[Gary Eickmeier]

I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.

So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

So would Behringer have a way of working around these problems or am I
correct in being nervous about it?

Gary Eickmeier

--
Gary Eickmeier

[PStamler]

On Saturday, August 9, 2014 6:08:54 PM UTC-6, Gary Eickmeier wrote:
I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.



So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

A question and a couple of comments. First, did you measure the system's gain with and without the DEQ in circuit? If it's introducing a few dB of loss, that would account for the subjective impression of poorer sound. One way to do this is to have the DEQ set flat, play back a 1kHz test tone, and measure the voltage at the speaker terminals. Remove the DEQ from the chain and do the measurement again.

Second, the DEQ *does* have an input gain control, does it not? If that's set so low that the meter is barely moving, you indeed have a problem with gain staging; the output from a consumer hi-fi receiver is less than it's designed to operate from, and you could set the gain control so you see a little more action from the meters. On the other hand, this is a 24-bit device; you can get away with low levels and still maintain linearity and low noise.

You really don't want to be "filling up" the bits; that means you're pushing the edge of digital clipping, which sounds horrible.

Set the input gain so the system gain is the same with the DEQ in and out of the circuit, and you should still be okay. Or turn up the gain so that the meters peak at about -12dBFS; that should get you way too much signal to the power amp, but if the latter has a level control you can turn it down again.

By the way, you wrote "But the analog inputs to the unit from my receiver were variable IAW the volume knob."

What does IAW mean?

Peace,
Paul

[Sean Conolly]

"Gary Eickmeier" wrote in message
...
I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.

So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of
available bits. But the analog inputs to the unit from my receiver were
variable IAW the volume knob. So how did I ever know that I was using all
of the bits in the equalizer? The output meter usually barely moved. The
input I doubt had enough gain for 16 bits to be filled up.

So would Behringer have a way of working around these problems or am I
correct in being nervous about it?

Hey there Gary,

If you're running a *very* low signal through the unit I could see where
quantization noise could start becoming a factor. I recall your receiver is
running at -10 consumer levels? As far as I know the DEQ only runs at +4,
and although you can adjust the volume it's a digital control upstream of
the D/A converter.

Buy or build a pad for the output, so you can run the output meters close to
50% and see if that helps.

Sean

[Frank Stearns]

"Gary Eickmeier" writes:

I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.

Not surprising at all.

So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

So would Behringer have a way of working around these problems or am I
correct in being nervous about it?

Depends on how they designed and built it. Knowing Behringer, probably not very well
-- or when they copied someone's design they cut every possible corner in power
supply quality, board quality, connector quality, caps, etc, etc. So I'd be nervous
too! And these days, I probably wouldn't worry as much about the converters as the
analog signal paths to and from those converters.

I've found the very best room EQ to be proper acoustic treatment -- control the bass
and reflections, add diffusion and small amounts of very carefully selected
reflection as needed; with those elements develop a reverb curve suitable for the
space, etc.

The results can be stunning. In fact, the new room here is nearly done. The mid and
top end has never sounded so good. The 20-40 hz range appears to be dead flat.
There's just a slight amount of murkiness in the 100-150 range; I'm halfway through
adding a set of traps tuned to this and will know soon whether a diaphragm
density change might be also appropriate for some of the other traps.

Maybe you need room EQ, maybe you don't. But do as much of the mechanical/phsyical
"EQ" first.

Frank
Mobile Audio
--

[geoff]

On 10/08/2014 12:08 p.m., Gary Eickmeier wrote:
I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.

So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

So would Behringer have a way of working around these problems or am I
correct in being nervous about it?

Gary Eickmeier


It is unlikely the unit itself stuffed anything up - more likely your
settings o some of the effects modules, or imagination.

There was some story back in 44k1/16bit days of a panel of people set up
with a live feed versus the same source with 10 AD-DA devices in series.
And they couldn't reliably tell the difference. the 2496 is years ahead
of any of those.


geoff

[geoff]

On 10/08/2014 6:04 p.m., Frank Stearns wrote:
"Gary Eickmeier" writes:


Depends on how they designed and built it. Knowing Behringer, probably not very well
-- or when they copied someone's design they cut every possible corner in power
supply quality, board quality, connector quality, caps, etc, etc. So I'd be nervous
too! And these days, I probably wouldn't worry as much about the converters as the
analog signal paths to and from those converters.


Not correct Frank. It it one of the Behringer 'goodies'. Even a relay
'true bypass'.

geoff

[Mike Rivers[_2_]]

On 8/9/2014 8:08 PM, Gary Eickmeier wrote:

I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.

So your system sounds better with the DEQ out of the signal path even
with all of its settings as bypassed as you can get? That's
disappointing. I wouldn't be surprised that there was a barely
noticeable change for no other reason than that there's another analog
component in the signal path, but you make it sound like a horror show
with it in line.

So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits.

Well, we used to, when there were only about 12 bits that we could count
on being reasonably accurate. With today's 24-bit systems, we prefer
more headroom so that we aren't using most of the bits most of the time.
The positive result is that what's converted between analog to digital
and back is more accurate. The downside is that you end up with an
analog output that may be rather low and you'll amplify the analog noise
when bringing it up to your normal listening level. This is why we have
the concept of "gain structure" - to get things working together
without, as you say, making the best use of the available bits, or
without clipping.

