Could use some advice. I just ordered the Onyx 400F 10x10 firewire
preamp/mixer/interface for my Shuttle XP daw. I plan to do some classical
piano recording using a four mic surround setup, and want to record at 192k.
I expect to be doing more of this in the future, including some live work.
I have done a lot of classical and jazz recording in the past (tape and
DAT); my experience with pop multitracking is minimal. I'm not enamored
with the Traktion 2 software that Mackie ships with the
preamp/mixer/interface.

I would like the software to cover simultaneous multi-channel real-time
recording, fairly extensive editing/splicing, perhaps some minor time
shifting of the surround tracks, and preparation of tracks for final
mastering, with the option of doing the mastering into DVD-A and Redbook
format myself. .

I'd appreciate your advice/input on which software you think best for this
purpose. I will have the Traktion, of course. And I could use Audacity, but
it won't do 192k. What beyond those two are most suitable for my purpose,
and what are their strengths and weaknesses?

Harry Lavo wrote:
Could use some advice. I just ordered the Onyx 400F 10x10 firewire
preamp/mixer/interface for my Shuttle XP daw. I plan to do some classical
piano recording using a four mic surround setup, and want to record at 192k.

I would like the software to cover simultaneous multi-channel real-time
recording, fairly extensive editing/splicing, perhaps some minor time
shifting of the surround tracks, and preparation of tracks for final
mastering, with the option of doing the mastering into DVD-A and Redbook
format myself. .

As an occasional Sequoia user, I'd suggest their next level down,
Samplitude. I would also recommend that you not push the 400F to 192
kHz. I haven't tried one at that speed, but conventional wisdom is that
a 192 kHz A/D converter that sounds better than a 96 kHz converter can
not be built from conventional components and therefore cannot be as
inexpensive as the 400F. Stick with
96 kHz and you'll have it as good as it gets.

The only advantage of 192 kHz over 96 kHz is that you can record what's
above
48 kHz. That really isn't of much use in music recording. Even if your
microphones had response up that high and all that high frequency
energy wasn't immediately absorbed by carpets, curtains, and the air
between the piano and microphone, there's nothing that can reproduce
it. You're just wasting disk space and sacraficing linearity. 192 kHz
sample rates exist today because the marketing departments needed
something new to advertise and it was cheap and easy.

"Chel van Gennip" wrote in message
...
On Sat, 22 Oct 2005 03:43:39 +0200, Harry Lavo wrote:

Could use some advice. I just ordered the Onyx 400F 10x10 firewire
preamp/mixer/interface for my Shuttle XP daw. I plan to do some
classical piano recording using a four mic surround setup, and want to
record at 192k. I expect to be doing more of this in the future,
including some live work.

Why do you want surround and 192K?

BTW, I use Samplitude and it seems to have quite a lot of surround
features, although I never used them. My version goes to 96K but newer
versions do 192k too.


Thanks, Chel. Exactly the kind of feedback I hoped for to create my "short
list".

"Mike Rivers" wrote in message
oups.com...

Harry Lavo wrote:
Could use some advice. I just ordered the Onyx 400F 10x10 firewire
preamp/mixer/interface for my Shuttle XP daw. I plan to do some
classical
piano recording using a four mic surround setup, and want to record at
192k.

I would like the software to cover simultaneous multi-channel real-time
recording, fairly extensive editing/splicing, perhaps some minor time
shifting of the surround tracks, and preparation of tracks for final
mastering, with the option of doing the mastering into DVD-A and Redbook
format myself. .

As an occasional Sequoia user, I'd suggest their next level down,
Samplitude. I would also recommend that you not push the 400F to 192
kHz. I haven't tried one at that speed, but conventional wisdom is that
a 192 kHz A/D converter that sounds better than a 96 kHz converter can
not be built from conventional components and therefore cannot be as
inexpensive as the 400F. Stick with
96 kHz and you'll have it as good as it gets.

