linuxaudio.org

--_91806bda-31d0-4852-8398-3abd0e223d60_

Content-Type: text/plain; charset="iso-8859-1"

Content-Transfer-Encoding: quoted-printable
Agree on the whole that summing is summing is summing. Most DAWs these=20

days will use floating point (either 32bit or double precision) and=20

A+B=3DC no matter what.
There may be dithering at the end but you are talking a single bit=20

possible difference between renders at points=2C which I don't believe=20

anybody is going to hear. I personally find claims that one dither=20

algorythm sounds better than another=2C especially at 24bits=2C can be=20

nothing but salesman talk.
Now how the DAW loads and converts your samples (eg mixing 44.1kHz and=20

48kHz samples within the same project=2C or using a different project same

 rate to all your samples) or if you are using an internal sampler to=20

play at different frequencies=2C does make a difference.  For example=20

compare Cubic vs Sinc (SinX/X) Interpolation and you very well might=20

start to hear differences (there are more methods but these are a couple

 of the most common.) You may also find that one software uses Cubic on=20

play and Sinc on rendering (as cubic is less processor intensive) and=20

thus there are subtle difference between them whereas another may use=20

Sinc at all times. While a lot of people disregard this as it's not part

 of the summing itself I argue it is part of the audio engine.
But every time I have ever come across the X sounds great=2C Y sounds=20

pants claims by people and then requests for examples to illustrate this

 they have either never materialised or the protagonist has ended up=20

backing down from their claim. Especially on the rendered waveforms=2C due

 to the fact I mentioned above of some DAWs only using the better=20

quality interpolation at that point.
As mentioned the characteristics of the internal effects and DSPs used=20

are going to have a lot more affect on your sound than any of this ever=20

will!
On the whole any differences that do exist are so small I really wouldn't w=

orry about them.
Dale.
 		 	   		  =
--_91806bda-31d0-4852-8398-3abd0e223d60_

Content-Type: text/html; charset="iso-8859-1"

Content-Transfer-Encoding: quoted-printable
Agree on the whole that summing is summing is summing. Most DAWs these=20

days will use floating point (either 32bit or double precision) and=20

A+B=3DC no matter what.
There may be dithering at the end but you are talking a single bit=20

possible difference between renders at points=2C which I don't believe=20

anybody is going to hear. I personally find claims that one dither=20

algorythm sounds better than another=2C especially at 24bits=2C can be=20

nothing but salesman talk.
Now how the DAW loads and converts your samples (eg mixing 44.1kHz and=20

48kHz samples within the same project=2C or using a different project same

 rate to all your samples) or if you are using an internal sampler to=20

play at different frequencies=2C does make a difference.&nbsp=3B For exampl=

e=20

compare Cubic vs Sinc (SinX/X) Interpolation and you very well might=20

start to hear differences (there are more methods but these are a couple

 of the most common.) You may also find that one software uses Cubic on=20

play and Sinc on rendering (as cubic is less processor intensive) and=20

thus there are subtle difference between them whereas another may use=20

Sinc at all times. While a lot of people disregard this as it's not part

 of the summing itself I argue it is part of the audio engine.
But every time I have ever come across the X sounds great=2C Y sounds=20

pants claims by people and then requests for examples to illustrate this

 they have either never materialised or the protagonist has ended up=20

backing down from their claim. Especially on the rendered waveforms=2C due

 to the fact I mentioned above of some DAWs only using the better=20

quality interpolation at that point.
As mentioned the characteristics of the internal effects and DSPs used=20

are going to have a lot more affect on your sound than any of this ever=20

will!
On the whole any differences that do exist are so small I really wouldn't w=

orry about them.
Dale.

=
--_91806bda-31d0-4852-8398-3abd0e223d60_--