But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

Doesn't it have an input level meter? How about a switch for nominal
operating level (-10/+4)? According to the manual, there's a switch to
change the maximum input level of the main XLR inputs from +12 to +22
dBu. Since the meter is showing a low level, put the switch in the +12
position and you'll get 10 dB more gain. Do you have "tape out/in" jacks
somewhere in your system? Tape output is nearly always fixed level,
somewhere near the design center for the unit. Of course it will vary
with the input source, but as long as you aren't clipping the input or
have to turn the volume knob way up to get a usable output level, the
tape output should be around the nominal operating level of the unit.
This, for home stereo gear, is usually in the ballpark of -10 dBV.

Still, the equalizer should be a unity gain device (plus or minus
whatever you do with the equalizer) so what you put in should be what
you get out. But there's all kinds of processing in there. Make sure
none of the dynamics stuff, the compressor, limiter, or dynamic EQ
processing is switched in. You might also try using the auxiliary output
jacks and use the I/O menu to feed those outputs from somewhere earlier
in the chain than what feeds the main outputs.

Now I don't have a DEQ2496, and I've never used one, so don't take this
as defending the product. But lots of people use them effectively, lots
of people just hate everything that Behringer makes. It sounds like
you've at least listened critically to it and can be objective about
whether the overall effectiveness is to make your system sound better or
worse. But explore the options and settings fully before you give up on it.

So would Behringer have a way of working around these problems

Yes. The input level switch, for one. RTFM

am I correct in being nervous about it?

You shouldn't be nervous, you should look for the problem and fix it
(which may mean taking the equalizer out of your system) if it bothers you.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Mike Rivers[_2_]]

On 8/9/2014 8:08 PM, Gary Eickmeier wrote:

But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

Again, RTFM and explore the controls. The Meter menu turns the display
into three pages worth of metering options, including showing the input
level.



--
For a good time, visit http://mikeriversaudio.wordpress.com

[Scott Dorsey]

Gary Eickmeier wrote:
So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

Where did you get this idea?

What makes you think "using all the bits" is significant or even useful?

Do you always operate your power amplifier at full tilt for fear you're
not getting all the output ower it's capable of?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Frank Stearns]

geoff writes:

On 10/08/2014 6:04 p.m., Frank Stearns wrote:
"Gary Eickmeier" writes:


Depends on how they designed and built it. Knowing Behringer, probably not very well
-- or when they copied someone's design they cut every possible corner in power
supply quality, board quality, connector quality, caps, etc, etc. So I'd be nervous
too! And these days, I probably wouldn't worry as much about the converters as the
analog signal paths to and from those converters.


Not correct Frank. It it one of the Behringer 'goodies'. Even a relay
'true bypass'.

Okay, fair enough. But particularly for audio, there are long life, highly reliable,
high-performance relays that even if seldom actuated, I'd want to be on the
high-performance end just so that the signal flow through the contacts isn't
degraded intermittently. In quantity, the good ones like this are $2-$5 each.

Then there are multi-purpose cheapies that run for pennies each in quantity. Maybe
they're just for switching line AC or other applications where unlike audio,
variances in the "quality" of the close doesn't matter. But you're never sure just
how "tight" the contact close will be, and how that connection will "hold" over time
(seconds, minutes, or days).

It's likely a safe bet which class of relay was used in a device at this price
point.

YMMV.

Frank
Mobile Audio

--

[Gary Eickmeier]

"Frank Stearns" wrote in message
news
"Gary Eickmeier" writes:

I played with the 2496 long enough to learn all of its functions and
features, did a few EQs with it, measured my speakers etc etc, but I got a
little nervous about having an additional A/D D/A in my system so I yanked
it out of there and the sound seems a lot better now, tighter, more
together, larger soundstage, etc.

Not surprising at all.

So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of
available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits
in
the equalizer? The output meter usually barely moved. The input I doubt
had
enough gain for 16 bits to be filled up.

So would Behringer have a way of working around these problems or am I
correct in being nervous about it?

Depends on how they designed and built it. Knowing Behringer, probably not
very well
-- or when they copied someone's design they cut every possible corner in
power
supply quality, board quality, connector quality, caps, etc, etc. So I'd
be nervous
too! And these days, I probably wouldn't worry as much about the
converters as the
analog signal paths to and from those converters.

Well... this is (seems to be) a very sophisticated piece of equipment. It
would be a shame if it had less than adequate components and converters in
it.

I've found the very best room EQ to be proper acoustic treatment --
control the bass
and reflections, add diffusion and small amounts of very carefully
selected
reflection as needed; with those elements develop a reverb curve suitable
for the
space, etc.

The results can be stunning. In fact, the new room here is nearly done.
The mid and
top end has never sounded so good. The 20-40 hz range appears to be dead
flat.
There's just a slight amount of murkiness in the 100-150 range; I'm
halfway through
adding a set of traps tuned to this and will know soon whether a diaphragm
density change might be also appropriate for some of the other traps.

Maybe you need room EQ, maybe you don't. But do as much of the
mechanical/phsyical
"EQ" first.

Frank
Mobile Audio

Yes, and thanks Frank. In fact, that is one of the reasons I pulled it out,
because the EQ it was calling for was barely off flat, as in not EQ'd at
all. I EQ to a room curve that has a lifted bottom and a falling high end.
My new speakers had that - along with the Velodyne - just naturally.