The only advantage of 192 kHz over 96 kHz is that you can record what's
above
48 kHz. That really isn't of much use in music recording. Even if your
microphones had response up that high and all that high frequency
energy wasn't immediately absorbed by carpets, curtains, and the air
between the piano and microphone, there's nothing that can reproduce
it. You're just wasting disk space and sacrificing linearity. 192 kHz
sample rates exist today because the marketing departments needed
something new to advertise and it was cheap and easy.

Thanks, Mike. Good feedback. Your comments converter quality are noted
and appreciated. I'll of
course listen both to 96k and 192k once I receive the 400F, using my own
mics (I may have to upgrade a pair). But my reason for the interest for
192k is not extended high-frequencies, but rather the fact that it is not
until this level of frequency that impulse response pre-echo becomes
theoretically inaudible. Disk space/processing power is not an issue for me
with only four channels recorded. My own listening to commercial releases
suggests to me that 192k "sounds" slightly more natural than 96k, and SACD
even more so. In reality, the impulse response and more natural
transient-attack is the only reason to which I can ascribe this
difference....but I have been recording chamber music off and on since the
late '60's and
I know "natural" sound when I hear it. I'll be recording a wonderful 9'
Steinway in a superb acoustic, and transient response will be an important
factor in the sound, I believe. To my ears so far, 192k and SACD
come closer than anything else in capturing accurate transient response..
I'd be recording in SACD if the software
were readily available. Ultimately, however, my own tests will determine
what I do. It is possible I'll take your advice and downgrade to 96k if the
converters are a problem, but I
don't want the software to limit me in this regard.

As to my interest in surround, I am a surround advocate, and especially for
chamber music, solo classical (as for the current project), and for jazz
ensembles where it is possible to recreate a credible "you are there"
in-room experience. I have always used purist mic'ng techniques, and plan
to do so to capture and prepare intimate surround recordings in DVD-A and/or
SACD.

Obviously, these disks will not be aimed at the mainstream. They will be
marketed as niche products to classical and jazz music lovers.

"Chel van Gennip" wrote in message
...
On Sat, 22 Oct 2005 17:01:32 +0200, Harry Lavo wrote:

But my reason for the interest for 192k is not extended
high-frequencies, but rather the fact that it is not until this level of
frequency that impulse response pre-echo becomes theoretically
inaudible. ..... My own listening to commercial releases suggests to me
that 192k "sounds" slightly more natural than 96k, and SACD even more
so.

I thought pre-echo's and transient problems are not a PCM problem, but
problems of formats like Mpeg and DSD ( as used in SACD)


No, actually SACD is better than PCM in this regard, but at 192k the
difference is small enough to be negligible.

Harry Lavo wrote:
But my reason for the interest for
192k is not extended high-frequencies, but rather the fact that it is not
until this level of frequency that impulse response pre-echo becomes
theoretically inaudible.

The "pre-echo" from the anti-aliasing and reconstruction
filters looks like HF ringing in the time domain but that's
not at all what it sounds like. That pre stuff is
necessasary to keep the response of the anti-alias filters
linear phase, i.e. having frequency independant group delay.
It could be eliminated by the use of a minimum phase
filter but that would alter the phase of the converted
signal in a frequency dependant way.

I'm not sure where the idea comes from that this "pre-echo"
has negative sounding artifacts. It is required for a
linear phase low pass.

Spikey things in the impulse response are what yield echo
sounding artifacts. Beyond that, looking at the form of an
impulse response in the time domain tells you almost nothing
about what it will sound like when used as a filter.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

"Bob Cain" wrote in message
...


Harry Lavo wrote:
But my reason for the interest for
192k is not extended high-frequencies, but rather the fact that it is not
until this level of frequency that impulse response pre-echo becomes
theoretically inaudible.

The "pre-echo" from the anti-aliasing and reconstruction filters looks
like HF ringing in the time domain but that's not at all what it sounds
like. That pre stuff is necessary to keep the response of the
anti-alias filters linear phase, i.e. having frequency independant group
delay. It could be eliminated by the use of a minimum phase filter but
that would alter the phase of the converted signal in a frequency
dependant way.

I'm not sure where the idea comes from that this "pre-echo" has negative
sounding artifacts. It is required for a linear phase low pass.

