How does one go about routing a d&b system?

[Tom Hares]

So, here’s a query for the group.

From next April we are hoping to instigate a rolling replacement for
our speaker system to retire our aging EAW rig.
Our current favourite replacements are d&b E12 for a number of reasons
(recognisable brand, sounds very good, long serving life, available to
several hire companies locally, lightweight, runs off two core cable).
This would include getting D12 amps for these.

And this is where the questions start:

How do users of d&b speakers utilise processing?
We currently have a 32in/32out MediaMatrix system (which we’d like to
include in the upgrading process at some point). What I don’t
particularly want is to have the AD/DA conversion (and latency) of the
MediaMatrix with another AD/DA taking place at the amps.

Do other users have a XTA/bss/MediaMatrix/DME frontend which then goes
AES-EBU to the amps? Or an analog matrix mixer? Or a 32 channel
soundcard set-up and a software based matrix mixer? Or just lots of
tie lines and link cables between amps?
We generally have to change how many audio lines get sent up on a
weekly basis - one week it'll be a stereo pair that needs to control
the entire system - the next it'll be individual lines to main
speakers, subs, delays ... so any matrix needs to be adaptable

Note that I’m not worried about the control of the audio – it’s the
physical routing that's the issue here.
All the tie lines to the amp room are analog copper – there is a
potential for Ethernet based digital audio but that means involving
the IT department and I’d honestly prefer not to. So we are looking at
XLR based signals (and being able to use euroblock connections will
save on re-wiring ...)

A couple of extra considerations – some of the rig (particularly our
100v delay lines) will not be d&b based so would definitely need to go
through some processing that isn’t R1.
And the rolling replacement program means that there would be some
swopping of analog and digital outputs over the next few years – I
suspect we’d make do with link cables for the next couple of years but
there’s currently no backup for the MediaMatrix – if it fails, we lose
all processing.

I can supply more details and reasons should people require.

Thanks in advance.

Tom

[Charlie Richmond]

On Thu, 10 Dec 2009, Tom Hares wrote:

> AES-EBU to the amps? Or an analog matrix mixer? Or a 32 channel
> soundcard set-up and a software based matrix mixer?

Since you mention it, Tom, you can use SoundMan-Server with any combination of
audio interfaces that works for you (Ethersound/Cobranet/AES/MADI/whatever) in a
standard computer to achieve a software based matrix mixer with programmable
delay as well:

http://www.richmondsounddesign.com/virtual-sound-system.html

Yes, the focus of those pages is control but it can also just be a really good
basic routing system, controllable from multiple locations using telnet if you
want to go that way.

Talk to me offlist if you want.

Good luck!
Charlie

| - Charlie Richmond - Richmond Sound Design - Skype: charlierichmond - |
| - http://www.RichmondSoundDesign.com "Performance for the Long Run" - |
| -------- SoundMan-Server -- the ultimate Virtual Sound System ------- |
| -------- LinkedIn: http://www.linkedin.com/in/charlierichmond ------- |
| -------- Facebook: charlie.richmond Twitter: charlierichmond -------- |

[Charles Coes]

In my limited experience with big d&b systems several approaches were taken.  For the first show we used the matrix in the console (DM2k) to build zones that were then sent directly to the D12s where eq and delay were taken care of.  Signal was sent via AES/EBU from the console, and looped for the amps in the line arrays.  For the other setup, with a simpler console and designers who were less interested in control over the room in tech we had bss units between the console and the amps taking care of the matrixing.  Signal in this case was analog.  There was no stage monitoring in this case, so the small latency added by the extra AD/DA was just in place of what would have been added in the amps anyway.  I didn't notice any big sonic difference between the two scenarios.

I hope that helps.

Charles Coes

cco...@gmail.com

"When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." - Buckminster Fuller

--
PLEASE edit unnecessary text out of your reply.

To unsubscribe or set your delivery options, see the web page:
http://groups.google.com/group/theatre-sound-list

Please take non-theatre-sound topics to private email.

[Tom Hares]

Hi,

thanks for the responses on and off-list

Our stage monitors travel a different path and are routed separately
to the rest of the system so that's not an issue - but if a company
brings in a digital console, which goes analog to the processing, goes
back to digital to be processed, back to analog to go to the amps to
converted back and forth again ...

Yes digital lines to the amps are in order - but our cable runs from
desk positions (and yes I do mean plural) really don't lend themselves
for easy runnings.

I'll look into Soundman-Server in more detail. I guess that something
MAX based would work as well - though I have no experience in
programming so that may be a tall order.

It does seem as though d&b are missing a trick here - a black box a
few U deep with half a dozen slots for input/output choices running
R1/R10 and able to accept a thin client (or a wireless connection)
surely couldn't be that tricky to make up ... or have I missed a bunch
of patents somewhere ...

I'm still open for other peoples comments.

And if anyone has tried sending AES-EBU signal down a couple of
hundred metres of ten year old mic cable through a couple of patch
bays ...

Cheers

Tom

2009/12/10 Charles Coes <ccoes...@gmail.com>:

[Charlie Richmond]

On Mon, 14 Dec 2009, Tom Hares wrote:

> It does seem as though d&b are missing a trick here - a black box a
> few U deep with half a dozen slots for input/output choices running
> R1/R10 and able to accept a thin client (or a wireless connection)
> surely couldn't be that tricky to make up ... or have I missed a bunch
> of patents somewhere ...

No, you haven't missed anything. I have been talking with d&b for several years
now trying to convince them they need to come to the table with digital
distribution but they remain adamant for reasons that they lay on their
engineers in Germany who apparently are not willing to talk with me about it.

> And if anyone has tried sending AES-EBU signal down a couple of
> hundred metres of ten year old mic cable through a couple of patch
> bays ...