I need to do alot more listening to them but right now I am in hog heaven
without the EQ, so I can still use it for RTA work if I get some comparison
speakers in to test some time, but I just don't need it inserted into my
system any more.

Gary Eickmeier

[Gary Eickmeier]

PStamler wrote:
A question and a couple of comments. First, did you measure the
system's gain with and without the DEQ in circuit? If it's
introducing a few dB of loss, that would account for the subjective
impression of poorer sound. One way to do this is to have the DEQ set
flat, play back a 1kHz test tone, and measure the voltage at the
speaker terminals. Remove the DEQ from the chain and do the
measurement again.

I confess no I did not.

Second, the DEQ *does* have an input gain control, does it not? If
that's set so low that the meter is barely moving, you indeed have a
problem with gain staging; the output from a consumer hi-fi receiver
is less than it's designed to operate from, and you could set the
gain control so you see a little more action from the meters. On the
other hand, this is a 24-bit device; you can get away with low levels
and still maintain linearity and low noise.

It has a little button called the MAX switch that "raises the maximum level
present at the MAIN inputs/outputs from +12 to +22 dBu." I have no idea what
that does, nor does the book tell me.

You really don't want to be "filling up" the bits; that means you're
pushing the edge of digital clipping, which sounds horrible.

Set the input gain so the system gain is the same with the DEQ in and
out of the circuit, and you should still be okay. Or turn up the gain
so that the meters peak at about -12dBFS; that should get you way too
much signal to the power amp, but if the latter has a level control
you can turn it down again.

By the way, you wrote "But the analog inputs to the unit from my
receiver were variable IAW the volume knob."

What does IAW mean?

In Accordance With


Peace,
Paul

[Gary Eickmeier]

Sean Conolly wrote:
"Gary Eickmeier" wrote in message
...

Hey there Gary,

If you're running a *very* low signal through the unit I could see
where quantization noise could start becoming a factor. I recall your
receiver is running at -10 consumer levels? As far as I know the DEQ
only runs at +4, and although you can adjust the volume it's a
digital control upstream of the D/A converter.

Buy or build a pad for the output, so you can run the output meters
close to 50% and see if that helps.

Sean

That sounds like a great idea Sean. I like the equalizer, but not if it is
going to be a detriment to my sound! I will figure out how to do that and
report back.

Gary

[Scott Dorsey]

Gary Eickmeier wrote:

It has a little button called the MAX switch that "raises the maximum level
present at the MAIN inputs/outputs from +12 to +22 dBu." I have no idea what
that does, nor does the book tell me.

It changes the reference operating level, adding a pad on the front end
and gain on the back end. You really, really need to get the Yamaha Sound
Reinforcement Handbook and read the introduction to operating levels.

In the professional world, analogue operating levels are standardized, it's
not like in the consumer world where outputs are all over the place.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Sean Conolly]

"Dave Plowman (News)" wrote in message
...
In article ,
Sean Conolly wrote:
If you're running a *very* low signal through the unit I could see where
quantization noise could start becoming a factor. I recall your
receiver is running at -10 consumer levels? As far as I know the DEQ
only runs at +4, and although you can adjust the volume it's a digital
control upstream of the D/A converter.

I/O levels are switchable between pro and domestic. Sadly only both
together. It can be useful if the in and out can be switched
independently.


Hmmm - check through the manual - and there it is! There's a switch next to
the RTA mic input labeled 'MAX', which I always thought was for the mic
input level. The manual says it's for switching the input/output levels as
you indicated.

Thanks for that - I had to use the text search on the manual to find it
though.

Sean

[Gary Eickmeier]

Scott Dorsey wrote:
Gary Eickmeier wrote:
So my main question would be: Whenever we have a digital component
in the system, especially a digital recorder, we try and make best
use of available bits. But the analog inputs to the unit from my
receiver were variable IAW the volume knob. So how did I ever know
that I was using all of the bits in the equalizer? The output meter
usually barely moved. The input I doubt had enough gain for 16 bits
to be filled up.

Where did you get this idea?

What makes you think "using all the bits" is significant or even
useful?

Do you always operate your power amplifier at full tilt for fear
you're not getting all the output ower it's capable of?
--scott

Scott, how often do you make digital recordings without the max gain
possible without clipping? Maybe in a 24 bit system you can go without the
top half dozen bits lighting up, but you wouldn't purposely try to record at
the 8 or 10 bit level.

Gary

[Mike Rivers[_2_]]

On 8/10/2014 11:42 AM, Gary Eickmeier wrote:
It has a little button called the MAX switch that "raises the maximum level
present at the MAIN inputs/outputs from +12 to +22 dBu." I have no idea what
that does, nor does the book tell me.

I explained it in my post earlier today.

Non-Google translation of "raises the maximum level
present at the MAIN inputs/outputs from +12 to +22 dBu." is

Changes the input sensitivity so that input clipping occurs at +12 dBu
or +22 dBu

You need to learn the language of manuals. Knowing something about
specs, conventions, and operating levels helps, too.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Gary Eickmeier]

Mike Rivers wrote:
On 8/9/2014 8:08 PM, Gary Eickmeier wrote:


So your system sounds better with the DEQ out of the signal path even
with all of its settings as bypassed as you can get? That's
disappointing. I wouldn't be surprised that there was a barely
noticeable change for no other reason than that there's another analog
component in the signal path, but you make it sound like a horror show
with it in line.