Spikey things in the impulse response are what yield echo sounding
artifacts. Beyond that, looking at the form of an impulse response in the
time domain tells you almost nothing about what it will sound like when
used as a filter.


Months and years before I ever became aware of PCM pre-echo, or that 192k
pcm
had impulse pre-echo that fell within the window of audible discrimination,
I
knew that SACD and high-res PCM sounded better than CD, and that it seemed
to be in the area of transient realism that the differences lay. So I may
be wrong in attributing what I hear to the this aspect (although I don't
think so). I suspect that it is more audible than has been traditionally
thought, and that it is one of the things some people react "badly"
to in CD sound. It is after all "un-natural" in that it does not occur
anywhere in nature....and our ear/brain complex has been highly conditioned
by eons of evolution attuned to the natural world. It is also possible that
it is a "learned" ability to hear it, as are most other artifacts.


For what it is worth, here is just the latest reference to the issue I have
run across:


"Aside from the sampling frequency.s ability to create a large frequency
response
in a digital system, this digital audio parameter also affects several other
elements of

encoded audio, particularly the stereo localization and impulse response.
These two

attributes are tied together because they are both affected by the
time-domain spread

created by the sampling rate. Since the sampling process is essentially a
quantization of

samples on a time scale rather than on an amplitude scale, signals occurring
between

those instances are cropped. In a 44.1 kHz system the time between each
sample is fairly

large, with approximately 22.68 Us between each sample. On the other hand,
192 kHz

PCM samples are only 5.1 Us apart, and DSD has an even greater time domain
resolution

with a mere .357 Us between samples. As a result, the 192 kHz PCM and DSD
systems

are better suited to respond to transients accurately. Figure 16
demonstrates how this

response time actually affects the system.s ability to reproduce a
transient, as each system

was fed a .6 dB block input (click) of a 3 µs duration. The resulting graph
shows that the

DSD and 192 kHz PCM systems respond the fastest and most accurately; whereas
the 48

kHz PCM system not only distorts the signal, but also takes a much longer
time to even

react. This distinction, most visible in the large width of the 48 kHz
sampling frequency

reproduction, is audibly apparent in transient events as a ringing or
..bell-like. sound.

The binding factor between a digital system.s transient response and stereo

localization abilities is a psychoacoustic principle relating that .most
people can hear a

time delay of 15 milliseconds or more. (Moorer 1). If two sounds are played
more than

15 milliseconds apart, they will be audibly perceived as two distinctly
separate sounds.

Obviously a 44.1 kHz system does not have a large enough sample rate time to
utilize

this principle to its fullest extent, as seen in the smearing of the audio
reproduced in a 48

kHz system, causing an imaging issue that results in a blurring of the
stereo soundstage.

This is a direct result of the same factors that cause the .ringing. sound
in transient

events. In contrast to the extremely inaccurate width of the 48 kHz
reproduced sample

seen in Fig. 16, both DSD and 192 kHz PCM are able to accurately reproduce a
3µs click

almost instantaneously, without any blatantly apparent inaccuracies."

Source: Honors research project done in the Fall of 2004 by Brandon M.
Schexnayer. It is called "Reshaping Digiital Audio: DSD Encoding as a
Viable Alternative to PCM.". It can be found at:
http://gearslutz.com/board/attachmen...achmentid=6491

Harry Lavo wrote:
Months and years before I ever became aware of PCM pre-echo, or that 192k
pcm
had impulse pre-echo that fell within the window of audible discrimination,
I
knew that SACD and high-res PCM sounded better than CD, and that it seemed
to be in the area of transient realism that the differences lay. So I may
be wrong in attributing what I hear to the this aspect (although I don't
think so).

The problem is that you can't really construct an accurate experiemnt.
A
48 kHz converter is a different animal from a 96 kHz converter, which
is a different animal from a 192 kHz converter, and DSD is different
from all of them. You can't just change the sample rate and compare
them. Well, I guess you could if you used the same filters and
restricted the bandwidth so that you would't have aliasing at the
lowest sample rate, but that wouldn't prove what you wanted to prove.

