From alsa-user-owner@alsa.jcu.cz  Mon Oct 26 05:53:38 1998
Received: from io.nu (qmailr@dogbert.advancenet.net [209.44.30.54])
	by marvin.jcu.cz (8.9.1a/8.9.1) with SMTP id FAA02563
	for <alsa-user@alsa.jcu.cz>; Mon, 26 Oct 1998 05:52:10 +0100
From: sterwill@io.nu
Received: (qmail 749 invoked by uid 1000); 26 Oct 1998 10:52:06 -0000
Message-ID: <19981026105206.748.qmail@io.nu>
Subject: Turtle Beach Noise
To: alsa-user@alsa.jcu.cz, alsa-devel@alsa.jcu.cz
Date: Mon, 26 Oct 1998 10:52:06 +0000 (GMT)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Reply-To: alsa-user@alsa.jcu.cz
Sender: alsa-user-owner@alsa.jcu.cz
Precedence: list

Here's a minor update to the noisy Turtle Beach card problems that 
me and others had seen.  

The other day, while doing other internal maintenance, I opened 
up my case and re-arranged the cards.  The sound card is the very
bottom card, the video cards (Millennium II and Pure3D) are the
very top cards.  There is at least 7" between the sound and video.  
I figured that that wasn't quite enough.  :)  I took a Pepsi can, 
and cut off the ends.  I rolled the aluminum into a flat sheet, 
about 10" long and 5" wide.  I taped over the sharp edges with 
rubber electrician's tape to protect my fingers.

Carefully avoiding contact with metal peripheral card parts,
this sheet is attached to the slot cover above my card and wraps
over the entire top of the TB Malibu, around the back, and back
under.  With the case on, there is a small (indirect) non-shielded
path to the card by way of the side of the case.  I'm not expecting
total absence of noise, but maybe some improvement.  I seem to
have got it with the combination of a rearrangement and shielding,
but a little noise is still there.

Anyway, I've noticed that after playing any file through /dev/dsp
(mpg123 in this case, the regular OSS/PCM output) the noise is
bad.  Every channel is muted BUT the main volume.  After 20 seconds
the noise (from mouse movement, video updates, hard drive activity)
quiets down to about half volume.  After another 15 or so seconds,
the noise gets quieter.  It's the same patterns from the same
input methods, but it's softer.  After a minute it's almost entirely 
gone.  These times are approximately the number of seconds after the
/dev/dsp device is closed by the application.

This seems like the card performing some automatic gain control 
to reduce noise during idle times.  Has anyone heard of TB's
cards doing this for some sane, predictable reason?  I like this
card quite a bit so far, but it's horrible with headphones.  My GUS
didn't do this (but it had its own noise).

Oh yeah, the TB Malibu also has some loud clicking when the PCM
device is opened and closed, but I can live with that.

-- 
Shaw Terwilliger (twig@advancenet.net)