Some of it could be psychoacoustic. But I am just asking about the AD/ DA
converter part.

So my main question would be: Whenever we have a digital component
in the system, especially a digital recorder, we try and make best
use of available bits.

Well, we used to, when there were only about 12 bits that we could
count on being reasonably accurate. With today's 24-bit systems, we
prefer more headroom so that we aren't using most of the bits most of the
time. The positive result is that what's converted between analog to
digital and back is more accurate. The downside is that you end up with an
analog output that may be rather low and you'll amplify the analog
noise when bringing it up to your normal listening level. This is why
we have the concept of "gain structure" - to get things working
together without, as you say, making the best use of the available bits,
or
without clipping.

Yes, fairly obvious, and the basis of my question.

But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the
bits in the equalizer? The output meter usually barely moved. The
input I doubt had enough gain for 16 bits to be filled up.

Doesn't it have an input level meter? How about a switch for nominal
operating level (-10/+4)? According to the manual, there's a switch to
change the maximum input level of the main XLR inputs from +12 to +22
dBu. Since the meter is showing a low level, put the switch in the +12
position and you'll get 10 dB more gain. Do you have "tape out/in"
jacks somewhere in your system? Tape output is nearly always fixed
level, somewhere near the design center for the unit. Of course it
will vary with the input source, but as long as you aren't clipping the
input or
have to turn the volume knob way up to get a usable output level, the
tape output should be around the nominal operating level of the unit.
This, for home stereo gear, is usually in the ballpark of -10 dBV.

Still, the equalizer should be a unity gain device (plus or minus
whatever you do with the equalizer) so what you put in should be what
you get out. But there's all kinds of processing in there. Make sure
none of the dynamics stuff, the compressor, limiter, or dynamic EQ
processing is switched in. You might also try using the auxiliary
output jacks and use the I/O menu to feed those outputs from
somewhere earlier in the chain than what feeds the main outputs.

Now I don't have a DEQ2496, and I've never used one, so don't take
this as defending the product. But lots of people use them effectively,
lots of people just hate everything that Behringer makes. It sounds like
you've at least listened critically to it and can be objective about
whether the overall effectiveness is to make your system sound better
or worse. But explore the options and settings fully before you give
up on it.
So would Behringer have a way of working around these problems

Yes. The input level switch, for one. RTFM

am I correct in being nervous about it?

You shouldn't be nervous, you should look for the problem and fix it
(which may mean taking the equalizer out of your system) if it
bothers you.

OK, I will try it again and study those gain settings and the various
meters. I know they have tried to think of everything, and this is a very
special product that may deserve greater attention from the owner....

Gary

[Scott Dorsey]

Gary Eickmeier wrote:
Kludge writes:
Where did you get this idea?

What makes you think "using all the bits" is significant or even
useful?

Do you always operate your power amplifier at full tilt for fear
you're not getting all the output ower it's capable of?

Scott, how often do you make digital recordings without the max gain
possible without clipping? Maybe in a 24 bit system you can go without the
top half dozen bits lighting up, but you wouldn't purposely try to record at
the 8 or 10 bit level.

The 8 bit level would be 96dB down from maximum in a 24 bit system. That is
pretty damn far down.

You have available to you an outrageous amount of dynamic range in a modern
digital system. Don't be afraid to use it. A little matter of 12 or 24 dB
is not going to be a worry.

There was a need to keep tight control over levels in digital systems back
in the early 1980s when converters weren't very linear, and back then a lot
of people would give out advice to "use all the bits." Those days are long,
long gone, thank God. Back then if you recorded with peaks at -60dBFS on the
PCM1610, the sound was audibly distorted and buzzy. Today with modern sigma
delta converters, I can record at -60dBFS and you probably won't even notice
the increase in noise floor since the ambient noise of the hall is _still_
above the converter noise floor. It is a different era.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[geoff]

On 10/08/2014 11:57 p.m., Scott Dorsey wrote:
Gary Eickmeier wrote:
So my main question would be: Whenever we have a digital component in the
system, especially a digital recorder, we try and make best use of available
bits. But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits in
the equalizer? The output meter usually barely moved. The input I doubt had
enough gain for 16 bits to be filled up.

Where did you get this idea?

What makes you think "using all the bits" is significant or even useful?

Do you always operate your power amplifier at full tilt for fear you're
not getting all the output ower it's capable of?
--scott



But Scott, pretty much ALL power amplifiers are indeed running at 'full
tilt' all the time. Level control is an attenuator to the input signal, no ?

;-)

geoff

[Sean Conolly]

"Gary Eickmeier" wrote in message
...
Mike Rivers wrote:
On 8/9/2014 8:08 PM, Gary Eickmeier wrote:


So your system sounds better with the DEQ out of the signal path even
with all of its settings as bypassed as you can get? That's
disappointing. I wouldn't be surprised that there was a barely
noticeable change for no other reason than that there's another analog
component in the signal path, but you make it sound like a horror show
with it in line.

Some of it could be psychoacoustic. But I am just asking about the AD/ DA
converter part.

So my main question would be: Whenever we have a digital component
in the system, especially a digital recorder, we try and make best
use of available bits.

