"Sander" wrote in message
news:5YX4c.4773$EV2.34823@amstwist00...
Preben Friis wrote:

Take a look at: http://www.rolls.com/data/adi6man.pdf

I can see how the placement of text in that schematic confuses you but
it says:

R1: 10K
Attenuator: P100K FX

Oh... my bad. I just remembered how I calculated that the impedance was
wrong, but did not remember why so I took a too quick look on the schematic
and noted the wrong value.

Using a 100k pot instead, parallel with R3 (100k) parallel with the input
impedance of the actual amplifier, can you get to 100k total impedance?

/Preben Friis

Preben Friis wrote:


Oh... my bad. I just remembered how I calculated that the impedance was
wrong, but did not remember why so I took a too quick look on the schematic
and noted the wrong value.

Using a 100k pot instead, parallel with R3 (100k) parallel with the input
impedance of the actual amplifier, can you get to 100k total impedance?

/Preben Friis

First you encounter 10k (R1 or R2) in series with the rest of the circuit.

After that you have the attenuator (100K) and R3(100K).
the resistance of this combination depends on the slider position of the
pot and is somewhere between 50K (least attenuation) and 100K (max
attenuation) for a total imput impedance of somewhere between 60 and 110
K. That will make up the main part.

Then in parallel with R3 we still have the rest of the circuit; mainly
C1, R4 and R6 which is where we are starting to go into frequency
dependant behaviour in comparison with the (at least theoretically)
purely resistive parts up to here.

Sander

"Sander" wrote in message
news:4eZ4c.4793$EV2.34799@amstwist00...
First you encounter 10k (R1 or R2) in series with the rest of the circuit.

After that you have the attenuator (100K) and R3(100K).
the resistance of this combination depends on the slider position of the
pot and is somewhere between 50K (least attenuation) and 100K (max
attenuation) for a total imput impedance of somewhere between 60 and 110
K. That will make up the main part.

Right... but at 110k you will have eliminated the signal. When using a DI
with passive transducers it is not likely that the attentuator is used at
all.

Then in parallel with R3 we still have the rest of the circuit; mainly
C1, R4 and R6 which is where we are starting to go into frequency
dependant behaviour in comparison with the (at least theoretically)
purely resistive parts up to here.

Yes... and that will lower the impedance even more. I suspect that R4 and R8
is not 3k3 as shown.

Actually the more I look at that shematic, the more confused I get. It looks
like a long tailed pair, but instead of having the emitters connected to a
constant current source it has the collectors connected to a LED wich
delivers is constant voltage drop. Without ground lift, Q2 will never
generate any signal, so the output signal is not balanced. The voltage to
the output is sourced by 4k7 resistors, so this will effectively limit the
output impedance to a point higher than that. Connect the output to a low
impedance mic input and it will distort pretty much.

The result is a box with a low input impedance, high output impedance, it
dampens the signal, distorts and does not balance it. Not what I would
expect from an "Active DI" that according to Recording Mag has "clean
sound".

This can't be true. Please, someone... tell me where I went wrong ...

/Preben Friis

"Preben Friis"
Yes, they will gladly tell you that it's horrible, it's crappy, it sucks
..... about something that they have never heard. That is what I am
attempting to fight here.

Preben - look at the title of this thread. Behringer amps clearly do
not "Rule", nothing at any low price point does actually, it's that kind of
"Behringer Rules" sales hyperbole that I find objectionable. People who are
not group regulars appear here from time to time with little to say except to
shill Behringer gear in this manner, and I think probably gear is being handed
out to guys if they will help spread some positive "buzz" and we're just lucky
enough to be on the "list".

But you've been having a nice substantive discussion on the specs of the
ADA8000 and no one has a problem with that, there are many other examples, no
one says "all Behringer gear is horrible, is crappy" as a blanket response to
real evaluations of gear.

I think if Behringer really cared what the RAP community strata of users
thought they'd wise up and have someone like Ty Ford or Harvey Gerst evaluate
some of their gear. But it appears they don't really care what we think, ok
fine. Except that when they have minions drop in postings full of juvenile
"Behringer Rules" tripe from time to time, that's a bit insulting really. And
that does suck...


Will Miho
NY Music & TV Audio Guy
Off the Morning Show! & sleepin' In... / Fox News
"The large print giveth and the small print taketh away..." Tom Waits

Arny Krueger wrote:

He has a pair of AudioRails. Delighted to death with 'em. He has this odd
idea that CAT-5 is easier to manage than a 24 channel recording snake. He
has both, so I guess we should believe him. ;-)

He also has a 24 track ADAT-interface recorder.

Maybe the AudioRail folks should also point out the synergy between 3216s
and AudioRails if they haven't..

This seems to be a mongo-high tech solution for live sound and location
recording on a tiny budget.

As long as you need 24 tracks or less, I think it's the price leader. Now how about a PC card that will turn that RJ45 into an ASIO driver? Or maybe he can hack up a driver for some existing Ethernet card? Dunno what sort of signaling he's using but I doubt it's Ethernet.