If you have a converter that can be set to different sample rates, even
though it's the same converter, there are different components and
different algorithms used when you change sample rates. So you're
always comparing Winesaps to Macintoshes.

There's nothing wrong with a subjective listening test and an opinion,
but what you hear that you think sounds better may be an inaccuracy
that you just happen to like.

Harry Lavo wrote:

Months and years before I ever became aware of PCM pre-echo, or that 192k
pcm
had impulse pre-echo that fell within the window of audible discrimination,
I
knew that SACD and high-res PCM sounded better than CD, and that it seemed
to be in the area of transient realism that the differences lay. So I may
be wrong in attributing what I hear to the this aspect (although I don't
think so). I suspect that it is more audible than has been traditionally
thought, and that it is one of the things some people react "badly"
to in CD sound. It is after all "un-natural" in that it does not occur
anywhere in nature....and our ear/brain complex has been highly conditioned
by eons of evolution attuned to the natural world. It is also possible that
it is a "learned" ability to hear it, as are most other artifacts.

Consider that a perfect brick wall low pass filter such as
would ideally be used as an anti-alias preconditioning
filter or as the reconstruction filter on the output end has
an infinite amount of pre-ringing and post-ringing.

What is passed by such a filter is all frequencies up to the
cutoff frequency with no magnitude modification at all.
Above that cutoff all frequencies are exactly blocked. The
filter has the property of introducing only a fixed delay
into the signal path. All frequencies that are passed have
the exact same phase relationship coming out as going in.
What artifacts would such a component introduce other than
chopping off everything above the cutoff?

In practice, since infinity is a long time, these filters
are windowed, that is they are multiplied by a function
which smoothly truncates both the pre and the post at some
practical length. This has the effect of introducing some
slight rolloff and some slight frequency dependant phase
modification, moreso as you get closer and closer to the
cutoff. By reducing the "ringing" you degrade the performance.

For what it is worth, here is just the latest reference to the issue I have
run across:

With all due respect, Harry, there is a great deal of
nonsense in what you quoted from Schexnayer. About the only
accurate thing is that with higher sample rates, transient
response is faster. This is a direct consequence of the
wider bandwidth and does not imply that you can hear beyond
the lower bandwidth to appreciate the difference in speed.
If the higher transient rate requires bandwidth beyond the
ear's, no difference will be heard. This has been
demonstrated. I think Arny has some demonstrations of it on
his site that you can ABX yourself.

There are, in fact, some negative consequences of using a
linear phase (pre-ringing) filter for signifigant low end
contouring but that's a different matter.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

On Sat, 22 Oct 2005 18:59:18 -0700, Bob Cain
wrote:

Harry Lavo wrote:

Months and years before I ever became aware of PCM pre-echo,
or that 192k pcm
had impulse pre-echo that fell within the window of audible discrimination,

refutation snipped but saved for study

There are, in fact, some negative consequences of using a
linear phase (pre-ringing) filter for signifigant low end
contouring but that's a different matter.

But, but, but... this is the interesting part!
No fair,

Chris Hornbeck
Gen. Miller, Gen. Sanchez, Donald Rumsfeld, President Bush.

Chris Hornbeck wrote:
On Sat, 22 Oct 2005 18:59:18 -0700, Bob Cain
wrote:


Harry Lavo wrote:


Months and years before I ever became aware of PCM pre-echo,
or that 192k pcm
had impulse pre-echo that fell within the window of audible discrimination,


refutation snipped but saved for study

There are, in fact, some negative consequences of using a
linear phase (pre-ringing) filter for signifigant low end
contouring but that's a different matter.


But, but, but... this is the interesting part!
No fair,

Essentially, if you are using a linear phase filter to boost
the low end of a drum, for example, after the filtering you
will find lots of low frequency energy in the waveform
starting signifigantly before the percusive hit that should
start the thing off. I think I can hear this in comparison
to a minimum phase filter with the same magnitude response
where the boosted LF is all after the main hit. I haven't
taken the care to do it ABX, though, just AB blind, and
should. What I hear, in general, with use of linear phase
filters in the mid and lower band for contouring seems to be
a dispersion in time but since that's what my eyes tell me
from the waveform, the impression may be contrived by my
brain to match my eyes.