Well, we used to, when there were only about 12 bits that we could
count on being reasonably accurate. With today's 24-bit systems, we
prefer more headroom so that we aren't using most of the bits most of the
time. The positive result is that what's converted between analog to
digital and back is more accurate. The downside is that you end up with
an
analog output that may be rather low and you'll amplify the analog
noise when bringing it up to your normal listening level. This is why
we have the concept of "gain structure" - to get things working
together without, as you say, making the best use of the available bits,
or
without clipping.

Yes, fairly obvious, and the basis of my question.

But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the
bits in the equalizer? The output meter usually barely moved. The
input I doubt had enough gain for 16 bits to be filled up.

Doesn't it have an input level meter? How about a switch for nominal
operating level (-10/+4)? According to the manual, there's a switch to
change the maximum input level of the main XLR inputs from +12 to +22
dBu. Since the meter is showing a low level, put the switch in the +12
position and you'll get 10 dB more gain. Do you have "tape out/in"
jacks somewhere in your system? Tape output is nearly always fixed
level, somewhere near the design center for the unit. Of course it
will vary with the input source, but as long as you aren't clipping the
input or
have to turn the volume knob way up to get a usable output level, the
tape output should be around the nominal operating level of the unit.
This, for home stereo gear, is usually in the ballpark of -10 dBV.

Still, the equalizer should be a unity gain device (plus or minus
whatever you do with the equalizer) so what you put in should be what
you get out. But there's all kinds of processing in there. Make sure
none of the dynamics stuff, the compressor, limiter, or dynamic EQ
processing is switched in. You might also try using the auxiliary
output jacks and use the I/O menu to feed those outputs from
somewhere earlier in the chain than what feeds the main outputs.

Now I don't have a DEQ2496, and I've never used one, so don't take
this as defending the product. But lots of people use them effectively,
lots of people just hate everything that Behringer makes. It sounds like
you've at least listened critically to it and can be objective about
whether the overall effectiveness is to make your system sound better
or worse. But explore the options and settings fully before you give
up on it.
So would Behringer have a way of working around these problems

Yes. The input level switch, for one. RTFM

am I correct in being nervous about it?

You shouldn't be nervous, you should look for the problem and fix it
(which may mean taking the equalizer out of your system) if it
bothers you.

OK, I will try it again and study those gain settings and the various
meters. I know they have tried to think of everything, and this is a very
special product that may deserve greater attention from the owner....

... but to be fair, I've had one for many years and never knew that the MAX
button wasn't for the RTA mic. It is in the manual, but it ain't hard to
miss either.

Sean

[Sean Conolly]

"Gary Eickmeier" wrote in message
...
Scott Dorsey wrote:
Gary Eickmeier wrote:
So my main question would be: Whenever we have a digital component
in the system, especially a digital recorder, we try and make best
use of available bits. But the analog inputs to the unit from my
receiver were variable IAW the volume knob. So how did I ever know
that I was using all of the bits in the equalizer? The output meter
usually barely moved. The input I doubt had enough gain for 16 bits
to be filled up.

Where did you get this idea?

What makes you think "using all the bits" is significant or even
useful?

Do you always operate your power amplifier at full tilt for fear
you're not getting all the output ower it's capable of?
--scott

Scott, how often do you make digital recordings without the max gain
possible without clipping? Maybe in a 24 bit system you can go without the
top half dozen bits lighting up, but you wouldn't purposely try to record
at the 8 or 10 bit level.

Powers of 2... if you only used half of the 96 dB range you're still using
23 bits - think about that.

The analog section won't have the same dynamic range, and that's why you do
want to make sure the signal is up out of the mud.

Sean

[Ron C[_2_]]

On 8/10/2014 5:10 PM, geoff wrote:
On 10/08/2014 11:57 p.m., Scott Dorsey wrote:
Gary Eickmeier wrote:
So my main question would be: Whenever we have a digital component in
the
system, especially a digital recorder, we try and make best use of
available
bits. But the analog inputs to the unit from my receiver were
variable IAW
the volume knob. So how did I ever know that I was using all of the
bits in
the equalizer? The output meter usually barely moved. The input I
doubt had
enough gain for 16 bits to be filled up.

Where did you get this idea?

What makes you think "using all the bits" is significant or even useful?

Do you always operate your power amplifier at full tilt for fear you're
not getting all the output ower it's capable of?
--scott



But Scott, pretty much ALL power amplifiers are indeed running at 'full
tilt' all the time. Level control is an attenuator to the input signal,
no ?

;-)

geoff

IMHO, "capable of" is not the same as "running at."

Just sayin' ...

==
Later...
Ron Capik
--

[Ron C[_2_]]

On 8/10/2014 7:07 PM, Sean Conolly wrote:
"Gary Eickmeier" wrote in message
...
Scott Dorsey wrote:
Gary Eickmeier wrote:
So my main question would be: Whenever we have a digital component
in the system, especially a digital recorder, we try and make best
use of available bits. But the analog inputs to the unit from my
receiver were variable IAW the volume knob. So how did I ever know
that I was using all of the bits in the equalizer? The output meter
usually barely moved. The input I doubt had enough gain for 16 bits
to be filled up.

Where did you get this idea?

What makes you think "using all the bits" is significant or even
useful?

Do you always operate your power amplifier at full tilt for fear
you're not getting all the output ower it's capable of?
--scott

Scott, how often do you make digital recordings without the max gain
possible without clipping? Maybe in a 24 bit system you can go without the
top half dozen bits lighting up, but you wouldn't purposely try to record
at the 8 or 10 bit level.