"Kurt Albershardt" wrote in message

Arny Krueger wrote:

He has a pair of AudioRails. Delighted to death with 'em. He has
this odd idea that CAT-5 is easier to manage than a 24 channel
recording snake. He has both, so I guess we should believe him. ;-)

He also has a 24 track ADAT-interface recorder.

Maybe the AudioRail folks should also point out the synergy between
3216s and AudioRails if they haven't..

Note to myself: They did. There's a diagram on the site that has ADA8000s,
3216s and other stuff on it.

This seems to be a mongo-high tech solution for live sound and
location recording on a tiny budget.

As long as you need 24 tracks or less, I think it's the price leader.

OK, I would like to be educated. What happens at 24 tracks?

Now how about a PC card that will turn that RJ45 into an ASIO driver?

Seems feasible, doesn't it?

It seems like the ADAT interface might live longer than the ADAT.

Or maybe he can hack up a driver for some existing Ethernet card?
Dunno what sort of signaling he's using but I doubt it's Ethernet.

The web site just about says that it is not Ethernet in so many words

Arny Krueger wrote:

This seems to be a mongo-high tech solution for live sound and
location recording on a tiny budget.

As long as you need 24 tracks or less, I think it's the price leader.

OK, I would like to be educated. What happens at 24 tracks?

Nothing, now that the € has annihilated our poor Dollar. The RME ADI-648 used to run around $1200 for 64 channels.





maybe he can hack up a driver for some existing Ethernet card?
Dunno what sort of signaling he's using but I doubt it's Ethernet.

The web site just about says that it is not Ethernet in so many words

Just maybe he's using the Ethernet PHY layer (as Sony does,) in which case there might be hope.

TonyP wrote:

People here just don't seem to understand the correlation between gain, and
S/N ratio.

Give a listen to the Gordon preamp at any gain setting...

--
ha

"WillStG" wrote in message
...
Preben - look at the title of this thread. Behringer amps clearly
do
not "Rule", nothing at any low price point does actually, it's that kind
of
"Behringer Rules" sales hyperbole that I find objectionable. People who
are
not group regulars appear here from time to time with little to say except
to
shill Behringer gear in this manner, and I think probably gear is being
handed
out to guys if they will help spread some positive "buzz" and we're just
lucky
enough to be on the "list".

I totally agree, those posts suck. They look like the work of
overenthusiastic teenagers. I really can't figure out why you are so worked
up about those. Actually they do more harm than good to the products they
describe.

But you've been having a nice substantive discussion on the specs of
the
ADA8000 and no one has a problem with that, there are many other examples,
no
one says "all Behringer gear is horrible, is crappy" as a blanket response
to
real evaluations of gear.

I've seen countless opinions about Behringer stuff here, that was never
based on any evaluations. I bet if someone asked if "Is Behringer XX6000
Ultra good for me" someone would reply "It sucks and it is copied from
someone else" even if that product didn't exist.
You yourself wrote another post in this thread, that I chose not to respond
to, since you also took the low road and drew conclusions of a product you
have never tried.

Quote from that post:

Be sure to measure it with your lab
equipment too, then post in terms we can all understand, like "Behringer
Bites"
and "Behringer sucks!" and

As the RMAA test proved, the dynamic range is close to the 100 dB
specification, which is more than what can be expected for the cheapest 8
channel ADA currently on the market. Actually it has better specs than the
best analog tape recorders, haven't it? So no, it does not suck, it does not
bite, but is sure stings that with cheap modern technology you can make
stuff that kicks ass to anything that was made ten years ago. (Now I'm
starting to sound like a commercial, so I'll stop here.)

I think if Behringer really cared what the RAP community strata of
users
thought they'd wise up and have someone like Ty Ford or Harvey Gerst
evaluate
some of their gear. But it appears they don't really care what we think,
ok
fine. Except that when they have minions drop in postings full of
juvenile
"Behringer Rules" tripe from time to time, that's a bit insulting really.
And
that does suck...

You mean the same Ty Ford, that responded with "To yoy sir, a high-tech
MEOW! And thanks for your vigilance." to a post that compared the ADA8000 to
digested tuna without hearing it? That would surely be an objective test ...
or not.

Yes... the "rules" posts suck... but I bet your newsreader has a block
function. I do think that they do care for constructive criticism, but as if
there are no reasonable way to respond to "Will Miho is an asshole", there
are no point for Behringer to respond to every "Behringer sucks" post. That
would just create an endless flamewar.
Have you ever seen an AKG representative respond to a S-x000 sucks post?
Have you seen Digidesign people respond to the criticism here? How did it go
with the Bombfactory guy? We don't see a lot of him around here any more....

Enough for now....

/Preben Friis

http://www.audiorail.com/ADA8000_RMAA_test.zip (406 KB)

Just getting back to this. The THD and IMD numbers and spectrals were
suspicious, so I looked at the waveforms, and they were clipped, even
though at -1 dBFS. I don't know whether this is the Behringer or RMAA.