Good luck there ;-)

I'd love to talk with you offlist more about this, of course.

[Brad Weber]

Tom asked:

> And this is where the questions start:
>
> How do users of d&b speakers utilise processing?

I've also run into similar concerns with other brands such as Nexo that have
dedicated processors or integrated processor/amps, however most of these
were in completely new systems and thus did not have some of the
considerations that you have. I like to let the speaker processor handle
the speaker/array processing and then use the matrix DSP for overall routing
and general house and subjective system tuning. It basically creates
another layer that you can let some people access while keeping them out of
the speaker processing.

> We currently have a 32in/32out MediaMatrix system (which we'd like to
> include in the upgrading process at some point). What I don't
> particularly want is to have the AD/DA conversion (and latency) of the
> MediaMatrix with another AD/DA taking place at the amps.
>
> Do other users have a XTA/bss/MediaMatrix/DME frontend which then goes
> AES-EBU to the amps? Or an analog matrix mixer? Or a 32 channel
> soundcard set-up and a software based matrix mixer? Or just lots of
> tie lines and link cables between amps?

I have always tried to have one A/D conversion and then stay in the digital
domain as much as possible after that. That's typically digital consoles
with AES or Cobranet/EtherSound/DANTE/A-Net/etc. out of the console and
between devices. Not that all of these are without some latency of their
own, but very low in many cases. Of course if your existing copper is not
AES compatible then it may not be usable for that purpose.

Don't necessarily eliminate network audio solutions as there is a difference
between that and Ethernet networks. You can have a networked audio system
using Cobranet, EtherSound, DANTE, etc. and not be tied into any data
networks, in fact I prefer to keep the two as separate as possible. If your
IT department thinks they need to be involved in anything using UTP (CAT5,
5e, 6) cabling they may be very surprised to find many non-Ethernet based
networks using it and that many network audio schemes will not work on the
data networks.


It sounds like the integration of the MediaMatrix system may be the most
difficult aspect to address. One aspect of this is that given the scope of
the other system changes then regardless of what amps and speaker you use it
sounds as though the programming of the MediaMatrix will probably have to
also change significantly. That is good in that the programming can be
changed to match whatever you do but it also has the downside of
representing a significant investment in the system that it sounds like you
may soon replace. One significant consideration is that replacing the
MediaMatrix with a new device is probably going to be more cost effective if
done as part of the speaker upgrade as that potentially avoids investing in
reprogramming the MediaMatrix and then sunsequent programming for whatever
you use to replace it.

Another aspect is if, or when, the inputs to the DSP may change from analog
to digital. If you ever move to a digital console or other digital sources
then it makes sense to stay digital through to the speaker processors. If
that is part of the upgrades planned then that may also add to the cost
effectiveness of changing out the MediaMatrix now. But if a change to a
digital console may occur well after replacing the MediaMatrix then it might
make sense to look at DSP devices that could make the same transition. Some
of the matrix DSP systems such as the BSS London BLU and Peavey Nion
offerings have modular I/O so you could literally configure them as needed
and then later swap the inputs from analog to digital if desired.



Brad Weber
muse Audio Video
Marietta, GA

[Charlie Richmond]

On Mon, 14 Dec 2009, Brad Weber wrote:

> of the matrix DSP systems such as the BSS London BLU and Peavey Nion
> offerings have modular I/O so you could literally configure them as needed
> and then later swap the inputs from analog to digital if desired.

All very good advice. And my suggestion also is completely modular since it
uses any combination of standard computer audio interfaces.

Charlie

[Richard B. Ingraham]

> -----Original Message-----
> From: theatre-s...@googlegroups.com [mailto:theatre-sound-
> li...@googlegroups.com] On Behalf Of Tom Hares
> Sent: Monday, December 14, 2009 2:32 PM
> To: theatre-s...@googlegroups.com
> Subject: Re: [theatre-sound-list] How does one go about routing a d&b
> system?
>
> It does seem as though d&b are missing a trick here - a black box a
> few U deep with half a dozen slots for input/output choices running
> R1/R10 and able to accept a thin client (or a wireless connection)
> surely couldn't be that tricky to make up ... or have I missed a bunch
> of patents somewhere ...

I don't think they are missing much here. I mean how many other products
are already out there that they would be competing against. Most of their
speakers have all the processing you're likely to need right in the DSP
contained in the amps themselves. So all you need is something to
distribute your signals. Often that's the digital console itself. Or any
number of DSP boxes available on the market.

I don't think it's really all that big of a deal to use analog outputs from
the desk to the amps. Even with really nice d&b speakers unless the analog
lines connecting them are garbage there isn't going to be any noticeably
loss in signal quality, at least in my experience anyway. The tiny extra
amount of latency that is added for one more A-D-A conversion is going to be
insignificant in most applications. Unless you just have a ridiculous
number of DSP boxes or the like in your signal chain (from mic or source to
speaker) the amounts of latency we are talking about are equivalent to
moving the speakers a foot or two upstage. (or a foot further away from
where they are aiming)

Sure, if you can put together a package where there is less A-D-A
conversions that is going to be ideal of course. And given a reasonable
option I would go that route. But at the end of the day I'm not going to
throw in unnecessary boxes and format convertors or the like, just so I can
hook up my amps via a digital connection, unless there really is a need for
that.

Just my opinion of course. And not knowing your exact situation, maybe
there is a need in your venue.


Richard B. Ingraham
RBI Computers and Audio
http://www.rbicompaudio.20m.com

[Tom Hares]

Hi

answering a few posts here so please forgive me if I jump a little.

The idea of having the D12 doing speaker dedicated processing that
incoming engineers get no access to and a separate processing unit
that is available for show use was one advantage of keeping a
processing matrix (MediaMatrix or otherwise).

