Does anyone do a three way s/w active crossover implemented on the PC or a
PC card?

Paul

Paul wrote:
Does anyone do a three way s/w active crossover implemented on the PC or a
PC card?

Could be done. Not sure why you want to do it...? (The hardest part is
that you need a card with three channels of output; six if you want stereo.)

(And of course you also need three or six channels of amplifier
downstream. Tri-amped is theoretically better, but in my experience it's
also much easier to get wrong.)

Joe Kesselman wrote:

Paul wrote:

Does anyone do a three way s/w active crossover implemented on the PC
or a PC card?


Could be done. Not sure why you want to do it...? (The hardest part is
that you need a card with three channels of output; six if you want
stereo.)

Well, the reason is that active crossovers using DSPs are very useful in top end
systems. However, machines like Soundweb cost a small fortune.

I was thinking that something like a 6+1 SoundBlaster could be used with the PC
running the DSP algorithms. Alternatively, a dedicated D/A PC card (which are
not hard to find).

--
Dirk

The Consensus:-
The political party for the new millenium
http://www.theconsensus.org

François Yves Le Gal wrote:

On Wed, 05 Oct 2005 10:16:59 +0100, Dirk Bruere at Neopax
wrote:


Well, the reason is that active crossovers using DSPs are very useful in top end
systems. However, machines like Soundweb cost a small fortune.


You may consider the Behringer DCX2496, available for less than 200 GBP. It
gives excellent results when properly used.

http://www.behringer.com/DCX2496/index.cfm?lang=ENG

IIRC someone else mentioned this, but I had assumed the cost would be
prohibitive. For £200 it's amazing!
I now have a choice of solutions following the clues posted here.
Thanks to all who replied.

BTW, both me and Paul are working on the project in case anyone wondered about
name changes etc.

--
Dirk

The Consensus:-
The political party for the new millenium
http://www.theconsensus.org

Personal opinion: If you're putting "top end systems" and "soundblaster"
in the same post, I think you're just a mite confused.

"Joe Kesselman" wrote in
message

Personal opinion: If you're putting "top end systems" and
"soundblaster" in the same post, I think you're just a
mite confused.

Agreed.

My multichannel computer digital audio interface of choice
would be a Delta 1010LT at the very least.

As long as digital crossovers cost more than a good PC, the
idea of using a PC as a signal processor for routine
applications like crossovers made some sense.

The Berhinger DCX2496 pretty well nailed that!

For applications needing complex equalization, the Behringer
DSP1124 (for which there appears to be a 2496 upgrade in
limited distribution) fits the bill.

There is some company that advertises in the back of AudioXpress that
they have a standlone DSP box that can be used as a crossover. Analogue
inputs and outputs and you can download your own code from a PC. Comes
with some sample code. Looks like a nice box for prototyping custom dsp
applications.
--scott


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

Arny Krueger wrote:

"Joe Kesselman" wrote in
message


Personal opinion: If you're putting "top end systems" and
"soundblaster" in the same post, I think you're just a
mite confused.


Agreed.

Any opinions on the latest Creative card, the X-Fi Elite?

My multichannel computer digital audio interface of choice
would be a Delta 1010LT at the very least.

As long as digital crossovers cost more than a good PC, the
idea of using a PC as a signal processor for routine
applications like crossovers made some sense.

The Berhinger DCX2496 pretty well nailed that!

For applications needing complex equalization, the Behringer
DSP1124 (for which there appears to be a 2496 upgrade in
limited distribution) fits the bill.

Also need the DEQ1024 probably.
Being lazy... does all the Behringer stuff interconnect digitally?

--
Dirk

The Consensus:-
The political party for the new millenium
http://www.theconsensus.org

"François Yves Le Gal" wrote in
message
On Wed, 05 Oct 2005 09:16:44 -0400, Joe Kesselman
wrote:

Personal opinion: If you're putting "top end systems"
and "soundblaster" in the same post, I think you're just
a mite confused.

The new X-Fi cards are really good.

Well, maybe good enough.

The band-splitting seems scary for perfectionists.

They still resample except in the trivial cases.

"François Yves Le Gal" wrote in
message
On Wed, 5 Oct 2005 09:17:58 -0400, "Arny Krueger"
wrote:

For applications needing complex equalization, the
Behringer DSP1124 (for which there appears to be a 2496
upgrade in limited distribution) fits the bill.

The DSP1124P is fitted with 24-bit AD and DA converters
sampling at around 48 KHz. It's quite old by digital
standards and can be considered obsolete.

