naturerecordists
[Top] [All Lists]

Re: bass rolloff question and "head room"

Subject: Re: bass rolloff question and "head room"
From: Rob Danielson <>
Date: Sun, 15 Feb 2004 04:01:50 -0600
>In a message dated 2/13/04 6:44:03 PM Pacific Standard Time,
> writes:
>
>>  There are those that worry that sound levels will not have enough detai=
l
>>  in the lower levels, but that's quite a way down. I've cranked in
>>  30-40dB of gain into a quiet recording in post and everything I hear is
>>  still plenty good
>
>Dear Walt,
>
>The above reflects my concern. On the Portadisc, if one records at -50dB w=
ill
>the signal be sufficient to capture all the sound? I guess there is some
>point where the input is just not enough to allow the machine to digitize
>sufficient "bits" to make a good recording. Any idea where this might
>be..-60..-50..-40?
>
>My best and thanks for all the continued help.
>
>John
>John V. Moore Nature Recordings
>


Hi John.
Concrete numbers for how under-saturation affects quality,.. I did a
little test that spelled out a few things for me. Hopefully they'll
make sense to you.

I took a recording I made in the middle of the night in a rural
setting with mkh 20's->MP2-2 (fully cranked)->Tascam DA-P1. The
ambient sound level was around 20-24dB-- very quiet. The peaks in
this recording have 1% saturation or -38dB-- with pretty even
amplitude.  I computed that this file is using about 10 of the
available 16 bits.  I took one channel and made a stereo file from it
producing two identical channels. I made four copies of this.  I used
Peak (32 bit processing) and reduced the saturation of one channel
-15dB (to -53dB) and saved it. Then I normalized this truncated
channel and the control channel to -8dB peaks-- a reasonable
submastering level. I repeated this process for the other copies at
-20dB (or -58dB total); -30dB (or -68dB total) and -40dB (or -78dB
total). In theory, with the gear and ambient sound floor described
above, I couldn't have gotten less than about -60dB  saturation.

In playing the files back, there's a very noticeable, progressive
increase in hiss starting even at -53dB. About 2dB of hiss starts at
5K with the -53dB file and 6dB of hiss starting at 2K with the -68dB
file. The quality below -40dB is really aren't acceptable to my ears
and this chart made it easier for me to see why:

-38dB   10 bits resolution
-53dB   8  bits resolution
-58dB   6 bits resolution
-68dB   5 bits resolution
-78dB   3 bits resolution

In practice, I've found it very hard to obtain an acceptable degree
of transparency from sound files with 1% peaks-- that's -38dB. The
broadband hiss that is added is virtually impossible to notch out
with EQ. Its much easier to make a transparent recording from a file
that has  3% saturation or peaks to -28dB.  This is another example
why those wanting to record ambience in natural settings need clean
preamplification. A mic pre with only 50dB gain would not really be
able to kick out enough signal to capture much transparency in
natural settings below 35dB. Many of the portable usb mic pres that
have come on the scene for laptop recording have ~50dB gain.  Rob D.

  =3D =3D =3D =3D







________________________________________________________________________
________________________________________________________________________


"Microphones are not ears,
Loudspeakers are not birds,
A listening room is not nature."
Klas Strandberg
------------------------------------------------------------------------
Yahoo! Groups Links

<*> To visit your group on the web, go to:
     http://groups.yahoo.com/group/naturerecordists/

<*> To unsubscribe from this group, send an email to:
     

<*> Your use of Yahoo! Groups is subject to:
     http://docs.yahoo.com/info/terms/

------------------------------------------------------------------------



<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU