[Romeo Rondeau]

Do you own a computer soundcard with direct monitoring?

No, I don't own one, but I did review the TASCAM US-122 with direct
monitoring, a fully analog path between the input jack and the
headphone jack. There's essentially a four-input mixer inside the case.
Two of the inputs are from the returns from the computer (which are of
course delayed) and the other two inputs are from the mic preamps, with
a switch to put a single input in the center if that's what you want to
hear. A pot adjusts the balance between the playback and input. Works
fine. Zero measurable delay from the input jack to the headphone jack
when in the "no latency monitor" mode. At least less than 0.05 uS,
about the best I can resolve with my oscilloscope.

That unit is a different animal from most interfaces, what you're talking
about is parallel monitoring and you can actually do it on a Soundblaster.
I'm talking about soundcards that do monitor switching, so we are talking
about different animals here.


And you're still going through the converters. There is still latency.
Anytime you want to prove my point, I'll be here.

No contest there. The 400F has "low latency monitoring" which does not
depend on the ASIO driver running on the computer. It does not have
direct monitoring. "Low" is a very indefinite number. I don't know what
it is for this unit, as I've said. But I know it's greater than zero.

It's a mixer with a computer interface, can it do "input to tape" monitor
switching?


NEVER sum two sognals, one of which is delayed by a small amount! Don't
you
know that causes comb filtering?

OK, next time I sing, I'll stuff a rag down my throat to block the
acoustic path through my head.

If it were so much of a problem, then why don't more people complain about
it?


You STILL don't get what I'm talking about. I'm not talking about
summing electrical signals here, I'm talking about summing acoustic
pressure waves - one coming from the earphone and hitting the outer
surface of your eardrum, the other coming up your throat and hitting
the inside of your eardrum. Since both come from the same source (your
vocal cords) but through different paths, they will arrive at your
eardrum at different times. If you have the same pressure pushing on
opposite sides of the eardrum at the same time, the eardrum won't move
- hence you'll hear no sound. That's an extreme case of course, but
it's the mechasim at work here.

Mike, I know exactly what you're talking about, I'm not arguing that... what
I am arguing with you about is the degree at which it becomes a problem.


They aren't combined at a close enough amplitude to be a problem.

That might be true, or it might not. It depends on how loud you turn up
the phones. The reason why this isn't a problem for many people in
practice is because they want more volume in the phones. This is why I
suggest that you try an experiment where you take the time to vary the
headphone level as you're talking or singing rather than just set it so
that the phones are pretty darm loud like you or your clients like it.

Mike, I've been monitoring through the machines for years, so have a lot
of
other people. Why do we not have huge problems?

Because individuals are all different. I don't work with rock singers
who want their voice really loud in the phones.

You're saying this to the big "western swing" engineer? :-)


There is nothing wrong with my understanding. They are NOT identical! One
of
them is being filtered by your mouth and one of them is what resonates
through your skull, they are pretty dissimilar.

They're dissimilar in spectral power, but that just means that the
amplitude of some frequencies are different. Some will be cancelled to
a greater degree than others if the time relationship between the two
is such that some frequencies will be out of phase (which will almost
certainly be the case).

We agree on this.

[Dr. Dolittle]

Romeo Rondeau wrote:


I see them putting one earpiece on and monitoring their vocal from
the air, even back in the analog days.


That is what I do.

[Scott Dorsey]

Dr. Dolittle wrote:
Romeo Rondeau wrote:

I see them putting one earpiece on and monitoring their vocal from
the air, even back in the analog days.

That is what I do.

This was standard practice back in the fifties and sixties when studios
weren't providing sophisticated monitor mixes, but these days it is hard
to get people to do it. Funny thing, it works a lot better than most of
the fancy monitor mix hardware.

Whatever happened to the Roland matrix mixers anyway?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mike Rivers]

Romeo Rondeau wrote:

I did review the TASCAM US-122

That unit is a different animal from most interfaces, what you're talking
about is parallel monitoring and you can actually do it on a Soundblaster.

It is? You can? I guess that when there's more than one way to monitor,
you have to come up with a name for the old way.

I'm talking about soundcards that do monitor switching, so we are talking
about different animals here.

I'm not aware that any do. ASIO has hooks for that, and when the driver
tells the card that the program is in "auto monitor" mode, the card can
do a digiital loopback, sending what's going out to the computer back
through the card's D/A concerter - about 1.5 ms later than it went in.
A tape deck has a relay that connects the input jack to the output
jack. The output goes out about 0.0000000000000000000000000001
milliseconds after it went in.