Powers of 2... if you only used half of the 96 dB range you're still using
23 bits - think about that.

The analog section won't have the same dynamic range, and that's why you do
want to make sure the signal is up out of the mud.

Sean


Seen another way, you've gone from 16+ million possible
levels to 8+ million possible levels - think about that.

==
Later...
Ron Capik
--

[[email protected]]

Its 6 dB per bit

If you are 12 dB below full scale, you have 2 bits in reserve.

Maek

[PStamler]

Gary: Please do a test of the system's gain with and without the DEQ in place, with the switch set to +12, and again with the switch set to +22, and tell us the result. I still think the DEQ qas, in your original experiment, slightly attenuating the signal.

Oh, and you asked Scott how often he makes recordings without trying to use the whole dynamic range. I'm not Scott, and can't speak for him, but I often make recordings where the highest level is -20dBFS, and they sound fine. As they ought to, since the theoretical dynamic range of a 24-bit system under those circumstances is 124dB, meaning that whatever I'm recording is well above the muck and mud level.

[Gary Eickmeier]

Mike Rivers wrote:
On 8/10/2014 11:42 AM, Gary Eickmeier wrote:
It has a little button called the MAX switch that "raises the
maximum level present at the MAIN inputs/outputs from +12 to +22
dBu." I have no idea what that does, nor does the book tell me.

I explained it in my post earlier today.

Non-Google translation of "raises the maximum level
present at the MAIN inputs/outputs from +12 to +22 dBu." is

Changes the input sensitivity so that input clipping occurs at +12 dBu
or +22 dBu

You need to learn the language of manuals. Knowing something about
specs, conventions, and operating levels helps, too.

OK so at the +22 position, I can raise the gain going into it 10 dB more
than at the +12 position - and then there is a compensating 10 dB pad at the
Main Out jacks?

No wait a minute - that didn't make total sense either. Just thinking out
loud here...I want to put more gain into the AD converter. So I just turn up
the gain - but that makes the volume too loud in the room. So they let me
turn it up 10 dB more anyway but take if off the other end. Yah - I guess
that makes sense. So I want to keep it on the +22 position. So why don't
they just make it that way in the first place? Who needs to be limited to
the +12?

Gary

[Gary Eickmeier]

Sean Conolly wrote:
"Dave Plowman (News)" wrote in message
...
In article ,
Sean Conolly wrote:
If you're running a *very* low signal through the unit I could see
where quantization noise could start becoming a factor. I recall
your receiver is running at -10 consumer levels? As far as I know
the DEQ only runs at +4, and although you can adjust the volume
it's a digital control upstream of the D/A converter.

I/O levels are switchable between pro and domestic. Sadly only both
together. It can be useful if the in and out can be switched
independently.


Hmmm - check through the manual - and there it is! There's a switch
next to the RTA mic input labeled 'MAX', which I always thought was
for the mic input level. The manual says it's for switching the
input/output levels as you indicated.

Thanks for that - I had to use the text search on the manual to find
it though.

Sean

There are a lot of thngs about this manual that are like that - they seem to
be telling us something and then a minute later you are scratching your head
and have to go back in again. By playing with the unit and reading over
again you gradually get most of it. Probably a translation from German.

There is still one part that I don't get - the EQ function lets you draw
whatever curve you think is appropriate for your idea of a room curve. I
tried to memorize that curve, but it just will not work. I have to re-draw
it every single time I want to do another Automatic EQ. Do you know what I
mean Sean?

Gary

[PStamler]

On Sunday, August 10, 2014 7:13:06 PM UTC-6, Gary Eickmeier wrote:

There is still one part that I don't get - the EQ function lets you draw
whatever curve you think is appropriate for your idea of a room curve. I
tried to memorize that curve, but it just will not work. I have to re-draw
it every single time I want to do another Automatic EQ. Do you know what I

I don't know about the Automatic EQ, but as I recall the DEQ has a Save function, allowing you to save a particular room EQ curve. In fact, my recollection is that you can save several of them.

As far as why they included a function to switch from max=+12 to max=+22, well, back in the 1970s the Teac company was inventing "prosumer" recording equipment and the whole home-studio thing. At the time pro equipment was almost universally built with a nominal operating level of +4dBu; it had varying amounts of headroom, but it would typically clip at somewhere between +18dBu and +24dBu. Well, Teac was using cheap opamps in their new line of bargain-priced multitrack gear, and they couldn't handle professional levels very well, let along professional load impedances (typically 600 ohms).. So Teac, which soon became Tascam, invented and promulgated a new standard, wherein nominal level became -10dBV, which is about -7.8dBu. 20dB of headroom over that yields about +12dBu.

This new "standard" was not recognized by any professional standards committee, but it became the de facto standard for a lot of cheap gear which was really consumer-grade, but was marketed as "prosumer" or "semi-pro". That included a lot of Behringer gear.

So the switch on the DEQ is to make it alternately compatible with prosumer gear (in the +12dBu position) or real professional gear (+22dB position -- by the way, that's a bit puny as a clipping point for pro gear, but I digress). It's really a half-assed idea, because the DEQ will work just fine with prosumer levels, but as you discovered that doesn't make the meters light up, and people get upset over that.