Do you have any ideas on this, Arny?

The new files re-posted above are of waveforms that pretty closely match
the output waveform amplitude with the RMAA test file waveform
amplitude, which is now at about -3 dBFS, as you suggested. This cleans
up the THD and IMD results.

Check this over.

Well, if it's the XR line of rackmount mixers, I've almost got 3 years on
mine (10/29/01) and there's only been excellent sound coming out of it. A
well thought out design that seems to not mind if it's moved around, even
though they are ideally for install work. I'm quite pleased that I didn't
go for the Venice.

--


Roger W. Norman
SirMusic Studio

"hank alrich" wrote in message
.. .
Scott Dorsey wrote:

The installed sound guys don't care so much about sound quality, but
they
are VERY sensitive to reliability issues. They want gear that is
intended
to be as bulletproof as possible, because it costs them a huge amount of
money to come out and replace something that fails under warranty. A
couple
service calls can turn a profitable contract into an unprofitable one
very
fast. And service calls once the warranty period has expired do not
make
customers happy.

Very different than the typical consumer/semi=pro customer.

This morning in the BSW catalog I noted the Crest rackmount mixers
offering a 5-year warranty. That'd appeal to the install guys, if the
kit lives up to the guarantee. And it just might.

--
hank alrich * secret__mountain
audio recording * music production * sound reinforcement
"If laughter is the best medicine let's take a double dose"

I saw, although have not researched, that Whirlwind has a cat5 system
similar to the AudioRail. I believe it's called the ESnake, but I doubt
it's a $500 64 channel solution.

Yep, just found it at http://www.whirlwindusa.com/esnake.html. Looks a
little high-end. And so what if the snake is only 7 pounds for 330 feet.
The unit looks like it makes up the difference! g

--


Roger W. Norman
SirMusic Studio

"Arny Krueger" wrote in message
...
"TonyP" wrote in message


"Arny Krueger" wrote in message
...

I met with a
friend last night who lives way 'cross town. He just happened to
mention that he has 3 ADA 8000s that he uses with the Behr digital
mixer. He's happy as a clam with the real world performance.

Doesn't surprise me at all, but some people do love to hate Behringer
:-)

How do you use 3 ADA8000's with one 3216 though?

Good question. The 3216 isn't the only ADAT-interface hardware that he
uses.

He has a pair of AudioRails. Delighted to death with 'em. He has this odd
idea that CAT-5 is easier to manage than a 24 channel recording snake. He
has both, so I guess we should believe him. ;-)

He also has a 24 track ADAT-interface recorder.

Maybe the AudioRail folks should also point out the synergy between 3216s
and AudioRails if they haven't..

This seems to be a mongo-high tech solution for live sound and location
recording on a tiny budget.

In article ,
"Roger W. Norman" wrote:

Well, if it's the XR line of rackmount mixers, I've almost got 3 years on
mine (10/29/01) and there's only been excellent sound coming out of it. A
well thought out design that seems to not mind if it's moved around, even
though they are ideally for install work. I'm quite pleased that I didn't
go for the Venice.

This morning in the BSW catalog I noted the Crest rackmount mixers
offering a 5-year warranty. That'd appeal to the install guys, if the
kit lives up to the guarantee. And it just might.

I would instal the Crest, I will not instal A&H gl series
George

George" wrote in message
...
I would instal the Crest, I will not instal A&H gl series

Oh, I'd damned sure install the Crest. No two ways about it. But I've just
as often thought about buying one more and using the bus multiplexer for
some of the gigs I do. I could always use 24 mono pres like these with 16
more stereo channels (8 more mono) that could be bussed to a recorder. I'm
not afraid to pre-mix! g I've been quite happy. JohnnyV and I did a gig
today and with 6 female a cappella going through SM57s, it was marvelous
(well, they did use a djembe, so I guess it wasn't totally without
instrument).

And with 57s. There's not a product in most lines under the quality of
Harrisons, Soundcrafts and others of that ilk that I've heard present 57s on
vocals pristinely. About where I would put it is somewhat shy of a John
Hardy with a 57, but not much, and that's saying a lot. You ought to hear
this baby with a Neumann KMS 105! g

--


Roger W. Norman
SirMusic Studio

" In article ,
"Roger W. Norman" wrote:

Well, if it's the XR line of rackmount mixers, I've almost got 3 years
on
mine (10/29/01) and there's only been excellent sound coming out of it.
A
well thought out design that seems to not mind if it's moved around,
even
though they are ideally for install work. I'm quite pleased that I
didn't
go for the Venice.

This morning in the BSW catalog I noted the Crest rackmount mixers
offering a 5-year warranty. That'd appeal to the install guys, if the
kit lives up to the guarantee. And it just might.

George

while it's routing and sound can't be faulted
your a better man than I am gung-a-din
never going to buy another desk with back mounted patch points(at least
not analouge patch points)
my gl totally frustrated me and the conversion (to standard desk style
patching)was a real PITA
George

Well, for a rack mount you simply have to improvise, but I got an idea from
a TLA unit that was straight 8 inputs and sat in a box angled on the top
with a slot on the bottom to accomodate cabling. You can easily pull the
box upright to affix cables, lay it down and mix. It's definitely a PITA
with the GigRig, but I use that for festivals and not with day setups and
I've found full peacefulness and being "Gung-a-din", I am a better man! g

There must be a reason that every job JohnnyV and I do needs to have the
Crest involved, along with the Mackie 1530s, although I know you don't like
them all that much. Hey, he choses these over his Klipsch and for the jobs
we do, the Mackies work just fine. I certainly won't think about putting
them down at the Kennedy Center if I get that job this year, but for crowds
of 300 to 500 they work. I have yet to push the Crest and the Mackies run
nice and cool. That's probably something that can't be said of others using
the Mackies and a Mackie 1604. Calm, cool and quiet. John came over today
and asked me if I'd heard distortion on some of the vocals and I had to look
at him like he just wasn't thinking. I don't believe I've ever been
involved with a local situation where the sound guys got so much notice and
an onstage thanks for the job.

You and I will have to talk about these Servodrives later, which is what I'm
planning on putting up at the Kennedy Center.

--


Roger W. Norman
SirMusic Studio

"George" wrote in message
...
while it's routing and sound can't be faulted
your a better man than I am gung-a-din
never going to buy another desk with back mounted patch points(at least
not analouge patch points)
my gl totally frustrated me and the conversion (to standard desk style
patching)was a real PITA
George

In article ,
"Roger W. Norman" wrote:

Well, for a rack mount you simply have to improvise, but I got an idea from
a TLA unit that was straight 8 inputs and sat in a box angled on the top
with a slot on the bottom to accomodate cabling. You can easily pull the
box upright to affix cables, lay it down and mix. It's definitely a PITA
with the GigRig, but I use that for festivals and not with day setups and
I've found full peacefulness and being "Gung-a-din", I am a better man! g


my solution to the gl2 was to build a desk stand with a dupilcate of the
patch panel, every connector.
while I was at it I paralled a multipin for the snake to the xlrs
all this mounted on the rear of the desk stand
George

"Arny Krueger" wrote in message
...
It seems like the ADAT interface might live longer than the ADAT.

Already the case isn't it :-)

TonyP.

"hank alrich" wrote in message
...
TonyP wrote:

People here just don't seem to understand the correlation between gain,
and
S/N ratio.

Give a listen to the Gordon preamp at any gain setting...

I'm sorry I'm not sure what your point is? Have they discovered a way to
defy the laws of physics in some way I'm not aware of?
Or does it use cryogenic cooling perhaps? Or maybe it just has very low
gain?

TonyP.

"Garth D. Wiebe" wrote in message


http://www.audiorail.com/ADA8000_RMAA_test.zip (406 KB)

Just getting back to this. The THD and IMD numbers and spectrals were
suspicious, so I looked at the waveforms, and they were clipped, even
though at -1 dBFS. I don't know whether this is the Behringer or
RMAA.

RMAA AFAIK just doesn't do this. You can do a virtual loop-back test with
RMAA by making a test file and immediately analyzing it without re-recording
it. This would show any errors in RMAA. I've done this and found no
problems.

Do you have any ideas on this, Arny?

I've seen this before. There's probably some minor gain-staging errors
inside the ADA8000. I suggested this before based on your much earlier
reports. This is not unusual.

Just another chapter from my book called "Don't record so %$#!! close to FS"

The new files re-posted above are of waveforms that pretty closely
match the output waveform amplitude with the RMAA test file waveform
amplitude, which is now at about -3 dBFS, as you suggested. This
cleans up the THD and IMD results.

And that's pretty much that. If you use RMAA in real-time mode, it sets the
peak level for the THD and IM tests so they max out around -3 dB FS. AFAIK,
this is consistent with AES recommendations for testing sound cards.

Some more tips about RMAA. It will automatically generate HTML of a web page
that if nothing else, is a good starting point for editing up one that meets
your own preferences. RMAA will save each test plot as a .PNG file which is
highly-compressed and should be displayable by anybody with a modern web
browser. You can mix and match these two facilities to quickly build a nice
customized online report. You can also save the test in a format that lets
any RMAA user modify his view of the data to suit including magnifying
certain ranges, as you have done. Saved test files can also be used to
compare products, or plot the performance of a product over a range of
operating conditions.

Finally, I prefer to run the IM test as a two-tone high frequency test
composed of 18 and 20 KHz if possible. There is an option in RMAA to modify
the test tones over a goodly range.

Here's my first shot at putting a RMAA test online:

http://www.pcavtech.com/pwramp/boostaroo/

Check this over.

Thanks for the new report.

Obviously, the actual clipping point is someplace between -3 dB and FS which
you can determine by experimentation. Here is an example of this kind of
test that I did and posted:

http://www.pcavtech.com/soundcards/L...R-vs-level.gif

I think this last retest clarifies *everything* or at least LOTS.

The Behringer spec sheet might be a tad optimistic, but it is close.

A set of tests at various levels between - 3 DB FS and FS could nail down
the actual clipping point. This might add a dB or two to the measured SNR
and DR performance, further closing the gap.

However, we should probably add about 3 dB to the DR & SNR performance
because it is a loopback test in order to estimate the performance of the
input or output side of the ADA 8000 when taken all by itself.

This makes the spec versus actual distance even smaller.

I don't see any reason to apologize for recommending the ADA8000.

IMO the ADA8000 measures well enough to be considered for critical
professional applications, based on just measurements. Perceived sound
quality is always the final test, but people I respect say it sounds very
good.

Really bad measurements IMO can disqualify a product for critical use. If it
sounds good when you use it, that should be enough. If it sounds really bad,
the problem is most likely someplace else than the ADA8000.

Garth D. Wiebe wrote:

suspicious, so I looked at the waveforms, and they were clipped, even
though at -1 dBFS. I don't know whether this is the Behringer or RMAA.

Do you have any ideas on this, Arny?

Just answering my own question, of course it could not be RMAA, because
it is the raw output from the ADA8000. Also, the clipping is not
perfectly digitally flat. So the ADA8000 must simply not be able to
drive all the way to 0 dBFS.

We were obviously sitting composing our posts at the same time.

Thanks again, Arny, for the additional pointers on the RMAA test.

I will add these to the post back at the ProSoundWeb review forum.

"Garth D. Wiebe" wrote in message

Garth D. Wiebe wrote:

suspicious, so I looked at the waveforms, and they were clipped, even
though at -1 dBFS. I don't know whether this is the Behringer or
RMAA.

Do you have any ideas on this, Arny?

Just answering my own question, of course it could not be RMAA,
because it is the raw output from the ADA8000. Also, the clipping is
not perfectly digitally flat. So the ADA8000 must simply not be able
to drive all the way to 0 dBFS.

Not an unusual situation. There was some minor error in gain-staging. Or
perhaps the mismatch was intentional to force clipping into the analog
domain where the particular parts used would produce cleaner clipping. I've
definitely seen analog-to-digital converters that were great until they
clipped and then went crazy. Clean clipping in an earlier stage can make a
part like this more practically usable, by never letting it go where it gets
into trouble.

I believe I had talked about doing just that in the Gig Rig, but I would
have done it on a 130 pin multi-connector, or two smaller connectors, one
running to a second rack, and it would have run about $1k, so I decided to
forget that idea! g Most of my stuff is kinda "mini" install where I have
the whole shebang loaded in for 4 or 5 days, so that works kinda well. But
for quick and easy, the box works just fine.

--


Roger W. Norman
SirMusic Studio

"George" wrote in message
...
In article ,
"Roger W. Norman" wrote:

Well, for a rack mount you simply have to improvise, but I got an idea
from
a TLA unit that was straight 8 inputs and sat in a box angled on the top
with a slot on the bottom to accomodate cabling. You can easily pull
the
box upright to affix cables, lay it down and mix. It's definitely a
PITA
with the GigRig, but I use that for festivals and not with day setups
and
I've found full peacefulness and being "Gung-a-din", I am a better man!
g


my solution to the gl2 was to build a desk stand with a dupilcate of the
patch panel, every connector.
while I was at it I paralled a multipin for the snake to the xlrs
all this mounted on the rear of the desk stand
George

"Roger W. Norman" wrote in message


I saw, although have not researched, that Whirlwind has a cat5 system
similar to the AudioRail. I believe it's called the ESnake, but I
doubt it's a $500 64 channel solution.

Yep, just found it at http://www.whirlwindusa.com/esnake.html . Looks
a little high-end. And so what if the snake is only 7 pounds for 330
feet. The unit looks like it makes up the difference! g

In all seriousness, I think you've scored a number of significant points
against the eSnake.

There doesn't seem to be a price for the eSnake *anyplace* on the web. That
big heavy full-o-parts box won't sell for no piddlin' $500.

I'm looking at preliminary design parameters for a rebuild of my church's
sanctuary. I can think of only a few reasons why there wouldn't be an
AudioRail system in it. They run along the lines of AudioRail going wheels
up before I get to the stage where we start buying stuff (still several
years down the road).

IOW, AudioRail looks like the total bomb for a room of any reasonable size
and complexity.

Of course I'm a gear slut and anything new excites me, especially if it's
affordable enough so I think I can buy some.

;-)

Arny Krueger wrote:
"Garth D. Wiebe" wrote in message


Garth D. Wiebe wrote:


suspicious, so I looked at the waveforms, and they were clipped, even
though at -1 dBFS. I don't know whether this is the Behringer or
RMAA.

Do you have any ideas on this, Arny?

Just answering my own question, of course it could not be RMAA,
because it is the raw output from the ADA8000. Also, the clipping is
not perfectly digitally flat. So the ADA8000 must simply not be able
to drive all the way to 0 dBFS.