And it's not just the in-house staff - we'll have over forty shows in
over a year, with some of those shows being festivals so there's a lot
of engineers each year. We've already got engineers running in RJ-45
and fibre to onstage amps - I'm sure it's only a matter of time before
the question gets asked if we can take a digital feed up to the amps -
particularly if there are E12's hanging up as our main speakers.

Personally, I wouldn't mind losing the current MediaMatrix set-up -
whether for Nion, BLU, DME, Soundman or something else. It's pretty
good and has been very useful but has no scope for upgrade and the
latency is on the order of several milliseconds whatever you are doing
- mostly fine for our predominantly dance based shows but has caught
us out once or twice.

My note about the IT department may be misleading - there are a couple
of RJ45 ports nearby that can be linked to our sound positions. The
patch bay for this is locked in a room that only our IT department
will normally access and the patch bay itself is a very scary wall to
look at - lots and lots (and lots) of unlabelled data patches - eek!
I presume it's standard CAT5 cable (the adapters that I can see appear
to be standard 4 pair UTP cable) - I haven't looked at audio networks
in too much detail over the last few years.
And I doubt that our existing copper would take AES (though I may be
trying out that experiment at some point soon ...)

I'm not expecting d&b to produce another processing box - I was
thinking more of an audio distribution box with lots of card slots to
accept a variety of inputs and enough processing to be able to route
them to a variety of outputs. There would be one AD conversion if
going from analog input to digital output and one DA conversion if
going from digital input to analog output. The rest would just be
control of the audio paths. This audio distro box would then be part
of the control network of the amps so it could tie into R1 if wanted,
or to any other processing control set-up. Otherwise all the user can
do is essentially use a big matrix mixer that has no other processing.
Maybe something for a company other than d&b now that I think about it
...

Richard - I fully understand what you are saying with regards to
adding unneccesary boxes. At it's simplest this revamp will be just a
straight swop from old speakers and amps to new. And it may be that
I'm worrying too much about what the additional digital change will be
in real terms. We've not really worried about it when using those
speakers in the past but then we've not run them through the
MediaMatrix when we've used them.

Right, back to the current show

Thanks for all the comments - they are useful and help to make sure
that we don't miss things

Cheers

Tom

[Jason Romney]

On Dec 15, 2009, at 4:06 AM, Richard B. Ingraham wrote:

> I don't think it's really all that big of a deal to use analog outputs from
> the desk to the amps.

I agree with Richard. I wouldn't lose any sleep over an extra A/D and D/A conversion in the amp. Speaking from experience with the d&b amps, the converters are clean. The worst thing that will happen here is that you'll gain 1 millisecond of latency. You can pretend the DSP in the amp isn't there if you want. Just use your MediaMatrix for all the system DSP. The amps will just make sure the loudspeakers are performing to spec.

So I would just replace the amps and loudspeakers with the new d&b equipment and integrate it into the existing setup. Down the road, you can worry about digital I/O.

Jason


--
Jason Romney
Sound Design Faculty
University of North Carolina School of the Arts
http://www.uncsa.edu

[Charlie Richmond]

On Tue, 15 Dec 2009, Jason Romney wrote:

> I agree with Richard. I wouldn't lose any sleep over an extra A/D and D/A
> conversion in the amp. Speaking from experience with the d&b amps, the
> converters are clean. The worst thing that will happen here is that you'll
> gain 1 millisecond of latency. You can pretend the DSP in the amp isn't there
> if you want. Just use your MediaMatrix for all the system DSP. The amps will
> just make sure the loudspeakers are performing to spec.

DSP that sits in amplifiers and processing and routing that sits in dedicated
hardware devices such as MMs are going to become obsolete eventually however ;-)

> So I would just replace the amps and loudspeakers with the new d&b equipment
> and integrate it into the existing setup. Down the road, you can worry about
> digital I/O.

Software based systems are going the be the way of the future....

C-)

[Richard B. Ingraham]

> -----Original Message-----
> From: theatre-s...@googlegroups.com [mailto:theatre-sound-
> li...@googlegroups.com] On Behalf Of Charlie Richmond
> Sent: Tuesday, December 15, 2009 8:13 PM
> To: theatre-s...@googlegroups.com
> Subject: Re: [theatre-sound-list] How does one go about routing a d&b
> system?

> DSP that sits in amplifiers and processing and routing that sits in
> dedicated
> hardware devices such as MMs are going to become obsolete eventually
> however ;-)
>

Yes and no. In the case of the d&b amps the DSP internal to those amps is
not just for EQ and Delay, but they are also dedicated processors to any of
d&b speaker models. So rather than having a generic amp(s) and adding an
external processor of some type (which probably would be one more DSP box of
some kind) that does functions such as EQ curves and the like for a
particular speaker model, d&b just adds all that into their dedicated amps.
I actually think it's a pretty nice way of working personally. To be clear,
I'm not totally certain what processing is happening for a given speaker
model with the d&b amps, but I do know you have to set the amps to the
correct model of speaker they are powering, or you can set them to a generic
preset for powering non d&b speakers as well, which I can only assume is
just a flat EQ curve with no loudness processing or the like. I'm sure if
you really want to know what those settings do d&b would be happy to tell
you. :-)

I like powered speakers as well, where the amp and DSP is all inside the
speaker. But you can achieve a bit more economy with d&b's approach since
you can hook up more than one speaker to each amp channel, so you're not
paying for an amp for each speaker, if you don't really need to do so.
(note: there are some d&b speaker models I think that don't allow for that,
so don't quote me...check with your d&b physician before taking..)

So I guess I disagree to a certain extent. I think we'll see more and more
powered speakers, where the DSP is in the speaker, or speakers that have
dedicated processing, be that in an amp or in addition to the amplifier.