It will be interesting to see if DSP1124P remains in
production after this product is generally available:

http://www.behringer.com/FBQ2496/index.cfm?lang=ENG

A DEQ2496 is a much better eq. solution IMHO.

I've become very partial to parametric eq, even though I
haven't sold or scrapped any of my graphics.

Scott Dorsey wrote:
Looks like a nice box for prototyping custom dsp applications.

Interesting, especially if they have a developers' kit for it.

(I'm still disappointed that the IBM/Motorola MWave family never got any
traction. Slice-multitasking DSPs with standardized APIs, capable of
doing everything from soft modem to voice-rec to synthesis...
Unfortunately there wasn't enough market yet, and they didn't figure out
how to create one. Sigh.)

I used a $7 Dolby 5.1 sound card to make all of the signals to control a
color laser projector. The only thing is that the outputs I needed are
analog. Do you want analog or digital outs for your separated signals?

http://www.akrobiz.com/laserboy/

James. )



"Joe Kesselman" wrote in message
...
Paul wrote:
Does anyone do a three way s/w active crossover implemented on the PC or
a
PC card?

Could be done. Not sure why you want to do it...? (The hardest part is
that you need a card with three channels of output; six if you want
stereo.)

"Dirk Bruere at Neopax" wrote in message
...

Also need the DEQ1024 probably.

You can do any two-channel equalisation you want on the PC, and have
different equalisation for different tracks.

If you are using a PC with multichannel cound card as a crossover, you have
to be certain that you will never run any program that bypasses the
crossover function, and puts out full-range sound to your HF and mid-range
drivers.

Tim

"Arny Krueger" wrote in message
...

The Berhinger DCX2496 pretty well nailed that!

For applications needing complex equalization, the Behringer
DSP1124 (for which there appears to be a 2496 upgrade in
limited distribution) fits the bill.

Can you comment on when the EQ facilities of the DCX2496 would fall short of
requirements for driving speakers, so that a separate EQ device would be
helpful?

Tim

James Lehman wrote:
I used a $7 Dolby 5.1 sound card to make all of the signals to control a
color laser projector.

What's the highest frequency that system actually has to pass. and how
accurately does it have to do so?

The question is whether it will work well enough to be worth the effort.
Which I think still requires understanding why we're specifically
solving the problem this way. How many people really have equipment and
a room where an active crossover and tri-amp setup will make an audible
difference vs. passive?

James Lehman wrote:

I used a $7 Dolby 5.1 sound card to make all of the signals to control a
color laser projector. The only thing is that the outputs I needed are
analog. Do you want analog or digital outs for your separated signals?

http://www.akrobiz.com/laserboy/

Nice pics.
However, I'm going to look at Behringer for now.

--
Dirk

The Consensus:-
The political party for the new millenium
http://www.theconsensus.org

I think the real value of bi or tri amping is the fact that there are no
speaker power reactive elements other than the speakers themselves. In other
words, there are no coils or caps anywhere near your speakers! It is much
easier to control precise frequency and phase of filters at the op amp
level, where you can have as much gain as you like. Not to mention the fact
that there is no need for coils at all; just resisters and small caps. Also,
there is no need to worry about how well the individual driver efficiencies
match each other.

James. )



"Joe Kesselman" wrote in message
...
James Lehman wrote:
I used a $7 Dolby 5.1 sound card to make all of the signals to control a
color laser projector.

What's the highest frequency that system actually has to pass. and how
accurately does it have to do so?

The question is whether it will work well enough to be worth the effort.
Which I think still requires understanding why we're specifically
solving the problem this way. How many people really have equipment and
a room where an active crossover and tri-amp setup will make an audible
difference vs. passive?

On Wed, 5 Oct 2005 11:23:31 -0400, "Arny Krueger"
wrote:

I've become very partial to parametric eq, even though I
haven't sold or scrapped any of my graphics.

Probably because it's a vastly better way of doing it.

James Lehman wrote:
there is no need to worry about how well the individual driver efficiencies
match each other.

You're able to adjust for those by adjusting the levels being sent to
each driver. On the other hand, as I found out when running a triamped
system, that means a whole bunch more controls which can be set
_incorrectly_; it took us about a year to get our triamped configuration
running as smoothly as the passive system had.

In Theory, triamping has significant advantages in letting you keep
strong signal in one band from messing up your ability to reproduce the
other two bands. In practice, I concluded that the active crossover and
its interaction with how the amps were set was one more complication
than I was really comfortable with. Worth doing perhaps, but I'm happy
to leave dealing with it to someone else.