Not an unusual situation. There was some minor error in gain-staging. Or
perhaps the mismatch was intentional to force clipping into the analog
domain where the particular parts used would produce cleaner clipping. I've
definitely seen analog-to-digital converters that were great until they
clipped and then went crazy. Clean clipping in an earlier stage can make a
part like this more practically usable, by never letting it go where it gets
into trouble.

Yes, I was thinking the same thing. The waveform edge corners are
nicely rounded, and if you look at the old THD spectral plots, the peaks
are down below -27 dBFS. Audible, to be sure, but I would expect worse
from a perfect digital clip, especially in the higher end of the spectrum.

"Garth D. Wiebe" wrote in message

Arny Krueger wrote:

"Garth D. Wiebe" wrote in message


So the ADA8000 must simply not be
able to drive all the way to 0 dBFS.

Not an unusual situation. There was some minor error in
gain-staging. Or perhaps the mismatch was intentional to force
clipping into the analog domain where the particular parts used
would produce cleaner clipping. I've definitely seen
analog-to-digital converters that were great until they clipped and
then went crazy. Clean clipping in an earlier stage can make a part
like this more practically usable, by never letting it go where it
gets into trouble.

Yes, I was thinking the same thing. The waveform edge corners are
nicely rounded,

You've got the advantage on me, I haven't seen the actual test waves.

and if you look at the old THD spectral plots, the
peaks are down below -27 dBFS.

That would be a consequence of the amount being clipped off being pretty
small. I estimated the mismatch as being on the order of 0.2 dB. This is
very typical. Only the finest ADCs can, IME go the last 0.2 dB without some
excess distortion. And you know what, it really shouldn't matter. In a
well-designed system, the expected peaks should be at least 10 dB below
peak. That leaves room for the *unexpected* peaks! ;-)

Audible, to be sure, but I would
expect worse from a perfect digital clip, especially in the higher
end of the spectrum.

Again, its not unusual for the clipping point to be different at various
frequencies, and lower at the highest frequencies.

This can come from a number of things. In a modern ADC there's a digital
filter. It's not unusual for there to be clipping in the digital filter. The
digital filter has more parameters being summed at high frequencies as a
rule, with complex timing that causes the signals being summed to be either
in-phase or out-of-phase or someplace in-between. At frequencies where a lot
of the signals being summed in the digital filter tend to be in-phase, the
probability of dynamic range problems is increased.

The *solution* is to build digital filters with accumulators and parameters
with lots of bits. This eats up silicon. Designing really-pretty-good
digital filters has been greatly facilitated by tools like Matlab. But, the
costs of implementing them ultimately comes down to square inches of
silicon.

The harmonics from a perfect digital clip roll off at 6 dB per octave. My
simulations of slight amounts of digital clipping with bias showed this
roll-off.

It strikes me, though, that sometime last night I read on the AudioRail site
that a full 64 channels (32/32) would run some $2.6 k, but then I was
talking to JohnnyV yesterday during a job about the unit and I could only
come up with having seen the $500 figure, sans converters. Now I'm confused
and haven't had enough coffee to do research today! g

--


Roger W. Norman
SirMusic Studio

"Arny Krueger" wrote in message
...
"Roger W. Norman" wrote in message


I saw, although have not researched, that Whirlwind has a cat5 system
similar to the AudioRail. I believe it's called the ESnake, but I
doubt it's a $500 64 channel solution.

Yep, just found it at http://www.whirlwindusa.com/esnake.html . Looks
a little high-end. And so what if the snake is only 7 pounds for 330
feet. The unit looks like it makes up the difference! g

In all seriousness, I think you've scored a number of significant points
against the eSnake.

There doesn't seem to be a price for the eSnake *anyplace* on the web.
That
big heavy full-o-parts box won't sell for no piddlin' $500.

I'm looking at preliminary design parameters for a rebuild of my church's
sanctuary. I can think of only a few reasons why there wouldn't be an
AudioRail system in it. They run along the lines of AudioRail going wheels
up before I get to the stage where we start buying stuff (still several
years down the road).

IOW, AudioRail looks like the total bomb for a room of any reasonable size
and complexity.

Of course I'm a gear slut and anything new excites me, especially if it's
affordable enough so I think I can buy some.

;-)

"Roger W. Norman" wrote in message


It strikes me, though, that sometime last night I read on the
AudioRail site that a full 64 channels (32/32) would run some $2.6 k,

I was under that impression for a while, too.

but then I was talking to JohnnyV yesterday during a job about the
unit and I could only come up with having seen the $500 figure, sans
converters. Now I'm confused and haven't had enough coffee to do
research today! g

The guy I know who actually has a working AudioRail setup said last Saturday
night that he bought two $500 boxes from them to have a working setup. I
think we were each only one beer down the line at the time, and this was my
one and only for the night, so this is probably a pretty good number.

;-)

The picture on the AudioRail site also seems to say that two $500 boxes are
needed. I'm convinced!