I certainly don't feel like dealing with things like x-over settings and the
like inside my mixing and/or playback computers. I don't mind doing things
like EQ and routing/matrixing and delays inside the computer. But I don't
want to deal with the EQ curves and other processing that is to designed to
make a particular speaker make and model relatively flat.

> > So I would just replace the amps and loudspeakers with the new d&b
> equipment
> > and integrate it into the existing setup. Down the road, you can
> worry about
> > digital I/O.
>
> Software based systems are going the be the way of the future....

I of course agree with you to the extent that you are talking about the EQ,
Routing and Delays needed for a particular show, or even for a
fixed/installed sound system. I'm perfectly happy to get rid of most
places where I might have used a Media Matrix, or Drive Rack or an Audiobox
(or insert your favorite here) in the past to do basic corrective system EQ
and delays and do all that in software instead. Look no further than my web
site for an example of a show that uses only computers to do all the mixing,
playback and system EQ, delay and routing: :-)


http://www.rbicompaudio.20m.com/pan09.html

The only external processing in that show is processors dedicated to the
Community Speakers, which are not shown in the line diagram.


However there are also times and places where you want your system EQ and
Delay settings in a system that is separate from the computers and/or
hardware you are using to mix the show as well. So of course you could
dedicate a separate computer just for that task that is locked down, so
guests in your venue couldn't tinker. But then it's just a computer
replacing a DSP box and I don't really see much advantage to that method
other than maybe you'll get a more flexible system (i.e. more routing and
features, etc.) for the same amount of money.

[Charlie Richmond]

On Tue, 15 Dec 2009, Richard B. Ingraham wrote:

> So I guess I disagree to a certain extent. I think we'll see more and more
> powered speakers, where the DSP is in the speaker, or speakers that have
> dedicated processing, be that in an amp or in addition to the amplifier.
>
> I certainly don't feel like dealing with things like x-over settings and the
> like inside my mixing and/or playback computers. I don't mind doing things
> like EQ and routing/matrixing and delays inside the computer. But I don't
> want to deal with the EQ curves and other processing that is to designed to
> make a particular speaker make and model relatively flat.

I don't disagree with this (except that it's much harder to update curves and
settings inside amplifiers and powered loudspeakers than it is inside both
computers and standalone central processors) but d&b have taken a decidedly old
school approach to giving their systems digital inputs so that all sorts of
digital 'processing' systems (including routing/matrix devices) can easily feed
them without additional conversions.

>> Software based systems are going the be the way of the future....
>
> I of course agree with you to the extent that you are talking about the EQ,
> Routing and Delays needed for a particular show, or even for a
> fixed/installed sound system. I'm perfectly happy to get rid of most
> places where I might have used a Media Matrix, or Drive Rack or an Audiobox
> (or insert your favorite here) in the past to do basic corrective system EQ
> and delays and do all that in software instead. Look no further than my web
> site for an example of a show that uses only computers to do all the mixing,
> playback and system EQ, delay and routing: :-)

Well, you know that we usually agree almost completely except in the details and
sometimes the semantics ;-) And usually I talk about the distant future whereas
you are always extremely exacting and thorough as you describe correctly how
things are for now and the immediate future ;-)

> However there are also times and places where you want your system EQ and
> Delay settings in a system that is separate from the computers and/or
> hardware you are using to mix the show as well. So of course you could
> dedicate a separate computer just for that task that is locked down, so
> guests in your venue couldn't tinker. But then it's just a computer
> replacing a DSP box and I don't really see much advantage to that method
> other than maybe you'll get a more flexible system (i.e. more routing and
> features, etc.) for the same amount of money.

Absolutely - or for even less money ;-) Thanks for making the point even more
pointedly and precisely as always ;-)

Charlie

[Richard B. Ingraham]

> -----Original Message-----
> From: theatre-s...@googlegroups.com [mailto:theatre-sound-
> li...@googlegroups.com] On Behalf Of Charlie Richmond
> Sent: Tuesday, December 15, 2009 10:40 PM
> To: theatre-s...@googlegroups.com
> Subject: RE: [theatre-sound-list] How does one go about routing a d&b
> system?
>

> I don't disagree with this (except that it's much harder to update
> curves and
> settings inside amplifiers and powered loudspeakers than it is inside
> both
> computers and standalone central processors) but d&b have taken a
> decidedly old
> school approach to giving their systems digital inputs so that all
> sorts of
> digital 'processing' systems (including routing/matrix devices) can
> easily feed
> them without additional conversions.
>

To be specific I wasn't talking about EQ curves that you would need to
update or change from show to show, or venue to venue if you're a tour. I'm
talking about the EQ settings for the speakers (and maybe a x-over or delays
that needed to align drivers within the speakers, etc..) themselves. I
wouldn't want to have to deal with those settings inside a 3rd party device
of some type. (whether than 3rd party device be a standard DSP box of some
type or a computer, or anything that might have that functionality)

Now granted, if I'm a sound shop and I supply d&b speakers, well then I
might need to change the settings in the amps on regular basis as this week
Amp #22 might power a E0 and next week it might be powering a Q7. But the
d&b amps do have a networking port so you can hook them all up to a computer
and change their settings, etc...

I do agree that it would be nice if they have more options available for the
digital I/O port on their amps. But my feeling is that if you can afford
the d&b speakers and amps themselves, well then a few more conversion boxes
to convert from another digital format to AES/EBU is probably also within
your budget. :-)

>
> Well, you know that we usually agree almost completely except in the
> details and
> sometimes the semantics ;-)

Yes, I should have said that I disagree to a small extent. :-)

[Charlie Richmond]

On Tue, 15 Dec 2009, Richard B. Ingraham wrote:

> Yes, I should have said that I disagree to a small extent. :-)

Please, don't let me browbeat you ;-)

C-)

[Nick Kourtides]

On Dec 15, 2009, at 10:15 PM, Richard B. Ingraham wrote:
> ...But you can achieve a bit more economy with d&b's approach...