The 400F has "low latency monitoring"

It's a mixer with a computer interface, can it do "input to tape" monitor
switching?

It's a DIGITAL mixer with a computer interface, with the inherent
delays. I don't think the 400F does real monitor switching. One input
to the monitor mixer is a stereo return from the computer. You mix what
goes out that stream in the DAW. The 400F's monitor mixer is primarily
an input mixer. You can't bring 8 streams back from the computer and
mix them in the 400F.

If it were so much of a problem, then why don't more people complain about
it?

LIke I said, it isn't a problem if your headphone volume is high
enough. More people want the headphones turned up than turned down, so
most people are simply not aware of the effect. Of course it could be
that their voice sounds more natural to them BECAUSE it's loud enough
to swamp out the comb filtering effect.

If you want a second opinion about whether this is noticable, ask Bob
Ohlsson or John Klett. Both have more than a little studio experience,
and both have written extensively about monitoring and what can be
wrong with it.

Mike, I know exactly what you're talking about, I'm not arguing that... what
I am arguing with you about is the degree at which it becomes a problem.

Well, any time you want to argue degree, you'd to have a diffrence of
opinion. How rare do you like your steak?

[Mike Rivers]

Romeo Rondeau wrote:

Most of the really good singers don't even monitor with both sides of the
cans, anyway. I see them putting one earpiece on and monitoring their vocal
from the air

Yeah, because it sounds better that way when they don't have a good mix
in the headphones. It also looks like they know what they're doing.

[Mike Rivers]

Scott Dorsey wrote:

Whatever happened to the Roland matrix mixers anyway?

I'm not sure what Roland you're talking about but the Avion and Hear,
albeint digital, let a singer create and control his own mix. The Oz
Audio headphone mixer was kind of a small scale "more me" mixer with a
main stereo mix input, a reverb return, and four dedicated inputs that
could be mixed with the main stereo input, to four separate headphone
outputs. Mackie has what appears to be (to the best of my recollection)
essentially a copy of the Oz as their HMX-56 headphone amplifier.

[hank alrich]

Romeo Rondeau wrote:

NEVER sum two sognals, one of which is delayed by a small amount! Don't
you
know that causes comb filtering?

This is somehow much more of a practical problem when the two signals
are combined in one speaker cone (or headphone). When the sources are
physically independent, the brain copes remarkably well.

Thank you. Mike and Hank have the modern recording world coming to a halt
because of this "problem" which hasn't been an issue since digital machines
came out.

Bull****. We get araound it, using what's between our ears. It is not a
problem because we make it not a problem. But we also don't pretend it
doesn't happen.

Though, come to think of it, based on much of what I hear, bringing "the
modern recording world" to a ****ing halt for some wake up calls might
make good sense. There's enough shiny **** around to last for minutes,
such is its worth.

--
ha

[hank alrich]

Federico wrote:

This are the preamps I'm currently using:
http://www.yamaha.com/yamahavgn/Docu...evised0305.pdf and
they have been modified (mostly condensers).
It has two set of outputs: 8-channels on DB25 (I use this one) and 4 pairs
of euroblock connectros. Both +4dB
Are you telling me that I can use them both at one time? One to my recorder
and one for monitoring? That would be fantastic!

I'm pretty sure that'll work fine. I downloaded the .pdf twice, but both
times it somehow got corrupted. If one wasn't supposed to use both
outputs, I'd expect Yamaha to state that in the docs somewhere. Have you
read any bads news of that kind? g

--
ha

[Romeo Rondeau]

I'm talking about soundcards that do monitor switching, so we are talking
about different animals here.

I'm not aware that any do. ASIO has hooks for that, and when the driver
tells the card that the program is in "auto monitor" mode, the card can
do a digiital loopback, sending what's going out to the computer back
through the card's D/A concerter - about 1.5 ms later than it went in.
A tape deck has a relay that connects the input jack to the output
jack. The output goes out about 0.0000000000000000000000000001
milliseconds after it went in.

Tascam DA-88's and Alesis ADATS don't have relays, they monitor through the
converters.

Well, any time you want to argue degree, you'd to have a diffrence of
opinion. How rare do you like your steak?

Duh!

[Romeo Rondeau]

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

Romeo Rondeau wrote:

Most of the really good singers don't even monitor with both sides of the
cans, anyway. I see them putting one earpiece on and monitoring their
vocal
from the air

Yeah, because it sounds better that way when they don't have a good mix
in the headphones. It also looks like they know what they're doing.

Most of them tell me that it's because they prefer to hear their voice
naturally, not through a microphone and headphones.

