PS: make that ...superior technique, money, or a soldering iron. I
followed up with Mike Rooke and will soon be doing some experiments that
should hold the budget down.
Another rabbit hole to go down!
Thanks Mike!
On Sun, Dec 9, 2012 at 10:42 AM, Keith Smith <> wrote:
> At this point, we might do well to re-read some early posts in this
> thread, such as David and Dan, below.
>
> Dan, I'm still working on my response to your plea to avoid noise
> reduction -haven't forgotten you at all.
>
> I'm presently accumulating some data using David's 'under the bed clothes=
'
> test with the PMD 661 and various transducers at gain settings from 0 to
> 10. So far I have an NTG1 and a peizo crystal(with Jensen DI) on a tin ca=
n
> top. Both yield -80 to -81 dBFS at a gain setting of 0. I thought the Pei=
zo
> would do much better. Am I hitting the published 65dB S/N wall? I'll
> carry on with some quieter mics I have this week (with a spreadsheet to
> follow), but I'm thinking that actually using more than 16 bits is a
> challenge solved only by superior technique and money.
>
> Keith
>
>
> On Tue, Dec 4, 2012 at 12:26 AM, Dan Dugan <> wrote:
>
>> **
>>
>>
>> > With very quiet backgrounds and low noise pro mics and inputs or
>> > mixers, 24 bit recording theoretically gives you another 8 bits, which
>> > is nice but probably difficult to justify. I say "theoretically" as
>> > the dynamic range at 24 bits is 23 times 6dB, or 138dB which is much
>> > greater than any mic input stage could ever come near. With studio
>> > mixers, you don't want to worry about noise, so 24 bit is used, but
>> > with wildlife recordings, this increases file sizes.
>> >
>> > Thermal noise of the air is around 0phons (it's frequency dependant)
>> > and the best pro mic systems will have an "excess noise over thermal"
>> > of around 12 dB. Ear damage starts around 90 phons, so if you need to
>> > record this level without adding noise, use 24 bits, but you won't
>> > find a listening setup which can reproduce this range. :-)
>>
>> 24-bit codes are written, but the actual signal-to-noise in the real
>> world is 21 bits maximum due to the analog parts of the signal chain.
>> Fade-outs in a DAW will use all the bits, which is why you must add dith=
er
>> noise at the very last stage when reducing to 16 bits for CD distributio=
n.
>>
>> I use 24-bit recording because it gives you extra headroom to work with.
>> You can record low for safety in the field and boost it in post without =
any
>> loss of quality. Now that storage is larger and cheaper, the larger file
>> size doesn't matter much.
>>
>> -Dan
>>
>>
>>
>
>
>
> --
> Keith Smith
>
> Keith Smith Trio, Northern Lights =96 Altai Khangai - www.keithsmith.ca
> Photography - www.mymountains.ca
>
>
--
Keith Smith
Keith Smith Trio, Northern Lights =96 Altai Khangai - www.keithsmith.ca
Photography - www.mymountains.ca
|