Message-ID: <3D56A231.4F1AA5A@usa.net>
Date: Sun, 11 Aug 2002 19:43:13 +0200
From: "Alberto di Bene" <dibene@usa.net>
Organization: None
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
Subject: LF: Re: Limiter in Jason reception
References: <01c24142$3ac60c60$0eb63bd0@eng-laptop>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

What else can one do in a rainy Sunday afternoon ?
Implementing new functions in Jason, of course....:-)
Thanks David, John and others for your comments and tests.
I prepared a test version of Jason, which then, if the reports
will be positive, will be released as V0.95.

I implemented an 'Hendrixizer' (selectable either via the
Options menu or with Ctrl-H) in the following way, which
maybe is different from what you did with Cool Edit, John.

First I normalize the input samples to the range -1 ... 1
Then I compute the 50th root of each sample (preserving
the sign). Then I denormalize it again.
This, if visualized, corresponds to a very steep curve that
raises abruptly at the very beginning of the scale, then slopes
gently to horizontal. Quantitatively, a signal of -60 dB from
full scale is brought to -1.2 dB,   -20 dB to -0.4 dB, etc.
up to 0 dB which of course is mapped to itself.

I am not quite sure it has a positive result on the reception,
but then, I performed only minimal testing.  I would be very
grateful if someone could test it in real marginal reception
conditions, maybe comparing it with hard-limiting, done either
in hardware or with Cool Edit, thanks.

In addition, this pre-0.95 implements saving to a WAV file,
and saving/playback to/from a so-called JAS file, which contains
the I/Q samples after the decimator. Both savings are activated
by a couple of new buttons on the panel, which toggle between
red and black, indicating whether saving is active or not.

Reports are greatly welcome, thanks

You can download this test version from here :
http://www.weaksignals.com/bin/jaspak_v095_test.exe

73  Alberto  I2PHD