[Lorin David Schultz]

Romeo Rondeau wrote:

Thank you. Mike and Hank have the modern recording world coming to
a halt because of this "problem" which hasn't been an issue since
digital machines came out.



Oh bull**** Romeo. Just because it doesn't bother you doesn't mean the
problem doesn't exist. And just because others mention it doesn't mean
they're flailing and moaning about it. Lighten up.

As for it not existing since digital machine came out, I just plain
don't know what the hell you're talking about, unless you've always had
a monitor path that doesn't take a nice little scenic cruise through
eight layers of software before it gets back to the artist. My 1992
vintage SAW rig with the CardD had plenty of throughput delay, thank
you. So did a 3324 if you decided to take a round trip through the
internal electronics rather than lifting the monitor source at the
console input.

Ultimately the situation is now one where many artists either accept or
adapt to the ever-so-slightly delayed monitor signal (hell, many of 'em
have never experienced it any other way). The fact that it is tolerated
does NOT mean the issue doesn't exist. So it doesn't impede your work.
That's great. Why does that prompt you to act like a dick towards
people who find it less than ideal?

--
"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)

[Romeo Rondeau]

Oh bull**** Romeo. Just because it doesn't bother you doesn't mean the
problem doesn't exist. And just because others mention it doesn't mean
they're flailing and moaning about it. Lighten up.

I never said the problem doesn't exist, can't you people read? I said it
wasn't usually a problem. I even stated why. Mike got confused as to my
terminology referring to "direct monitoring" and got us sidetracked, then
Hank got excited about taking another potshot at me and did so, that's all.


As for it not existing since digital machine came out, I just plain don't
know what the hell you're talking about, unless you've always had a
monitor path that doesn't take a nice little scenic cruise through eight
layers of software before it gets back to the artist. My 1992 vintage SAW
rig with the CardD had plenty of throughput delay, thank you. So did a
3324 if you decided to take a round trip through the internal electronics
rather than lifting the monitor source at the console input.

We aren't talking about monitoring through the software, go back and read
the thread. We were talking about monitoring through the A/D and D/A
converters.


Ultimately the situation is now one where many artists either accept or
adapt to the ever-so-slightly delayed monitor signal (hell, many of 'em
have never experienced it any other way). The fact that it is tolerated
does NOT mean the issue doesn't exist. So it doesn't impede your work.
That's great. Why does that prompt you to act like a dick towards people
who find it less than ideal?

You don't understand the issue that we're talking about either, read the
thread. How was I acting like a dick? Quotes please...

[Lorin David Schultz]

Romeo Rondeau wrote:

We aren't talking about monitoring through the software, go back
and read the thread. We were talking about monitoring through the
A/D and D/A converters.

I know. So am I. The off-the-cuff remark about "eight layers of
software" was a poor choice of words.

It's only recently that one could reasonably configure a system for
relatively low throughput delay (I prefer that phrase to the use of the
term "latency" in this context). Even DA88s and ADATs held up the
signal long enough to be noticeable.

Even now, *most* systems do enough buffering and processing that *most*
people aren't getting their signals back to them in a couple
milliseconds like it seems you are.

I don't have the luxury of being able to test the affects of 1.5ms
throughput -- my choices are instantaneous analog or slow digital -- but
I suspect that you're right about signals coming back that fast being
good enough to not screw up a singer. Like you said, it's the
equivalent of standing a couple feet away from the mic.



How was I acting like a dick? Quotes please...

You made some remark about Mike and Hank, suggesting that they were
doing a Chicken Little. It struck me as insulting.

Still, I came on way too strong in my reply to you. I came home in a
****y mood and dumped on you. I apologize for that.

--
"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)

[Mike Rivers]

Romeo Rondeau wrote:

Tascam DA-88's and Alesis ADATS don't have relays, they monitor through the
converters.

I know. And I wondered if this would be a problem. Most people said no.
I tried it myself and found that sometimes they were wrong. If you can
work around the problem, for example, by sending the live vocal to the
headphone mix by a direct analog path, fine. If you routinely work in a
way so that the problem is masked (a loud headphone mix) that doesn't
mean the problem doesn't exist. It only means that it doesn't bother
you or your clients. That's fine, too.

[Mike Rivers]

Romeo Rondeau wrote:

We aren't talking about monitoring through the software, go back and read
the thread. We were talking about monitoring through the A/D and D/A
converters.

Who cares how the delay is created? It's still a delay. The quest for
"low latency drivers" is an attempt to add as little as possible to the
delay through the converters. If you were working on a pure analog
system, you wouldn't put the singer's voice through a 2ms delay and
send it back to his headphones, would you? That's what you're doing
when you monitor through the converters.