Richard, this is the funniest sentence I've heard in several days.
I've literally just picked myself up off the floor. And now going to
bed.

laughing,
Nick

[Richard B. Ingraham]

> -----Original Message-----
> From: theatre-s...@googlegroups.com [mailto:theatre-sound-
> li...@googlegroups.com] On Behalf Of Nick Kourtides
> Sent: Wednesday, December 16, 2009 12:17 AM
> To: theatre-s...@googlegroups.com
> Subject: Re: [theatre-sound-list] How does one go about routing a d&b
> system?
>

Yea, I can see that. LOL. :-)

But when the S**t costs that much.... well if you can save a D6 or D12 or
two.....

I was just comparing it to other high end speakers that are self powered.
So you have no choice but 1 amp, 1 speaker. But you're right. It likely
all falls under the category of "if you can afford these speakers, what's an
extra amp or two..." And in the install I have with d&b, we have 1 amp
channel per speaker. :-)

[Jim vanBergen]

Has anyone here shared/explained their perspective on using AES with D&B, or did I miss that? I'm the production engineer on a show designed by Scott Lehrer right now, and we're using AES from Yamaha consoles into & out of DMEs into D6 & D12 amps. It's working quite well. Rental equipment provided by PRG Audio, including AES snakes. While I personally have touted fiber and MADI, Scott's design and use of digital I/O in this system is working wonderfully, and I think it's a mistake to NOT consider AES for driving d&b systems when the infrastructure works so well. 

--
Jim van Bergen
AudioArt Sound, NYC
917-826-1626
vanber...@gmail.com

[Brad Weber]

Charlie said:
> Software based systems are going the be the way of the future....

Keeping in mind that this is a generic view, there are software based audio
mixing systems other then SoundMan that have been getting a lot of
discussion lately in other forums and that seem to be gaining in popularity
among some users, I have to say that maybe in the future it will move more
toward software, but for now I am still a bit wary. There is a current
thread going on another forum regarding an 'older' DSP device for which the
programming software can only be run on Windows 95 or 98. I have run into
similar situations with both DSP units and digital consoles. Anything
software based can be dependent upon the operating system and thus can
become obsolete unless the developer keeps the software both compatible with
new Operating Systems and supports backward compatibility with older
Operating Systems, ideally doing so via free or minimal cost updates so that
you aren't stuck with the decision of a potentially large investment or
abandoning what you have.

I have also recently had two desktop computers, one just over a year old,
experience main hard drive crashes that rendered the computers useless until
the hard drives could be replaced and the systems rebuilt. While RAID
arrays, mirrored drives, etc. can minimize the potential downtime in
critical situations, I am currently a little wary of devices that depend
upon hard drives for operation. If nothing else, if a device crashes I like
it to keep passing audio in the most recent configuration and not just cease
to function.

Richard said:
> Most users are used to multiple banks these days and one function
> I really like is a user assignable layer for the faders. (something
> that I
> hate is missing on the Yamaha M7) So that you can assign any input or
> output or DCA channel to the surface, exposing all the "handles" that
> you
> need to mix a show on a single layer. That way you don't really need
> dedicated faders to groups or the like. (also, 8 groups would likely
> not be
> enough in my book, it depends on what those are actually controlling)

While I like the concept of user defined layers in general, I'm less
thrilled with the actual implementation on consoles that do not have
electronic scribble strips. Multiple layers, especially with one layer
being user defined and potentially varying and others being software
assigned to physical I/O, can be a awkward when having to deal with the
small labeling areas allowed on many entry level digital consoles.

A similar capability that I like is being able to switch layers for groups
of faders rather than just for all of them. Being able to do something like
switch a bank of eight faders between some inputs and DCAs or Groups without
affecting any of the other faders can be very useful. And being able to
easily put a graphic EQ on faders is bonus points.

Richard said:
> To be specific I wasn't talking about EQ curves that you would need to
> update or change from show to show, or venue to venue if you're a tour.
> I'm

> talking about the EQ settings for the speakers (and maybe a x-over or
> delays

> that needed to align drivers within the speakers, etc..) themselves. I
> wouldn't want to have to deal with those settings inside a 3rd party
> device
> of some type. (whether than 3rd party device be a standard DSP box of
> some
> type or a computer, or anything that might have that functionality)

And:

I agree. I have always been a proponent of separating speaker and array
processing from the other processing even if it is all done in the same box,
if nothing else that aspect of the system tuning can usually be performed
under much better conditions in advance of installation and then be locked
out. Dedicated processors preprogrammed for specific speakers takes that to
another level and typically with very good results. My experience using
dedicated speaker processors for particular boxes and arrays, be they
integrated into amps (such as d&b or Nexo) or stand alone (such as those
from EAW, Nexo, dbx and others), is that they greatly simplify the system
setup and improve the performance for most applications. The manufacturers
have time and resources (human, equipment and facilities) not available to
most people that they can apply to optimizing the processing for their
speakers, and in some cases for arrays of their speakers. This is not just
EQ and crossovers but also time alignment and, very important for some
applications, complex limiting and protection. You still need to tune the
system for the environment, specific installation and individual preferences
as well as aligning to backline, delay for fills, etc., but with dedicated
processors the work related to getting the speaker or array itself tuned and
processed is done for you and better then the vast majority of people would
do on their own.

Charlie said:
> I don't disagree with this (except that it's much harder to update
> curves and settings inside amplifiers and powered loudspeakers than it
> is inside both computers and standalone central processors) but d&b
> have taken a decidedly old school approach to giving their systems
> digital inputs so that all sorts of digital 'processing' systems
> (including routing/matrix devices) can easily feed them without
> additional conversions.

