[Steven Sullivan]

Please help ease my mind and tell me whether I've thoguht this through correctly..

I've been transferring my 2-channel SACDs (analog output) to PCM digital lately (so I can add
them to my personal FLAC archive). Soundcard is the Audiophile 2496, recording software is
Audition 1.0. The player is a Yamaha S2500 universal DVD player, which has a rated 'max
output voltage (0 dB) = 2 V rms' for SACD sources.

The 2496 has a rated 'analog input peak = +2 dBV'

When I set the player's internal SACD channel level trim/boosts at 0 (in the range -6 to +6),
leave all internal bass management, delay etc OFF, and feed the two-channel analog output
directly into the 2496 L/R ...the recording clips (goes into the red, flattops in the
waveform) on loud parts. This is visible both in Audition and in the monitor mixer panel of
the Audiophile 2496. (The SACD in this case was one of the Peter Gabriel remasters that was
released in the last year or two. It has no CD layer or multichannel layer, so I'm certainly
accessing the 2-channel SACD tracks).

My thought is that I'm seeing the input sensitivity of the 2496 at work here. Trimming input
level with the *soundcard* software mixer doesn't stop the clipping (it just occurs at a lower
digital level), but trimming the channel levels in the *player* menu (to ~ -5) does; then I
get normal-looking waveforms that peak at around -1 dbFS.

Am I right that the issue is the soundcard, and the player is not 'really' clipping at output
at its default channel level value (0 trim/boost)? According to handy online converters, +2dBV
= ~1.26 V rms, and thus the analog input peak limit of the 2496 is some 0.74 volts lower than
the max output of the player (2 V rms). If the soundcard had a high enough input peak limit,
I'd see the full waveform without having to trim levels in the player.....right?

Lastly, and not so important because I generally use a *digital* (HDMI or IEEE 1394)
connection for SACD playback, the only spec given for my AVR's audio input section is a line
input sensitivity/impedance of 335 mv/47 kohm -- so it takes 335 mv at input to achieve the
amps's rated power output. There is no spec given for 'analog input peak' voltage of the audio
line inputs. How could I determine if the analog output peaks of the player clip in the AVR?




___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[Arny Krueger]

"Steven Sullivan" wrote in message

Please help ease my mind and tell me whether I've thoguht
this through correctly..

I've been transferring my 2-channel SACDs (analog output)
to PCM digital lately (so I can add them to my personal
FLAC archive). Soundcard is the Audiophile 2496,
recording software is Audition 1.0. The player is a
Yamaha S2500 universal DVD player, which has a rated 'max
output voltage (0 dB) = 2 V rms' for SACD sources.

The 2496 has a rated 'analog input peak = +2 dBV'

+2 dBV is about 1.2 volts

When I set the player's internal SACD channel level
trim/boosts at 0 (in the range -6 to +6), leave all
internal bass management, delay etc OFF, and feed the
two-channel analog output directly into the 2496 L/R
...the recording clips (goes into the red, flattops in
the waveform) on loud parts. This is visible both in
Audition and in the monitor mixer panel of the Audiophile
2496. (The SACD in this case was one of the Peter Gabriel
remasters that was released in the last year or two. It
has no CD layer or multichannel layer, so I'm certainly
accessing the 2-channel SACD tracks).

Letsee - apply 2 volts to something that is guaranteed to clip at 1.2 volts.
Seems pretty predictable.

My thought is that I'm seeing the input sensitivity of
the 2496 at work here. Trimming input level with the
*soundcard* software mixer doesn't stop the clipping (it
just occurs at a lower digital level),

That's because this adjustement is in the digital domain, and follows the
converters.

but trimming the
channel levels in the *player* menu (to ~ -5) does; then
I get normal-looking waveforms that peak at around -1
dbFS.

That's called "doing it right"

Am I right that the issue is the soundcard, and the
player is not 'really' clipping at output at its default
channel level value (0 trim/boost)?

You don't know for sure unless you do something about the excessive
sensitivity of the sound card in the analog domain.

According to handy
online converters, +2dBV = ~1.26 V rms, and thus the
analog input peak limit of the 2496 is some 0.74 volts
lower than the max output of the player (2 V rms). If
the soundcard had a high enough input peak limit, I'd see
the full waveform without having to trim levels in the
player.....right?

Agreed.

Lastly, and not so important because I generally use a
*digital* (HDMI or IEEE 1394) connection for SACD
playback, the only spec given for my AVR's audio input
section is a line input sensitivity/impedance of 335
mv/47 kohm -- so it takes 335 mv at input to achieve the
amps's rated power output. There is no spec given for
'analog input peak' voltage of the audio line inputs. How
could I determine if the analog output peaks of the
player clip in the AVR?

Find a signal source with a variable output that is known to itself not
clip. Most traditional audio generators do this, because internally they
have a fixed signal source that they attenuate with a passive attenuator to
provide the desired output.