Norm said:
I have two CDRoms. Recently I started getting intermittent
static during playback of music CDs. I took my box to
a very good tech because I suspected a loose connection
somewhere. But he found nothing amiss with the connections.
Notes:
1. It occurs on both CDs, which are different brands (One
is a Sony, the other is LG.
2. Music ripped from CDs in either drive has no static and
mp3s/waves play fine with no static.
3. I have a CD cleaner CD which both cleans the lens and
applies a gaussing effect to reduce/remove static
interference, no effect.
4. My soundcard is a SoundBlaster XFi Extreme PCI card.
Anyone seen something similar?
There are several ways to play CDs.
One is via a cable with four pins on the end. It runs from the back
of the CD drive, to the sound card.
Modern computers also support DAE or digital audio extraction.
That first appeared on optical drives with IDE interfaces.
You can select that option in Windows. The music is read as
blocks of data over the cable, and fed digitally to the sound system.
There should be no opportunity for noise (barring digital mixing of
a noise source coming from somewhere).
http://www.farstone.com/images/support/vdr-cd-rom.gif
A third method, is I've seen a two wire interface on the drive,
and the claim is, it's a digital interface. It's supposed to be
SPDIF, implying operation at around 6MHz, but I've never been
able to find a spec for it. SPDIF normally would be carried on
a coaxial cable, and be relatively low amplitude and controlled
impedance. The two pin interface on the drive, could be a
"TTL level" or digital signal kind of output (that would be
higher amplitude than SPDIF), without the nice properties of the
SPDIF coaxial output. I've never run into someone who has succeeded
in connecting it up
I don't have any sound systems, with an input
for it. And if the intention was to connect it to SPDIF_in, then an
adapter would be needed of some sort. And I've never seen mention
of an adapter either.
*******
My guess would be, you're using the analog method. Switch to
DAE and try again. The analog method with the four pin connector
(typically a white connector body), can pick up noise, and
the quality of the thing receiving the signal, may not be that
good (it could be pseudo-differential and have poor common mode
rejection).
In addition to the analog cable picking up noise, a second
noise source is the hardware mixer in the sound chip. If two
inputs are enabled at the same time, one being the CD analog,
the other being some other input like microphone, if the
gain is cranked on the microphone, it could pick up digital
noises from the motherboard. In some cases, you can hear
"mouse noises" when the mouse moves. Or, you can hear sound
whenever packets are sent over the Ethernet RJ-45 interface.
Now, people don't call that "static", that's more of a
"tone" or "musical burbling".
"Clicks and pops" in playback, can come from an "underrun" when
digital samples aren't delivered to the sound card in time,
and its buffer runs dry. On older systems, it might be
"Delayed Transaction" that needs to be enabled in the BIOS,
or perhaps "PCI Latency" needs to be dropped to a lower
value (like perhaps 16). I had one system, where the thing
was *very* picky about PCI Latency, and only one value came
close to working. If you drop PCI Latency too low (I tried it),
the system becomes sluggish from the additional bus traffic.
Modern systems just don't seem to have those problems any more,
as Delayed Transaction is enabled and there is no disable
for it in the BIOS. And PCI Latency no longer seems to be
a problem. I can use 32 or 64, without consequence. Some
people use a setting of 128, when doing I/O benchmarks to
impress their friends, and high settings allow a PCI card to be
more of a "bus hog". When there are devices with real time
requirements, using such a high setting is a mistake, because
the real time device will suffer occasional starvation (underrun).
*******
So the first part of debugging this kind of problem, is
deciding what kind of noise it is. And that's pretty
hard to communicate over USENET.
Paul