Programming DSP in some powered speakers is no big deal as you are starting
to see powered speakers with network connectivity, the Tannoy V-Net products
are a good example. I fairly recently completed a project where the
operator could sit in the audience with a wireless laptop or tablet PC and
not only operate the console but also access and interact with the speaker
processing, house processing, EtherSound interface, wireless mics, outboard
preamps and even the amplifiers. Factor in their wireless comms, whose
programming they could also access if they wanted to, and they can pretty
much do everything a traditional FOH would and possibly even more while
sitting in any seat with a laptop and wireless connection. On the other
hand, they can give visiting artists access to only the aspects they want
them to access while they can still monitor what is happening or override
anything done.

Richard said:
> I don't think it's really all that big of a deal to use analog outputs
> from

> the desk to the amps. Even with really nice d&b speakers unless the
> analog
> lines connecting them are garbage there isn't going to be any
> noticeably
> loss in signal quality, at least in my experience anyway. The tiny
> extra
> amount of latency that is added for one more A-D-A conversion is going
> to be
> insignificant in most applications.

I have found that it does indeed depend on the boxes and the application.
For example, IEMs are much more sensitive to latency while delay fills are
going to be delayed anyways and thus the latency doesn't usually matter for
them. It can become an issue for stage spill from any arrays or clusters
mounted high over stages, on a recent project I had to drop a cluster
several feet from the desired position in order to avoid the spill being
perceived as an echo on stage due to the signal path latency being added to
the natural acoustic time of travel. And where larger I/O counts are
required, some devices may use CobraNet or similar methods to interconnect
boxes, so any latency related to those hops must also be accommodated.

Another consideration may be whether the you are applying delay
compensation. With matrix DSP devices each path created inside the box is
often different, they may have different processing or involve connections
to or from a slave expander, which can result in varying processing and path
latencies for different paths. Some processors offer the option to apply
delay compensation that automatically compensates for those differences,
applying a delay based on the greatest latency path (not related to desired
delay applied, only to processing and internal path delay) and in effect
keeping all paths time aligned through the box. This can be desirable when
trying to keep signals aligned but can increase the latency for some paths.

[Richard B. Ingraham]

> -----Original Message-----
> From: theatre-s...@googlegroups.com [mailto:theatre-sound-
> li...@googlegroups.com] On Behalf Of Brad Weber
> Sent: Thursday, December 17, 2009 8:56 AM
> To: theatre-s...@googlegroups.com
> Subject: RE: [theatre-sound-list] How does one go about routing a d&b
> system?
>

> A similar capability that I like is being able to switch layers for
> groups
> of faders rather than just for all of them. Being able to do something
> like
> switch a bank of eight faders between some inputs and DCAs or Groups
> without
> affecting any of the other faders can be very useful. And being able
> to
> easily put a graphic EQ on faders is bonus points.
>

I'm assuming you're referring to the CentraLogic section on the M7, or
similar features in other desks, but that's the only one I've worked with
that has this type of functionality. I've tried and tried to make that work
for me... on the M7 that is, and I have to say I just hate it. Some of it
I'm sure is that everyone set their M7 to work a tad different, so when I
walk up to one to adjust things, sometimes the centralogic section will
follow if you hit a channel selection button on the channel faders and
others set it up so it doesn't, so it always keeps me guessing as a
designer. :-) It works OK as long as I'm in a place where I can take the
time to set it all up to my liking. But even so I find it more of a
distraction and hassle than helpful.

I like layers better as long as I have enough physical faders to run the
show logically with little to no swapping of what a fader does during a
show. It's why in some regards I actually prefer a LS9 to the M7. (if I
have another box to do matrix mixing post the LS9, I actually would rather
work on a LS9 than the M7, most days in fact) I can usually set up a show
to be run on 32 (33 if you make good use of the Master Stereo Bus fader)
faders with no swapping at all during the show. I will expose the "handles"
the board op needs during the show and hide the stuff they don't need to run
the show in an unexposed layer. I've found this works very well, especially
since I'm often working with students or less experienced board ops. My
main issues with the M7 is that the DCA faders can only be assigned to
inputs, not outputs, and more importantly there are only 8 of them, which I
have found out is rarely enough, even on modest cast size shows. At least
in my experience and the way I like to have musicals mixed anyway. It just
leads to way to many DCA reassigns during a show, in my not so humble
opinion.

Granted that doesn't mean I should dismiss the concept just because I don't
like the way it's implemented on one particular model of console. :-)

The gradual move towards software based systems excites me because then you
can pick and choose the size of control surface you need for a given
situation. (in theory anyway... someday..) If I need only 16 handle for
the board op, then that's all you put in the show. If you need 40, then put
together a surface (or multiple surfaces) with up to 40 faders.

> I have found that it does indeed depend on the boxes and the
> application.
> For example, IEMs are much more sensitive to latency while delay fills
> are
> going to be delayed anyways and thus the latency doesn't usually matter
> for
> them. It can become an issue for stage spill from any arrays or
> clusters
> mounted high over stages, on a recent project I had to drop a cluster
> several feet from the desired position in order to avoid the spill
> being
> perceived as an echo on stage due to the signal path latency being
> added to
> the natural acoustic time of travel. And where larger I/O counts are
> required, some devices may use CobraNet or similar methods to
> interconnect
> boxes, so any latency related to those hops must also be accommodated.
>

Yes of course. I was responding mostly to the original poster's specific
situation. I understand that keeping an eye on the total latency of your
system from input to output can be very important and how important it is
will vary depending on the given circumstances.

> latencies for different paths. Some processors offer the option to
> apply
> delay compensation that automatically compensates for those
> differences,
> applying a delay based on the greatest latency path (not related to
> desired
> delay applied, only to processing and internal path delay) and in
> effect
> keeping all paths time aligned through the box. This can be desirable
> when
> trying to keep signals aligned but can increase the latency for some
> paths.