[Scott Dorsey]

Mike Rivers wrote:
Scott Dorsey wrote:

Whatever happened to the Roland matrix mixers anyway?

I'm not sure what Roland you're talking about but the Avion and Hear,
albeint digital, let a singer create and control his own mix. The Oz
Audio headphone mixer was kind of a small scale "more me" mixer with a
main stereo mix input, a reverb return, and four dedicated inputs that
could be mixed with the main stereo input, to four separate headphone
outputs. Mackie has what appears to be (to the best of my recollection)
essentially a copy of the Oz as their HMX-56 headphone amplifier.

Back in the seventies, Roland made a thing that was like an oversize
version of the Oz. It was like the Crest matrix mixer but it wasn't
modular like the Crest. For a while, everybody was using it, then it
seemed to disappear.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Romeo Rondeau]

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

Romeo Rondeau wrote:

Tascam DA-88's and Alesis ADATS don't have relays, they monitor through
the
converters.

I know. And I wondered if this would be a problem. Most people said no.
I tried it myself and found that sometimes they were wrong. If you can
work around the problem, for example, by sending the live vocal to the
headphone mix by a direct analog path, fine. If you routinely work in a
way so that the problem is masked (a loud headphone mix) that doesn't
mean the problem doesn't exist. It only means that it doesn't bother
you or your clients. That's fine, too.

FWIW, I believe that the Sony versions of the DA-88's have true input
monitor switching on their I/O cards.

[Romeo Rondeau]

Even now, *most* systems do enough buffering and processing that *most*
people aren't getting their signals back to them in a couple milliseconds
like it seems you are.

I'll admit, I'm spoiled with the RME stuff. For me, they have been the only
way to reliably get your DAW to function like a tape machine, but yes there
is converter delay and yes I can tell that it's there if I try.


I don't have the luxury of being able to test the affects of 1.5ms
throughput -- my choices are instantaneous analog or slow digital -- but I
suspect that you're right about signals coming back that fast being good
enough to not screw up a singer. Like you said, it's the equivalent of
standing a couple feet away from the mic.



How was I acting like a dick? Quotes please...

You made some remark about Mike and Hank, suggesting that they were doing
a Chicken Little. It struck me as insulting.

I didn't mean it that way, it was meant to be tongue in cheek, I apologize.


Still, I came on way too strong in my reply to you. I came home in a
****y mood and dumped on you. I apologize for that.

Oh hell, it's hard to tell how someone really means something when you are
just looking at words, no harm done.

[Romeo Rondeau]

I don't have the luxury of being able to test the affects of 1.5ms
throughput

Do you have a delay unit (real hardware, not a plug-in) that goes down
to 1.5 ms? If so, do this:

Plug a mic into one channel of a mixer. Use a pre-fader send to send
the mic input out to the delay. Bring the delay output back into a
mixer channel and pan it to the center. Plug a set of headphones into
the mixer's output and put them on.

If the delay unit is digital, done even bother to set it to 1.5ms. Set it to
0, the converter throughput will be around 1.5ms. Unless of course the unit
compensates for it's own converter latency, which I would doubt it does. It
should be easy to test on a workstation. They ARE useful for some things :-)

[Romeo Rondeau]

We aren't talking about monitoring through the software, go back and read
the thread. We were talking about monitoring through the A/D and D/A
converters.

Who cares how the delay is created? It's still a delay. The quest for
"low latency drivers" is an attempt to add as little as possible to the
delay through the converters. If you were working on a pure analog
system, you wouldn't put the singer's voice through a 2ms delay and
send it back to his headphones, would you? That's what you're doing
when you monitor through the converters.

Okay, okay! Enough already! :-) I'll just switch to recording rock and roll
and give my clients bad headphone mixes :-) That will solve the problem...
It's a joke, Mike... relax... have some herbal tea and some spicy Indian
snacks :-)

[Dr. Dolittle]

What??

What kind of monitoring sytem do you have that you couldn't get a $50
distribution amp and monitor the actual input signal?

Lorin David Schultz wrote:

Oh bull**** Romeo. Just because it doesn't bother you doesn't mean the
problem doesn't exist. And just because others mention it doesn't mean
they're flailing and moaning about it. Lighten up.

As for it not existing since digital machine came out, I just plain
don't know what the hell you're talking about, unless you've always had
a monitor path that doesn't take a nice little scenic cruise through
eight layers of software before it gets back to the artist. My 1992
vintage SAW rig with the CardD had plenty of throughput delay, thank
you. So did a 3324 if you decided to take a round trip through the
internal electronics rather than lifting the monitor source at the
console input.

