Help PLEASE? (Dave Hill??)

AFSDMS wrote on 2/7/2001, 11:08 AM
Nobody replied to the question I posed and I cannot seem to
get the email support to do anything but give me ASP errors.

This was posted on the Forum and no one had any ideas. This
can make a big difference in how successful someone can be
in using the Vegas line.

I need to better understand how the internals of Vegas
Audio/Video handle the various tweaks I apply to sound.

Example: I have a track that was recorded 10 dB low, so I
Normalize it. In the track, I end up reducing the fader to -
3 dB. Within the track I have a volume envelope that
highlights certain parts.

So, are each of these volume changes handled independently,
meaning that the opportunity for adding artifacts and
quantization (sp) noise is drastically increased?

Or, are the changes kind of 'compiled' so that the combined
change for any given point in time is combined so it is
only applied once.

Naturally I've given a really simple case, since my mixes
also have track FX and one or two assignable FX busses,
some reverb, and some changes at the bus including a
possible fader tweak and, finally, the dither for
downsampling to 16-bit from my native 24-bit .wavs.

Any inside info on how Vegas works would be appreciated.

Thanks much.

Wayne Munn, ASPP

Comments

Rednroll wrote on 2/7/2001, 12:23 PM
Well it seems like you're putting a little too much tought
into this processing. If this was in the analog domain, I
would more clearly understand your concerns. Once audio is
recorded into Vegas it becomes nothing, but a number and
all your processing that you do on it is still nothing, but
a math equation. IE..1+1 still equals 2, (1+1)-2+5-3 also,
still equals 2. There's no bit round offs, no A to D
conversions, no FX's processing,No magic, just direct
volume changes....all just basic math, don't try and turn
it into a calc problem.

Brian Franz
SonyEPM wrote on 2/7/2001, 12:32 PM
We use floating point in the signal path and side step the
whole problem. Try this simple test to get a feel for
what's going on:

If you have a single Vegas audio track with the track fader
at +10dB and the master fader at 0dB it obviously distorts.
But when if you bring the master fader down to -10dB you
get normal sounding audio back, not a quieter version of
the distorted audio.

AFSDMS wrote on 2/7/2001, 1:14 PM
Brian,

Please look at the helpful reply I got from someone who
understands the issue I raised and the basic differences
between analog and digital audio. The important point is
that the internal calculations are done in floating point
thus not coming up against the problems of working in a
limited 16 or 24 bit model.

Similar issues are faced in digital imaging, an area of
technology I also use extensively and in which I review a
number of products. In digital imaging the artifacts of
applying multiple processes to images with limited
resolution, typically 16-bits per color channel, are easily
demonstrated. They must be understood of digital image
technology becomes just another way for someone to louse up
an image:-)

The digital domain is far from perfect, especially if you
don't take into consideration the inherent . Understand
those limitations and you can do remarkable things.

Wayne Munn
AFSDMS wrote on 2/7/2001, 1:23 PM
Dave,

Thanks very much for clarifying the internal operation of
Vegas. Vegas obviously has a lot of features for post-
production that make it amazing in comparison with an
analog mixer, but this capability of avoiding front-end
overload really puts it in a class by itself. I'd never
quite thought of it in those terms.

As I mentioned to Brian, it is important to understand what
is going on to get the most from the product, Vegas in this
case.

BTW, is it necessary to have a "Pay for Support" number in
order to now send a query be email? When I put in NA the
form kept returning page errors.

Thanks very much.

Wayne Munn
Rednroll wrote on 2/7/2001, 4:53 PM
Hmmmm....seems like our answers coincide with each other.
Except he said +10dB + -10dB equals 0dB....I said 1+1 still
equals 2, just like 1+1-5+3+2 Still equals 2. Maybe I
should have threw in the floating point calculation point
to make things more confusing. Sorry, I'll try and
communicate on a higher plain of thinking next time.