Dear Tony,

the mic preamps in the ADA8000 do not use op-amps - they are based on
a low noise conjugate-pair transistor pre-amp circuit similar to that
used in our mixing consoles.
We do utilise TL074 op-amps in the line-level circuitry, and the LM339
is used in its intended role as a comparator.

Best regards,

Andy Cullen
Customer Support
BEHRINGER Spezielle Studiotechnik GmbH






"TonyP" wrote in message . au...
"Arny Krueger" wrote in message
...

And that ain't worst case.

And it aint the best case either!
People here just don't seem to understand the correlation between gain, and
S/N ratio.


Looking at

http://srforum.prosoundweb.com/viewt...944cdcb38834bc
b087811cf1a ,
I find the following semiconductor compliment:

Alesis Semiconductor AL1101 A/D converters
Alesis Semiconductor AL1201 D/A converters
Alesis Semiconductor AL1402 ADAT optical decoder
Alesis Semiconductor AL1401A ADAT optical encoder
Toshiba TORX176 Toslink receiver
Toshiba TOTX176 Toslink transmitter
ST Microelectronics TL074C and LM339 op amps

Many of these parts are digital interface chips which would be less
suspect
in terms of their effect on analog signal quality. I've seen spec sheets
for
the Alesis parts, but little else. Therefore they are suspect to me. I
don't
have any special problems with TL074s as moderate-impedance, moderate
gain,
line level parts. I've got questions about the use of them and LM339s as
mic
preamps.


I would be very surprised if the ADA8000 didn't use 4580 op-amps for the mic
pre's like all their other mixers.

Remember the thing is designed to go with their DDX3216 digi mixer. For line
level inputs only, they should add a mic-pre bypass switch.
If they want to sell these things to accompany the new BCA2000 for multi
channel recording, then I recommend they do a new version with that feature.

TonyP.

"Andy Cullen" wrote in message
om
Dear Tony,

the mic preamps in the ADA8000 do not use op-amps - they are based on
a low noise conjugate-pair transistor pre-amp circuit similar to that
used in our mixing consoles.
We do utilise TL074 op-amps in the line-level circuitry, and the LM339
is used in its intended role as a comparator.

Thanks for the clarification. Based on the latest tech test results,
whatever you're doing, it is reasonably effective.

TonyP wrote:

"hank alrich" wrote in message
...

TonyP wrote:


People here just don't seem to understand the correlation between gain,
and S/N ratio.


Give a listen to the Gordon preamp at any gain setting...


I'm sorry I'm not sure what your point is? Have they discovered a way to
defy the laws of physics in some way I'm not aware of?
Or does it use cryogenic cooling perhaps? Or maybe it just has very low
gain?


It uses no feedback--changing gain dynamically reconfigures the circuit. The net result is an EIN that remains nearly constant at gain settings from 35 dB to 70 dB.

Gain still affects the S/N, but the realworld performance is excellent under a wide range of conditions.

http://www.gordonaudio.com/

"Kurt Albershardt" wrote in message


It uses no feedback--changing gain dynamically reconfigures the
circuit. The net result is an EIN that remains nearly constant at
gain settings from 35 dB to 70 dB.

Gain still affects the S/N, but the realworld performance is
excellent under a wide range of conditions.

http://www.gordonaudio.com/

I dunno. As a rule, any claim of "no feedback" in a real-world audio amp is
a false claim.

If they said that they don't vary gain by varying feedback, that could be
true.

If they said that there was no loop feedback, that could be true as well.

However, it's exceedingly hard to build a real world amplifier without some
kind of local or loop feedback, usually both.

I've also got problems with the claim that "The primary source of distortion
in any preamplifier is gain.". Conventional wisdom is that the primary
source of distortion is signal amplitude. Obviously the two are related,
but that involves another independent variable - input signal level.

Hey, the amps might be great, but the advertising seems a tad weak.

TonyP wrote:

Give a listen to the Gordon preamp at any gain setting...

I'm sorry I'm not sure what your point is? Have they discovered a way to
defy the laws of physics in some way I'm not aware of?
Or does it use cryogenic cooling perhaps?

No, that'd lose all the "warmth". g

Or maybe it just has very low
gain?

Different topology, no feedback. Gives quite a different result.

You'd probably enjoy a look at:

http://www.gordonaudio.com/

--
ha

Arny Krueger wrote:

"Kurt Albershardt" wrote

It uses no feedback--changing gain dynamically reconfigures the
circuit. The net result is an EIN that remains nearly constant at
gain settings from 35 dB to 70 dB.

Gain still affects the S/N, but the realworld performance is
excellent under a wide range of conditions.

http://www.gordonaudio.com/

I dunno. As a rule, any claim of "no feedback" in a real-world audio amp is
a false claim.

If they said that they don't vary gain by varying feedback, that could be
true.

If they said that there was no loop feedback, that could be true as well.

However, it's exceedingly hard to build a real world amplifier without some
kind of local or loop feedback, usually both.