Nature doesn't give us any linear phase frequency response
contouring. Rooms (reflections and delays) give mixed phase
response and most physical systems are close to minimum
phase. I think the ear/brain prefers minimum because nature
is biased that way.

Had I been in an ivory tower doing research, one of the
things I'd like to have done is develop a metric for phase
minimality, maybe fast factoring to see where the zeros are
on the complex Z plane (they should all be inside the unit
circle for minimality), to get a better look at where things
like mics, speakers and rooms really are in that metric space.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

On Sun, 23 Oct 2005 14:51:22 -0700, Bob Cain
wrote:

Nature doesn't give us any linear phase frequency response
contouring. Rooms (reflections and delays) give mixed phase
response and most physical systems are close to minimum
phase. I think the ear/brain prefers minimum because nature
is biased that way.

I had to stop reading comp.dsp because it seemed
that nobody there understood this, and I'm too low
on the food chain to change the viewpoint.

Had I been in an ivory tower doing research, one of the
things I'd like to have done is develop a metric for phase
minimality, maybe fast factoring to see where the zeros are
on the complex Z plane (they should all be inside the unit
circle for minimality), to get a better look at where things
like mics, speakers and rooms really are in that metric space.

Whoosh! Saved for possible future understanding.

Thanks, as always,

Chris Hornbeck
Gen. Miller, Gen. Sanchez, Donald Rumsfeld, President Bush.

"Bob Cain" wrote in message
...


Chris Hornbeck wrote:
On Sat, 22 Oct 2005 18:59:18 -0700, Bob Cain
wrote:


Harry Lavo wrote:


Months and years before I ever became aware of PCM pre-echo,
or that 192k pcm
had impulse pre-echo that fell within the window of audible
discrimination,


refutation snipped but saved for study

There are, in fact, some negative consequences of using a linear phase
(pre-ringing) filter for signifigant low end contouring but that's a
different matter.


But, but, but... this is the interesting part!
No fair,

Essentially, if you are using a linear phase filter to boost the low end
of a drum, for example, after the filtering you will find lots of low
frequency energy in the waveform starting signifigantly before the
percusive hit that should start the thing off. I think I can hear this in
comparison to a minimum phase filter with the same magnitude response
where the boosted LF is all after the main hit. I haven't taken the care
to do it ABX, though, just AB blind, and should. What I hear, in general,
with use of linear phase filters in the mid and lower band for contouring
seems to be a dispersion in time but since that's what my eyes tell me
from the waveform, the impression may be contrived by my brain to match my
eyes.

Nature doesn't give us any linear phase frequency response contouring.
Rooms (reflections and delays) give mixed phase response and most physical
systems are close to minimum phase. I think the ear/brain prefers minimum
because nature is biased that way.

Had I been in an ivory tower doing research, one of the things I'd like to
have done is develop a metric for phase minimality, maybe fast factoring
to see where the zeros are on the complex Z plane (they should all be
inside the unit circle for minimality), to get a better look at where
things like mics, speakers and rooms really are in that metric space.



Like you, I have not done a rigorous test, but yesterday morning I did do
some blinded listening and informally tried to identify the CD layer and
2-Ch SACD layer on my C222ES. This is not instant switching, but takes
about 5 secs. It was difficult to identify, but I did get about 2/3rd's
right out of a total of perhaps 20 switches...I was not doing this
rigorously and can only approximate counts. Also, I was able to see after
each guess whether I was right or wrong, which is not good protocol.. I was
listening to bass drum and toms and what I think I heard is a slightly
"heavier" and "hollower" sound on CD than with SACD, which seemed "tauter".
Also, I think the CD image was a little vaguer (although I am less sure of
this than the "hollowness").

These observations seem somewhat consistent with your comments above, even
though you seem to dismiss the possibility that your low frequency filter
observation has anything to do with what I ascribe to SACD..

Harry Lavo wrote:

These observations seem somewhat consistent with your comments above, even
though you seem to dismiss the possibility that your low frequency filter
observation has anything to do with what I ascribe to SACD..

