Re: [Linuxsampler-devel] ADSDR 6th try

[Vladimir S. <ha...@so...>]

Christian,

I *think* it is 99% done. Keep in mind my inexperience with digital audio
software though. I'll clean it up this weekend and send you the result.

I'm not 100% sure it acts exactly like Gst but i tried to follow your emails
when implementing it.
I've added an ascii diagram describing the state machine. So if you don't
have time to review the code but could take a quick look at that diagram
that'd be great. Because if fundamentals are wrong it will not be worth your
time reviewing the code.

I have a few questions . . .
First one is about the "double sustain". Are we going to do that? If so, how
. . . i've hacked that in using a global and it seems to work ok, but i'd
like feedback on that.
Second question is about post attack hold. I was going to use
gig::Sample::SamplePeriod to calculate post attack hold period but it is
only 32 bits and that's in nanoseconds. 32 bits seems way too short for
nanoseconds. So i was not sure if i was missing something or not. Currently,
the post attack hold period is calculated incorrectly but i'm going to fix
it tonight or tomorrow at the latest.
Last question is about Release coefficient calculation for samples without
infinite sustain. I'm currently calculating it only once before the
trigger(). Sustain level is used in that calculation.
But if Decay2 has already lowered the level from sustain level, then the
Release coefficient is not updated and will actually produce a shorter
release phase. This sould not produce clicks because decay2 has already
lowered the level (if it hasn't lowered it enough then release phase will be
long enough).
I can't find any document to clarify if that is correct interpretation or
not. It makes sense, but if release phase must really be fixed in time
regardless of when it starts in decay2 phase, then coefficient will need to
be calculated before Release(), not before Trigger(). It's not a problem to
do it that way, but i'm just not sure what is the right way.

Mark has also reported getting some hiss and i was hoping we could
troubleshoot that to figure out if that is related to adsdr patch or not. I
have not noticed that on my setup. Unfortunately Mark is having some
problems with alsa at the moment so we can't really troubleshoot this right
now.

I'm onto command line options.

Regards,
Vladimir.

----- Original Message -----
From: "Christian Schoenebeck" <chr...@ep...>
To: "Vladimir Senkov" <ha...@fu...>; "Rui Nuno Capela"
<rn...@rn...>
Cc: <lin...@li...>
Sent: Saturday, January 31, 2004 3:24 PM
Subject: Re: [Linuxsampler-devel] ADSDR 6th try


> Es geschah am Samstag, 31. Januar 2004 03:56 als Vladimir Senkov schrieb:
> > Any other feedback on that adsdr patch? I have a semi free weekend ahead
> > and should be able to fix bugs or rewrite it completely depending on how
> > bad it is :) Or maybe clean up the code and get it ready for check in?
> > If anyone has time to review the code that would be greately
appreciated.
>
> Vladimir, Mark: Is the EG patch already finished? Means does the EG act
like
> the original Gst one? I'm currently busy, but I'll have a look at it next
> week if you think the patch is already complete.
>
> > Also, if there are no objections, i could look into those enhancement
> > requests that Mark made at the bottom of the e-mail. Should be pretty
> > straight forward i assume, but i don't want to do that if someone is
> > already looking into it :)
>
> You mean those command line options, right? If yes, I'd appreciate if
you'll
> take that task.
>
> Rui: Sorry, I'm still working on some things in the engine, but I'll start
> with the bison network interface next week.
>
> CU
> Christian
>
> P.S. these delays on this list are really annoying, maybe we should move
the
> mailing list away from sourceforge
>
>
>
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Linuxsampler-devel mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
>
>