I've also got problems with the claim that "The primary source of distortion
in any preamplifier is gain.". Conventional wisdom is that the primary
source of distortion is signal amplitude. Obviously the two are related,
but that involves another independent variable - input signal level.

Hey, the amps might be great, but the advertising seems a tad weak.

Arny, this Grant Carpenter guy is no lightweight; he has some history of
interesting design work. That preamp sounds unlike any other pre I have
ever auditioned, compared in the context of GR, Millennia and Grace. I
think in the context of this thing I'd avoid any fall-backs to "as a
rule". It costs a lot to build a pre the way Grant does, but I'm saying
people ought to hear this thing. Call him up; he's no recluse. He'll
talk about it. He's way into _something else_ judging only by what I
heard.

--
ha

Arny Krueger wrote:

http://www.gordonaudio.com/

I dunno. As a rule, any claim of "no feedback" in a real-world audio amp is
a false claim.

I'll defer to Grant on this one since I know just a little about his circuit topology.


If they said that they don't vary gain by varying feedback, that could be
true.

If they said that there was no loop feedback, that could be true as well.

AFAIK both of these are true. I know that the gain is controlled in each of the two stages by some rather elaborate re-biasing among other things.




I've also got problems with the claim that "The primary source of distortion
in any preamplifier is gain.". Conventional wisdom is that the primary
source of distortion is signal amplitude. Obviously the two are related,
but that involves another independent variable - input signal level.

Yes, Grant mentioned that getting decent performance out of the lower gain settings (and higher input levels) required phenomenal voltage swing capability.

"Preben Friis" wrote in message


Take a look at: http://www.gordonaudio.com/images/3preinta.jpg

Judging from the component count alone, they might have built an
amplifier for each gain step and then switches between them instead
of varying the gain of a single amplifier.

The block diagram at http://www.gordonaudio.com/images/3block.gif might
confirm that. By using two cascaded amps each with a limited number of gain
steps (3 or 4), they can get a goodly number of different gains (12 if
http://www.gordonaudio.com/images/3gc2chb.jpg can be taken at face value)
without a humungeous amount of circuitry. dBs add.

"Preben Friis"
Quote from that post:

Be sure to measure it with your lab
equipment too, then post in terms we can all understand, like "Behringer
Bites"
and "Behringer sucks!" and

I was being facetious Preben. You rushed in to defend Behringer without
consideration of the BS disrespect of this forum implicit in the many postings
of the ilk of "Behringer Rules!". My point was if you want have a serious
discussion, that is fine, but the sales hype postings have got to go, and you
were implicitly defending that type of BS.

. Actually it has better specs than the
best analog tape recorders, haven't it? So no, it does not suck, it does not
bite, but is sure stings that with cheap modern technology you can make
stuff that kicks ass to anything that was made ten years ago.

So F'n what! Maybe it's something I can use, maybe it's not, but the RAP
faq states this group shall not be used as a mass marketing sales forum. That
should be plain enough on the face of it.

You mean the same Ty Ford, that responded with "To yoy sir, a high-tech
MEOW! And thanks for your vigilance." to a post that compared the ADA8000 to
digested tuna without hearing it? That would surely be an objective test ...
or not.

Yes, that Ty Ford who is a consultant for equipment manufacturers and who
has written and reviewed a plethora of pro audio gear. Perhpas you are unaware
that Ty is as well respected and as qualified and objective in this regard as
they come.

His point like mine, since you seem to have had a cognitive dissonance on
this point, is he also does not appreciated this forum being the object of a
spam campaign, as we have seen in the past in exactly in the same manner as has
just been occuring. We're supposed to beleive a bunch of guys who are not
members of the group appear one day to tell us all how great B's gear is, and
it's not part of a marketing campaign of some sort? Wanna buy a bridge?

Will Miho
NY Music & TV Audio Guy
Off the Morning Show! & sleepin' In... / Fox News
"The large print giveth and the small print taketh away..." Tom Waits

Arny Krueger wrote:

http://www.gordonaudio.com/images/3block.gif
By using two cascaded amps each with a limited number of gain
steps (3 or 4), they can get a goodly number of different gains (12 if
http://www.gordonaudio.com/images/3gc2chb.jpg can be taken at face value)
without a humungeous amount of circuitry.

Gain is adjustable from 10 dB to 70 dB in the Model 3 (13 steps of 5 dB each.)

"Kurt Albershardt" wrote in message

Arny Krueger wrote:

http://www.gordonaudio.com/images/3block.gif
By using two cascaded amps each with a limited number of gain
steps (3 or 4), they can get a goodly number of different gains (12
if http://www.gordonaudio.com/images/3gc2chb.jpg can be taken at
face value) without a humongous amount of circuitry.

Gain is adjustable from 10 dB to 70 dB in the Model 3 (13 steps of 5
dB each.)

The circuit board picture seems to show two groups of amplifiers, with 4
elements each. That could give up to 16 steps, no?

http://www.gordonaudio.com/images/3preinta.jpg

Warning, my ability to analyze pictures of circuit cards has already been
questioned already once this week!

;-)