If you're curious about what I mean by "nominal level", you really need to read the Yamaha Sound Reinforcement Handbook. You'll save yourself a lot of heartache and needless worry if you do. A lot.

And you still need to find out if the DEQ, set flat, is introducing a slight attenuation into the playback chain. I'm betting it is, and that's the reason why the sound improved when you removed it from the chain.

Peace,
Paul

[Gary Eickmeier]

PStamler wrote:
On Sunday, August 10, 2014 7:13:06 PM UTC-6, Gary Eickmeier wrote:

There is still one part that I don't get - the EQ function lets you
draw whatever curve you think is appropriate for your idea of a room
curve. I tried to memorize that curve, but it just will not work. I
have to re-draw it every single time I want to do another Automatic
EQ. Do you know what I

I don't know about the Automatic EQ, but as I recall the DEQ has a
Save function, allowing you to save a particular room EQ curve. In
fact, my recollection is that you can save several of them.

Yes, 64 of them. But the one I am talking about apparently can't be saved.
If you put it in one of the memories, and go into the Auto EQ function, you
then cannot retrieve a memorized curve for the target, you have to rebuild
your curve all over again. ****er.

As far as why they included a function to switch from max=+12 to
max=+22, well, back in the 1970s the Teac company was inventing
"prosumer" recording equipment and the whole home-studio thing. At
the time pro equipment was almost universally built with a nominal
operating level of +4dBu; it had varying amounts of headroom, but it
would typically clip at somewhere between +18dBu and +24dBu. Well,
Teac was using cheap opamps in their new line of bargain-priced
multitrack gear, and they couldn't handle professional levels very
well, let along professional load impedances (typically 600 ohms). So
Teac, which soon became Tascam, invented and promulgated a new
standard, wherein nominal level became -10dBV, which is about
-7.8dBu. 20dB of headroom over that yields about +12dBu.

Interesting. I had - and still have - the Teac 4 channel Simul Sync reel to
reel recorder for my film work. Yes, film. I shot Super 8 Sound sync sound
in stereo and surround sound and had to do mixdowns from several tracks in
sync with the projector.

This new "standard" was not recognized by any professional standards
committee, but it became the de facto standard for a lot of cheap
gear which was really consumer-grade, but was marketed as "prosumer"
or "semi-pro". That included a lot of Behringer gear.

So the switch on the DEQ is to make it alternately compatible with
prosumer gear (in the +12dBu position) or real professional gear
(+22dB position -- by the way, that's a bit puny as a clipping point
for pro gear, but I digress). It's really a half-assed idea, because
the DEQ will work just fine with prosumer levels, but as you
discovered that doesn't make the meters light up, and people get
upset over that.

If you're curious about what I mean by "nominal level", you really
need to read the Yamaha Sound Reinforcement Handbook. You'll save
yourself a lot of heartache and needless worry if you do. A lot.

Getting yet another book on audio won't tell me what the damn Behringer
manual is trying to tell me. Can I google up what you're talking about?

And you still need to find out if the DEQ, set flat, is introducing a
slight attenuation into the playback chain. I'm betting it is, and
that's the reason why the sound improved when you removed it from the
chain.

Peace,
Paul

Could be, but I am not doing some A/B instant comparison when I switch it in
and out. But I guess I could check it if you think.

Gary

[Sean Conolly]

wrote in message
...
Its 6 dB per bit

If you are 12 dB below full scale, you have 2 bits in reserve.

Much better - thank you.

Sean

[geoff]

On 11/08/2014 4:54 p.m., Gary Eickmeier wrote:


Getting yet another book on audio won't tell me what the damn Behringer
manual is trying to tell me. Can I google up what you're talking about?


Can you ? You'd need to check the manual for Google - but it may not be
100% clear !

;-)

geoff

[Phil W[_3_]]

"Mike Rivers":
On 8/9/2014 8:08 PM, Gary Eickmeier wrote:

But the analog inputs to the unit from my receiver were variable IAW
the volume knob. So how did I ever know that I was using all of the bits
in
the equalizer? The output meter usually barely moved. The input I doubt
had
enough gain for 16 bits to be filled up.

Again, RTFM and explore the controls. The Meter menu turns the display
into three pages worth of metering options, including showing the input
level.

Well, most people in this group should have learned a while ago, that Mr
Know-it-all-best is not able to RTFM and conclude from what´s written there.
No matter, which subject. This particular moron refuses to go along well
with concepts and devices, which work well for most (if not all) other
users.

[Dave Plowman (News)]

In article ,
PStamler wrote:
As far as why they included a function to switch from max=+12 to
max=+22, well, back in the 1970s the Teac company was inventing
"prosumer" recording equipment and the whole home-studio thing. At the
time pro equipment was almost universally built with a nominal operating
level of +4dBu; it had varying amounts of headroom, but it would
typically clip at somewhere between +18dBu and +24dBu. Well, Teac was
using cheap opamps in their new line of bargain-priced multitrack gear,
and they couldn't handle professional levels very well, let along
professional load impedances (typically 600 ohms). So Teac, which soon
became Tascam, invented and promulgated a new standard, wherein nominal
level became -10dBV, which is about -7.8dBu. 20dB of headroom over that
yields about +12dBu.

