that we can store samples in a random subdir of an A3 session and
> I have successfully got LS to load and play soundfonts but it looks like A3
> has trouble loading such a session without moving anything atm so I've not
> been able to test what happens when I move the session just yet.

Recently a new release of synthclone was announced here, and as far as I
understood it it's a tool to create soundfonts in sfz, gig, etc.. Have a
look, maybe it works for you: https://code.google.com/p/synthclone/

Regards,
Philipp



More information about the Linux-audio-user mailing list