You mean there are boxes out there that don't do that automatically?
Interesting. Let me know which ones please so I can make sure to avoid
them! :-)

[mackerr]

On Dec 17, 10:45 am, "Richard B. Ingraham" <rbingra...@sbcglobal.net>
wrote:

> >
>
> You mean there are boxes out there that don't do that automatically?
> Interesting.  Let me know which ones please so I can make sure to avoid
> them!  :-)
>
> Richard B. Ingraham
> RBI Computers and Audiohttp://www.rbicompaudio.20m.com

Just about all of them. The only consoles I can think of off the top
of my head that DO include automatic delay compensation are the
Digidesign and Midas consoles. I'm sure someone will jump in with
others, but those are the ones I know about.

It is important with digital consoles to be aware of multiple paths
for a single signal as it will almost always cause comb filtering.

Mac

[Charlie Richmond]

On Thu, 17 Dec 2009, Brad Weber wrote:

> Charlie said:
>> Software based systems are going the be the way of the future....
>
> Keeping in mind that this is a generic view, there are software based audio
> mixing systems other then SoundMan that have been getting a lot of
> discussion lately in other forums and that seem to be gaining in popularity
> among some users, I have to say that maybe in the future it will move more
> toward software, but for now I am still a bit wary. There is a current
> thread going on another forum regarding an 'older' DSP device for which the
> programming software can only be run on Windows 95 or 98. I have run into
> similar situations with both DSP units and digital consoles. Anything
> software based can be dependent upon the operating system and thus can
> become obsolete unless the developer keeps the software both compatible with
> new Operating Systems and supports backward compatibility with older
> Operating Systems, ideally doing so via free or minimal cost updates so that
> you aren't stuck with the decision of a potentially large investment or
> abandoning what you have.

Absolutely! And this is precisely why I made the statement originally!
Software is no good unless it is kept up to date.

> I have also recently had two desktop computers, one just over a year old,
> experience main hard drive crashes that rendered the computers useless until
> the hard drives could be replaced and the systems rebuilt. While RAID
> arrays, mirrored drives, etc. can minimize the potential downtime in
> critical situations, I am currently a little wary of devices that depend
> upon hard drives for operation. If nothing else, if a device crashes I like
> it to keep passing audio in the most recent configuration and not just cease
> to function.

I used to include this sort of caveat but now there are 16GB solid state hard
drives that we actually sell for less than $150 which completely eliminates the
vagueries of mechanical hard drives, which always used to be the cog in the
works. True, flash memory isn't perfect either but it produces far less heat
and has an MTBF far greater than 'old fashioned' spinning drives ;-)

> Charlie said:
>> I don't disagree with this (except that it's much harder to update
>> curves and settings inside amplifiers and powered loudspeakers than it
>> is inside both computers and standalone central processors) but d&b
>> have taken a decidedly old school approach to giving their systems
>> digital inputs so that all sorts of digital 'processing' systems
>> (including routing/matrix devices) can easily feed them without
>> additional conversions.
>
> Programming DSP in some powered speakers is no big deal as you are starting
> to see powered speakers with network connectivity, the Tannoy V-Net products
> are a good example. I fairly recently completed a project where the
> operator could sit in the audience with a wireless laptop or tablet PC and
> not only operate the console but also access and interact with the speaker
> processing, house processing, EtherSound interface, wireless mics, outboard
> preamps and even the amplifiers. Factor in their wireless comms, whose
> programming they could also access if they wanted to, and they can pretty
> much do everything a traditional FOH would and possibly even more while
> sitting in any seat with a laptop and wireless connection. On the other
> hand, they can give visiting artists access to only the aspects they want
> them to access while they can still monitor what is happening or override
> anything done.

Flexibility is always the key and the user is the final arbiter in any case. My
point was more that a 10 year old amplifier with built in DSP is less likely to
continue to have current support and more likely for the user to have to say 'I
need to replace that piece of hardware' than a completely software based system
that has been incrementally updated and has had full support for those same 10
years.

[Richard B. Ingraham]

> -----Original Message-----
> From: theatre-s...@googlegroups.com [mailto:theatre-sound-
> li...@googlegroups.com] On Behalf Of mackerr
> Sent: Thursday, December 17, 2009 1:03 PM
> To: theatre-sound
> Subject: [theatre-sound-list] Re: How does one go about routing a d&b
> system?
>

> On Dec 17, 10:45 am, "Richard B. Ingraham" <rbingra...@sbcglobal.net>
> wrote:
> > >
> >
> > You mean there are boxes out there that don't do that automatically?
> > Interesting.  Let me know which ones please so I can make sure to
> avoid
> > them!  :-)
> >

> Just about all of them. The only consoles I can think of off the top
> of my head that DO include automatic delay compensation are the
> Digidesign and Midas consoles. I'm sure someone will jump in with
> others, but those are the ones I know about.
>
> It is important with digital consoles to be aware of multiple paths
> for a single signal as it will almost always cause comb filtering.
>

Are you talking about when you route a signal out of the console and then
bring it back into the console. Like when you would use some outboard piece
of gear? So the Digi and Midas will compensate for that? Or are you just
talking about adding in effects and/or other processing internal to the
console? If we're just talking about internal processing I thought they all
would compensate for that.

I could well be wrong, but I don't think more or less latency is added to
your signals when you turn EQs or Dynamics on and off in your typical
digital console, be that Yamaha or Mackie or Soundcraft, etc.. If they
don't all do that, it's certainly news to me.