I just think there is a signifigant difference between what
"brick wall" low pass filters used for anti-aliasing do at
the level of perception and what linear phase filters used
for equalization (contouring) do in the low and mid band.

A simple point, look at the frequency of the "ringing" in
the low pass filters that differentiate the cases you are
concerned with. It is very close to the cutoff. I don't
think that can be heard as ringing or time dispersion
because it is generally beyond the ear's bandwidth and is of
very short duration.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

"Harry Lavo" wrote:

[...] I'll be recording a wonderful 9' Steinway in a superb acoustic,
and transient response will be an important factor in the sound, I
believe.


Whacha gonna use for mics?

--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good

(Remove spamblock to reply)

"Harry Lavo" wrote in message


Could use some advice. I just ordered the Onyx 400F
10x10 firewire preamp/mixer/interface for my Shuttle XP
daw. I plan to do some classical piano recording using a
four mic surround setup, and want to record at 192k. I
expect to be doing more of this in the future, including
some live work. I have done a lot of classical and jazz
recording in the past (tape and DAT); my experience with
pop multitracking is minimal. I'm not enamored with the
Traktion 2 software that Mackie ships with the
preamp/mixer/interface.
I would like the software to cover simultaneous
multi-channel real-time recording, fairly extensive
editing/splicing, perhaps some minor time shifting of the
surround tracks, and preparation of tracks for final
mastering, with the option of doing the mastering into
DVD-A and Redbook format myself. .
I'd appreciate your advice/input on which software you
think best for this purpose. I will have the Traktion,
of course. And I could use Audacity, but it won't do
192k. What beyond those two are most suitable for my
purpose, and what are their strengths and weaknesses?

Audition/CEP does multi-channel recording, stereo recording,
editing, time-shifting, and all that good stuff. It has a
good reputation for sonic transparency if all you want to do
is clean recording. It can even work as audio test equipment
in a pinch, as well as do data generation and collection for
other audio analysis software.

Audition/CEP also has a very large collection of built-in
sonic effects, expressed in both traditional and academic
terms. It has more different kinds of equalizers than you
may have ever seen in one place. It has the ability to edit
and shift data in both the frequency domain as well the
more traditional time domain editing. It is widely used for
transcribing legacy recordings because it has so many
different effective options for noise reduction.

Adobe has their own NMTP and HTML online user groups with
are pretty agressively supported. Oh, and Audition/CEP
costs only a fraction of a lot of its competition.

I think that Audition/CEP works with sample rates up to at
least 10 MHz, so for your purposes Harry, its got a bit of
room for future expansion. ;-) Its also one of the few
things that John Atkinson and I actually agree about. Well,
maybe. ;-)

"Harry Lavo" wrote in message


But my reason for the interest for 192k is not extended
high-frequencies, but rather the fact that it is not
until this level of frequency that impulse response
pre-echo becomes theoretically inaudible.

Whose theory?

The usual criteria for temporal masking suggest that the
pre/post echoes from 44 Khz converters are not audible.

Disk space/processing power is not an issue for me with
only
four channels recorded. My own listening to commercial
releases suggests to me that 192k "sounds" slightly more
natural than 96k, and SACD even more so.

Maybe Harry we'll get you into some good recording software
like the stuff I used to assemble the files posted at
http://64.41.69.21/technical/sample_rates/index.htm , and
you'll finally get some real-world experience.

Well, a guy can hope...

"Lorin David Schultz" wrote in message
news:Aa07f.29029$y_1.2876@edtnps89...
"Harry Lavo" wrote:

[...] I'll be recording a wonderful 9' Steinway in a superb acoustic,
and transient response will be an important factor in the sound, I
believe.


Whacha gonna use for mics?


For starters, a quartet of Sony 270 small-diaphragm cardioid condensers
which are surprisingly good mics. These would be used in front-facing and
rearward-facing ORTF configuration, although I might switch the fronts to
X-Y.. However, I'm inclined to rent both a Royer SF-24 and a pair of
Earthworks TC-40 omni's for testing in the front mike position before I
commit and start the sessions.