[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [microsound] 7 points...


On 25 Apr 2006, at 17:06, j.kurtz wrote:


25 apr 2006 kl. 09.26 skrev Frank Barknecht: 

This is trivial, of course a recording is just that: A recording, not
the original. However once you recorded something digitally, this
specific recording can be transfered without any loss of information,
which is not true in the analoge realm. But that is trivial as well.


This is only true as long as you do no processing - only act as
storage. Not even changing the volume.

4b. While there are some interesting differences between analogue
and
digital recording media, these are not all that significant. A much
more striking difference is between a standalone digital recorder
and
a computer-based DAW.


Never having worked with standalone gear, I'd be interested to know:
what is this difference?


*Good* stand-alone gear at least sometimes has the processing power
to do things "right". A DAW (without any hardware-assist) is
limited to the processor doing everything else in the workstation.
And this is not enough. Even a simple operation as chang-of-volume
(a multiply) requires application of dithering not to add
distortion. Dithering must be added at every step and every
channel. This is very expensive in terms of processing power.

Many DAW:s also fall into the trap of using floating-point formats
which not only has rounding errors but also is a format that is a
bitch to apply correct dithering to. Most floating-point based
DAW:s do not even bother - they just hope the distortion from the
rounding errors will mask the quantization errors.

It can be argued that these errors are very small in a 24-bit fixed
(or the equivalent 32-bit float) but when you add them up (many
channels with many operations on each channel) this is very audible
and tiring for the ear.

Finally good stand-alone gear almost always have better analog
circuits/powersupplies than the ones inside a computer. This makes
for a sound with superior PRaT (Pace, Rythm and Timing).


I challenge you to tell the difference between a recording mixed on a
standalone recorder and digital desk and say a mix done entirely in
Protools.

I agree that some DAW software may not use the mathematically ideal
summing but generally the imperfections and artistic variations in
peoples mixing and production decisions and the characteristics of
the source material and so seem to have a far more obvious effect on
the final production far exceeding the strive for perfection in the
mixer.

Beyond this the final listening environment is far from perfect for
99.9% of environments and is usually hampered by the poor equipment
it's being played back on.  I suspect things are improving for the
general appreciation of production and mixing with the advent of the
iPod with high quality inner hear  headphones such as the Etymotic
stuff being used by lots of people on their morning commute, none the
less the exterior noise is not entirely eliminated but at least the
reproduction isn't squashed by a poor home stereo with poor speakers
and a wooden floor.

On the downside most of the stuff listened to has probably hit an
embarrassing amount of limiting and been mashed up by at least one
slightly nasty lossy compression.  God help them if they are
listening to a compilation album where the tracks ahve been through
multiple mastering processes, I've spoken to mastering engineers
who've been sent tracks as mp3s to use on albums, these same albums
which have been bought then re-encoded using another lossy codec
before finally being listened to by the punters.

What I suspect 4b is referring to above is perhaps the most
considerable change that DAWs have made to the music we listen to,
and that's the ability to do thousands of seamless edits on every
track should the mood take us.  And often it does, in an effort to
sound better, sharper, crisper and tighter than the next record we've
perfected the art of the edit.  Four vocal tracks can be made like
one, finger noise from fretboards can be perfectly eliminated or even
chopped up and synchronized to the beat.  I can take a chorus with a
wrong word and splice in the word from the second chorus in the track
where it's correct.

Not only this, I can then take these tools and use them in ways like
BT or Richard Devine might and find a sonic landscape that was
previously unimaginable.

So, to the point, whilst I'm sure there is a difference between the
sound quality of different digital systems, i suspect what most
people will perceive as the difference, if they aren't told to be
listening for sound quality at least, is perhaps the effect of the
tools available rather than the subtlety of the errors in the design
decisions in the DACs and so on.

It always reminds me of the talk in Hi-Fi magazines about the
difference between the sound quality of alternative 200 quid cables,
quite frankly I'm sure it's fascinating that one has an 'open woody
sound with a hint of vanilla' or whatever, but firstly I have to
wonder if you're listening to the right music, and then I think about
the studios I've worked in before, some brilliant music was heard but
I never saw a 200 quid cable, save the 20 yard long 64 core stuff and
wonder if perhaps if we want to worry about something perhaps the
listening environment is more important.

Roll on the day when houses get built with some decent acoustics in
the living room.



"3. ‘Fidelity’ is a red herring in sound recording. A recording
medium only ever works tolerably well within certain parameters. You
never get back exactly what you put in. This applies to digital and
analogue recording alike."

I think this is hampered again, more by the reproduction than the
recording.

Records can be made very well indeed.   Try a binaural recording made
on your own head with high quality microphones, a good preamp and
hardly any processing.

"6. Computer software is designed according to what successive
programmers have assumed is a reasonable or rational way of working.
Their assumptions may have been wrong."

This is vague enough to be right :)

Anyway, this is my first ever post to Microsound, so hello and all
that.  Sorry it's a huge rant about audio, but there ya go, gotta
start with something!

cheers, J.





---------------------------------------------------------------------
To unsubscribe, e-mail: microsound-unsubscribe@xxxxxxxxxxxxx
For additional commands, e-mail: microsound-help@xxxxxxxxxxxxx
website: http://www.microsound.org