Interesting hypothesis. I'd always assumed that domestic line level was
loosely based around what existed at the output of an AM radio receiver
and the input to its AF stages in the valve days. And it doesn't seem odd
to me that a maker like Teac might adopt this (near enough) too for
something which wasn't specifically aimed at the then pro market.

Pro line levels date back to the early days of telephony, and I'd say it
was first adopted by broadcasters who had to interface with this, rather
than for any intrinsic reason.

--
*Reality is a crutch for people who can't handle drugs.

Dave Plowman London SW
To e-mail, change noise into sound.

[None]

"Gary Eickmeier" wrote in message
...
Getting yet another book on audio won't tell me what the damn
Behringer
manual is trying to tell me.

Woooooooshhhhhh!

[None]

"Phil W" wrote in message
...
Well, most people in this group should have learned a while ago,
that Mr Know-it-all-best is not able to RTFM and conclude from
what´s written there.
No matter, which subject. This particular moron refuses to go along
well with concepts and devices, which work well for most (if not
all) other users.

He whines that he refuses to read the standard Yamaha text because it
won't explain the Behringer manual to him. Multi-level stupidity.

[Mike Rivers[_2_]]

On 8/10/2014 9:07 PM, Gary Eickmeier wrote:
...I want to put more gain into the AD converter. So I just turn up
the gain - but that makes the volume too loud in the room. So they let me
turn it up 10 dB more anyway but take if off the other end. Yah - I guess
that makes sense. So I want to keep it on the +22 position. So why don't
they just make it that way in the first place? Who needs to be limited to
the +12?

It's really simple, but you need to understand the difference between
gain and sensitivity.

You have a source with a fixed output level. According to the DEQ's
meter, it's too low for good performance. Look at the MAX switch. If
it's in the +22 position, set it to the +10 position and you'll see a
higher level on the DEQ's meters. If it's already set to +10 then that's
all the level you'll get. Live with it or put another stage of
amplification between your source and the DEQ.

The unit maintains (with everything bypassed) unity gain between input
and output, so the the level coming out will be the same as the level
going in. So whether the switch is set for +10 or +22, you won't get any
loss through the unit, but you won't get any gain, either.

--
For a good time, visit http://mikeriversaudio.wordpress.com

[Mike Rivers[_2_]]

On 8/11/2014 12:54 AM, Gary Eickmeier wrote:
PStamler wrote:
I don't know about the Automatic EQ, but as I recall the DEQ has a
Save function, allowing you to save a particular room EQ curve. In
fact, my recollection is that you can save several of them.

Yes, 64 of them. But the one I am talking about apparently can't be saved.
If you put it in one of the memories, and go into the Auto EQ function, you
then cannot retrieve a memorized curve for the target, you have to rebuild
your curve all over again. ****er.

That makes sense to me. When you're using Auto EQ, the DEQ is making the
decision as to what the curve should be, at least initially. In that
mode, it doesn't care about a curve that you used at one time.

Getting yet another book on audio won't tell me what the damn Behringer
manual is trying to tell me. Can I google up what you're talking about?

Paul's explanation of nominal operating levels is excellent. Study that.
It wouldn't hurt if you were able to make some actual measurements on
your system to see if Paul's suggestion that there's some loss through
the DEQ and the lower listening level is throwing you off. In lieu of
buying test equipment, try turning the volume up just a tad when you
have the DEQ in line and see if that fixes all of your problems in one
simple step.

I'm not going to read the manual to you, but look closely to see if
there's a control for output level in addition to the MAX switch. It
would be functionally similar to the make-up gain control that's found
on most compressors that allows you to bring the output level back to
nominal after the compressor reduces the signal level as it does its job.

You might get a better understanding of levels and how they're measured
by reading the Meter Madness article on my web site.



--
For a good time, visit http://mikeriversaudio.wordpress.com

[Mike Rivers[_2_]]

On 8/11/2014 6:09 AM, Dave Plowman (News) wrote:

I'd always assumed that domestic line level was
loosely based around what existed at the output of an AM radio receiver
and the input to its AF stages in the valve days. And it doesn't seem odd
to me that a maker like Teac might adopt this (near enough) too for
something which wasn't specifically aimed at the then pro market.

Actually, most consumer audio gear from before the solid state era
worked at a level somewhere between -20 and -15 dBu. TASCAM liked -10
dBV so users wouldn't be confuse it with dBm (power). They also promoted
the "voltage matched" interfacing concept of having a relatively low
output impedance feeding a relatively high input impedance. With this
setup, very little voltage is dropped across the output impedance and
nearly the full (open circuit) output voltage is what the input sees.

In a "power matched" connection with a 600 ohm source feeding a 600 ohm
input, while that provided the maximum power transfer, half the open
circuit voltage was dropped across the output when it was connected to a
load of equal impedance.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Scott Dorsey]

Gary Eickmeier wrote:

No wait a minute - that didn't make total sense either. Just thinking out
loud here...I want to put more gain into the AD converter. So I just turn up
the gain - but that makes the volume too loud in the room. So they let me
turn it up 10 dB more anyway but take if off the other end. Yah - I guess
that makes sense. So I want to keep it on the +22 position. So why don't
they just make it that way in the first place? Who needs to be limited to
the +12?

If you never use the digital inputs and outputs, there is no reason to ever
use the +12 position. However, if you use the digital inputs and outputs,
you may want your digital levels to correspond with some known analogue level
and that level may vary from facility to facility since there are two common
standards.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."