[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ARSCLIST] 2GB limit for audio file formats
> makes it possible for DSP algorithms to better distinguish noise
> from signal - the signal generated by a percussion instrument will
> look dramatically different from the noise generated by a scratch
> or crack in the record.
> You can then remove the noise without
> affecting the signal.
The noise, click, removal has nothing whatsoever to do with the actual
frequency content of the click, since the resynthesising algorithm is
looking before and after the click to determine what it should put in
there instead of the click.
> When using a Shure M44-7 cartridge, the signal from a percussion
> instrument and the noise generated by a scratch or crack look
> nearly the same, so when removing the noise you end up also
> removing a little bit of signal.
That would indicate to me that the software used by you has great
trouble synthesising a replacement for the click since when you remove
the noise after the click detection it uses the before and after audio
around the actual click to do the resynthesising.
In fact if the click detect isn´t able to detect the click due to:
> the noise generated by a scratch or crack look
> > nearly the same
Then it shouldn´t flag this part as a click since as you say they´re
nearly the same so there shouldn´t be a click there in fact.
> This is because the M44-7
> does not generate supersonic (inaudible high frequency harmonics)
> that the moving coil transducers do.
Dramatically different and Nearly the same just by going from a Shure
M44 to an Ortofon MC cartridge??
I beg to strongly differ since the actual phase response differences
between these two cartridges would be small in practice so I would
expect the wave forms to be almost the same.
Having used the Shure 500 and the Ortofon MC 2000 MK II with a claimed
frequency reponse to 60 kHz then I still see very little difference
between the actual waveforms in my Sonic Solutions DAW whether I use
44.1 or 96 kHz sampling.
What is lost is some small EHF above 20 kHz and depending on the click
detection algorithm used and it may or may not make any difference
whatsoever in fact.
> And you need the higher
> sampling rates to capture all of the noise so that the DSP
> algorithms can work better.
This may or may not be true depending on the software used.
The Mastering Room AB
Learn from the mistakes of others, you can never live long enough to
make them all yourself. - John Luther