Ultimately the situation is now one where many artists either accept or
adapt to the ever-so-slightly delayed monitor signal (hell, many of 'em
have never experienced it any other way). The fact that it is tolerated
does NOT mean the issue doesn't exist. So it doesn't impede your work.
That's great. Why does that prompt you to act like a dick towards
people who find it less than ideal?

[Dr. Dolittle]

Lorin David Schultz wrote:

It's only recently that one could reasonably configure a system for
relatively low throughput delay (I prefer that phrase to the use of the
term "latency" in this context). Even DA88s and ADATs held up the
signal long enough to be noticeable.

For monitoring inputs? Why go though all that stuff? (..)

[Romeo Rondeau]

"Dr. Dolittle" wrote in message
...


Lorin David Schultz wrote:

It's only recently that one could reasonably configure a system for
relatively low throughput delay (I prefer that phrase to the use of the
term "latency" in this context). Even DA88s and ADATs held up the signal
long enough to be noticeable.

For monitoring inputs? Why go though all that stuff? (..)

Punch and entire band to get an ending just right and you'll see. You needs
lots of channels, pain the the ass. That's why most folks monitor in sync
mode. That's what it's there for, afterall. If you like to work other ways,
that's fine. But eating up 48 channels just to record 24 is silly and
time-consuming.

[Dr. Dolittle]

Romeo Rondeau wrote:

But eating up 48 channels just to record 24 is silly and
time-consuming.


Is this a permanent set up? Not silly and not time-consuming at all. You
set it up once.

The most important component of a performance is the performance. And
how the performers hear themselves do affect the performances.

[Lorin David Schultz]

Dr. Dolittle wrote:

What??

What kind of monitoring sytem do you have that you couldn't get a
$50 distribution amp and monitor the actual input signal?


What are you talking about? You must be coming in to this late. The
whole point of the discussion is the difference between monitoring
directly off the analog source vs. through the converters. We're
discussing whether or not monitoring through the converters is a
problem.

I have no problems with my monitoring system, but thanks for your
concern. You will be happy to know that my clients get their mix off
the console, the way God intended. g

--
"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)

[Romeo Rondeau]

"Lorin David Schultz" wrote in message
news:0cItg.135367$S61.115370@edtnps90...
Dr. Dolittle wrote:

What??

What kind of monitoring sytem do you have that you couldn't get a
$50 distribution amp and monitor the actual input signal?


What are you talking about? You must be coming in to this late. The
whole point of the discussion is the difference between monitoring
directly off the analog source vs. through the converters. We're
discussing whether or not monitoring through the converters is a problem.

I have no problems with my monitoring system, but thanks for your concern.
You will be happy to know that my clients get their mix off the console,
the way God intended. g

So do mine :-) Through the converters, then to the console :-) 'Course I'm
using a digital console... and feeding my DAW with lightpipe output preamps,
running per input direct monitoring, so my latency isn't nearly as bad as
some. I still can't figure out what he's talking about, though.

[Mike Rivers]

Romeo Rondeau wrote:

"Direct monitoring" is not as direct as you think :-)

I thought mayb this thread had finally run its course, but yesterday at
the NAMM show, I was talkig with Marcus Ryle, the designer of the ADAT,
about some new Line 6 interface product and the subject got around to
latency in monitoring. He mentioned that the input/audo monitor modes
in the ADAT did NOT go through the A/D and D/A converters, but turned
the analog audio around right at the input. He recognized the potential
for monitoring problems and solved it by routing the signal in the same
way as the analog recorders that the ADAT intended to replace. He
belived (and rightly so, I think) that this would be an issue to anyone
making the transition from an analog multitrack recorder to the ADAT.

[Romeo Rondeau]

I thought mayb this thread had finally run its course, but yesterday at
the NAMM show, I was talkig with Marcus Ryle, the designer of the ADAT,
about some new Line 6 interface product and the subject got around to
latency in monitoring. He mentioned that the input/audo monitor modes
in the ADAT did NOT go through the A/D and D/A converters, but turned
the analog audio around right at the input. He recognized the potential
for monitoring problems and solved it by routing the signal in the same
way as the analog recorders that the ADAT intended to replace. He
belived (and rightly so, I think) that this would be an issue to anyone
making the transition from an analog multitrack recorder to the ADAT.

I stand corrected, although I know for a fact that this is not the case with
the DA-88.