I am fairly certain that the total through put on an Audiobox (just to use
an example I'm fairly familiar with) is the same number no matter how many
bands of EQ you add to any signal when you compare it to other signals that
have no processing. (obviously we need to remove adding delay from the
discussion, since in that case you want to add latency to those inputs or
outputs, etc...) :-)

I guess I could see where you add some delay to an input channel (although
that's not an option on the M7 or LS9) and split that same physical input to
another input where you don't add any delay, and then you mix those two
signals together some place along the way, well you would certainly get comb
filtering then, but that's pilot error in my opinion and not a fault of the
console.

Oh well.. it's certainly something interesting to think about and
investigate further when time actually allows.

[Charlie Richmond]

On Thu, 17 Dec 2009, Richard B. Ingraham wrote:

> I am fairly certain that the total through put on an Audiobox (just to use

That is correct and is also true for SoundMan-Server.

Unless, of course, additional delay is specifically added by the user ;-)

[Chip]

Charlie Richmond wrote:

>
> I used to include this sort of caveat but now there are 16GB solid state hard
> drives that we actually sell for less than $150 which completely eliminates the
> vagueries of mechanical hard drives, which always used to be the cog in the
> works. True, flash memory isn't perfect either but it produces far less heat
> and has an MTBF far greater than 'old fashioned' spinning drives ;-)

>
Fry's Elec sells a 32GB USB thumb sized flash drive for ~$80.

Chip 1

[Charlie Richmond]

On Thu, 17 Dec 2009, Chip wrote:

> Fry's Elec sells a 32GB USB thumb sized flash drive for ~$80.

The ones we use are SATA and much faster than USB drives, which is required for
streaming a lot of tracks of audio.

Charlie

[mackerr]

On Dec 17, 2:59 pm, "Richard B. Ingraham" <rbingra...@sbcglobal.net>
wrote:
> >

> Are you talking about when you route a signal out of the console and then
> bring it back into the console.  Like when you would use some outboard piece
> of gear?  So the Digi and Midas will compensate for that?  Or are you just
> talking about adding in effects and/or other processing internal to the
> console?  If we're just talking about internal processing I thought they all
> would compensate for that.
>
> I could well be wrong, but I don't think more or less latency is added to
> your signals when you turn EQs or Dynamics on and off in your typical
> digital console, be that Yamaha or Mackie or Soundcraft, etc..  If they
> don't all do that, it's certainly news to me.
>
>

No, I was talking about multiple paths through the console. Most of
the digital consoles I am familiar with have a fixed latency
independent of processing for a given route. Different routes however
have different latencies. The path; input>stereo has less latency than
input>group>stereo, or input>group>stereo>matrix. If you have an input
taking multiple paths to the same output, for instance to do parallel
processing like compressed and uncompressed drums (a common studio
technique), they need to follow the same route to have the same
latency. Sending the drums to stereo, and to a group that will be
compressed and that group to stereo will cause comb filtering.

Mac

[Chip]

Ah, whats 12MB/sec vs 1.2GB/sec? Just a letter in the alphabet. Should
have remembered that before showing my ignorance.

Chip 1

[Brad Weber]

Richard asked:

> You mean there are boxes out there that don't do that automatically?
> Interesting. Let me know which ones please so I can make sure to avoid
> them! :-)

Most 'drag and drop' programming DSP boxes I've worked with do not
automatically apply delay compensation, it usually has to be selected. You
may want some outputs time aligned but may not want to add any latency to
outputs such as IEM or PSM feeds, so you can typically turn it on or off. I
also remember that on some of the earlier 'drag and drop' programmable DSP
boxes, like the original MediaMatrix, enabling delay compensation often
added significantly to the processing used and in some cases it meant either
exceeding the available processing or adding another DSP card or two. With
newer devices the compensation required has generally decreased while the
DSP capacity has increased thus making delay compensation less of an issue
to accommodate.

Varying latency through consoles is a separate issue, as Mac has addressed.

[Tom Hares]

Hi Paul,

having had a look through the website, this does seem to be pretty
much what I had in mind. The D12's deal with the speakers and this is
an answer for a black box that gets the audio from whatever mixer is
in this week to the amps with the option to add processing for the
non-d&b parts of the rig.

There's a distinct bias towards the broadcast market and our local
dealer doesn't have an easily accessible price list which is never a
good sign ...

But a 6U box would do everything we needed and have spaces for future
use - presuming that we could use inputs from an analog card and route
them to an AES output card.
I'm presuming also that the analog XLR cards take up two slots rather
than the implied one
That DSP cards can be added is fine and I'm happy enough to learn a
new system and don't mind too much that it's not a 'standard' brand -
as long as the quality is there. There's always going to be a case of
'I've never used that name so I'd rather use my own'

I'm also a little worried that I asked for details on this five days
ago and have yet to hear anything - not that they are the only company
we are waiting to hear from as we approach Christmas.

If anyone is aware of other boxes that could do this, feel free to share.

Brad mentioned that the original MediaMatrix adds a fair amount when
you enable delay compensation - which, to be honest, I had forgotten
about.
We normally do enable it here and the latency of the system is around
7m/s for our normal configuration. If i take this off, the latency of
the outputs varies from 2 m/s to 7 m/s and I'm warned that some
devices have a latency of 1.5m/s between inputs

This is mostly due to having inputs on one breakout box (one set of
physical in/outs) being routed to outputs on a different breakout box
- by the time you add matrix mixing, limiters, EQ, delay, filters,
crossovers.
It's possible to trim this down by using parallels to send multiple
feeds to the inputs but sometimes we need lots of discrete inputs - it
saves time on the set-up if we have a known latency so the
compensation is normally enable.
It helps that all the onstage speakers travel a different signal path
so the internal latency is incorporated into the additional delay we
add for speakers in the house - it's possible to get a greater latency
if you don't take some care on the internal routing

Best wishes all

Tom