Let's suppose the following situation: I would like to connect the
Behringer ADA8000 device (8 channel mic preamp with ADAT output) and
hard disc audio recorder Fostex D160 via ADAT. ADA8000 generates 24-bits
digital signal at ADAT output, but Fostex D160 saves signal in 16-bits
resolution.
So, can you answer me:
1) Can I connect these 2 devices?
2) What kind of negative influence on the recorded signal will it cause?
> Daniel Fuchs wrote:
>
>
>>>2) What kind of negative influence on the recorded signal will it cause?
>
>
>>None, apart from the fact that the last 8 bits will simply be dropped.
>>So you get less resolution than the coverter could provide...
>
>
> Isn't that called "truncation", which leaves noise that is correlated to
> the signal, maybe offering nasties? Isn't that why we use dither?
Yes you can, but see other posts regarding "nasties".
Check-out:
http://www.behringer.com/SRC2496/index.cfm?lang=ENG I believe that it's not worth it buying something more expensive if you
are going to keep the ADA8000 in your chain. The ultramatch is about
(srp) $160 (street-price should be much cheaper).
Good luck,
Evangelos
%
Evangelos Himonides
IoE, University of London
tel: +44 2076126599
fax: +44 2076126741
"Allas to those who never sing but die with all their music in them..."
> Daniel Fuchs wrote:
>
> > > 2) What kind of negative influence on the recorded signal will it cause?
>
> > None, apart from the fact that the last 8 bits will simply be dropped.
> > So you get less resolution than the coverter could provide...
>
> Isn't that called "truncation", which leaves noise that is correlated to
> the signal, maybe offering nasties? Isn't that why we use dither?
Wouldn't that be a problem only where you actually have more than
16-bits of active signal? If the content has a noise floor within
-96dBFS then the signal should already be dithered by the recorded
noise?
Lars
--
lars farm // http://www.farm.se lars is also a mail-account on the server farm.se
aim: larsfarm@mac.com
"hank alrich" <walkinay@thegrid.net> wrote in message
news:1gs86jg.e88dee1sbqp1kN%walkinay@thegrid.net
> Daniel Fuchs wrote:
>
>>> 2) What kind of negative influence on the recorded signal will it
>>> cause?
>
>> None, apart from the fact that the last 8 bits will simply be
>> dropped. So you get less resolution than the coverter could
>> provide...
>
> Isn't that called "truncation", which leaves noise that is correlated
> to the signal, maybe offering nasties? Isn't that why we use dither?
This is clearly true in a theoretical sense.
Practically speaking, almost all real-world recordings have enough
uncorrelated noise that they dither themselves someplace around the 11-15
bit level.
> > > > 2) What kind of negative influence on the recorded signal will it cause?
> > > None, apart from the fact that the last 8 bits will simply be dropped.
> > > So you get less resolution than the coverter could provide...
> > Isn't that called "truncation", which leaves noise that is correlated to
> > the signal, maybe offering nasties? Isn't that why we use dither?
> Wouldn't that be a problem only where you actually have more than
> 16-bits of active signal? If the content has a noise floor within
> -96dBFS then the signal should already be dithered by the recorded
> noise?
Why would the recorded noise have the properties of good dither?
"hank alrich" <walkinay@thegrid.net> wrote in message
news:1gs9z38.jguhao1i6kk6dN%walkinay@thegrid.net
> Lars Farm wrote:
>> Wouldn't that be a problem only where you actually have more than
>> 16-bits of active signal? If the content has a noise floor within
>> -96dBFS then the signal should already be dithered by the recorded
>> noise?
> Why would the recorded noise have the properties of good dither?
Odds are that recorded noise at the -96 dB FS level lacks one or more of the
properties of good dither. However, recorded noise typically bottoms out at
far higher levels, almost always greater than -76 dB.
Obligatory disclaimer - the fact that there's a pretty fair chance that
recorded noise includes enough of the right kinds of noise to be good
dither, does not lessen the need to do things right - and apply good dither.
"Lars Farm" <see.bottom.of.page@farm.se> wrote in message
news:1gsavgh.1uhm0u714175lqN%see.bottom.of.page@farm.se
> hank alrich <walkinay@thegrid.net> wrote:
>
>> Why would the recorded noise have the properties of good dither?
>
> What is good dither noise? and what makes that good?
Good dither decorrelates quantization error from the signal and sampling
rate.
Quantization error will tend to otherwise become correlated with the signal
and/or sample rate, and thus become a coherent signal that the ear can
recognize as a distorted sound.
You are about to answer a thread that has been inactive for more than 6 months. If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.