"Tim Martin" wrote in message

"Arny Krueger" wrote in message
...

The Berhinger DCX2496 pretty well nailed that!

For applications needing complex equalization, the
Behringer DSP1124 (for which there appears to be a 2496
upgrade in limited distribution) fits the bill.

Can you comment on when the EQ facilities of the DCX2496
would fall short of requirements for driving speakers, so
that a separate EQ device would be helpful?

Yes, I've looked at this via the manual.

The DCX 2496 has a modest ( the actual number doesn't seem
to be specified in the user's manual) number of parametric
equalizer sections that you can insert at various places in
the signal path. I get the feeling that there might be like
a dozen of them or so. The number that are availble may
depend on your use of other features.

While I've heard the DCX2496 in actual use and fiddled with
it a bit, that equipment is on the other side of town and I
haven't had a chance to fiddle intensively to determine the
total capabilities of the DCX2496's parametric equalizers.

It seems to me that you would probably be able to use the
DCX2496 to do pretty intensive eq on a small subset of the
available channels, or seriously tweak each of the full
complement of channels.

"dizzy" wrote in message

On Wed, 5 Oct 2005 11:23:31 -0400, "Arny Krueger"
wrote:

I've become very partial to parametric eq, even though I
haven't sold or scrapped any of my graphics.

Probably because it's a vastly better way of doing it.

Agreed. I see graphics as learning devices for parametrics.

The best of all possible worlds are the parametrics with
graphical UIs. I have them in Audition and in my 02R96.
mixing console.

"dizzy" wrote in message
...
On Wed, 5 Oct 2005 11:23:31 -0400, "Arny Krueger"
wrote:

I've become very partial to parametric eq, even though I
haven't sold or scrapped any of my graphics.

Probably because it's a vastly better way of doing it.

Just to bring me up to speed, what is the difference between a graphic and a
parametric equalizer?

Graphic EQs have a set of fixed frequencies that they can either cut or
boost. Parametric EQ's allow you to move the frequency and, in some cases,
the bandwidth around the frequency, while still providing the same idea of
cutting or boosting.



wrote in message
...

"dizzy" wrote in message
...
On Wed, 5 Oct 2005 11:23:31 -0400, "Arny Krueger"
wrote:

I've become very partial to parametric eq, even though I
haven't sold or scrapped any of my graphics.

Probably because it's a vastly better way of doing it.

Just to bring me up to speed, what is the difference between a graphic and
a
parametric equalizer?

normanstrong wrote ...
Just to bring me up to speed, what is the difference between a graphic and
a parametric equalizer?

Seems like a good topic to add to the RAP FAQ.

Very basically, a "graphic" equalizer has a set of boost/cut controls
evenly spaced across the bandwidth of interest. Typically at 1-octave,
or 1/2-octave, or 1/3-octave, etc. intervals. Commonly implemented
with "slider" type controls, so that the front panel gives a "graphic"
view of the frequency response. For example...
http://www.dbxpro.com/231.htm

A "parametric" equalizer is so-named because additional "parameters"
can be controlled. Typically, in addition to boost/cut, you can also
control the center frequency and the "Q" (or "width") of the controlled
band. Physical parametric equalizers typically use a series of rotary
controls for the parameters (as compared with sliding/linear controls
on a "graphic" equalizer) For example...
http://www.speck.com/asc/asc_2.shtml

"James Lehman" wrote in message
news
Graphic EQs have a set of fixed frequencies that they can either cut or
boost. Parametric EQ's allow you to move the frequency and, in some cases,
the bandwidth around the frequency, while still providing the same idea of
cutting or boosting.

It would seem to me that it should be possible to position a microphone at
the listening position and then feed a sweep frequency into the system. The
parametric equalizer could then adjust the response of the system so that it
was flat at the listening position. Once the system is flat, it could then
be adjusted to taste by the listener. Has anyone done this?

Norm Strong

wrote:

It would seem to me that it should be possible to position a microphone at
the listening position and then feed a sweep frequency into the system. The
parametric equalizer could then adjust the response of the system so that it
was flat at the listening position. Once the system is flat, it could then
be adjusted to taste by the listener. Has anyone done this?

It's a common final step of setting up a listening or control room,
after sufficicient acoustic treatment had been done.
It only works optimally for one listening position, but it's better than
not doing it al all.

A similar technique is also often used for equalizing a PA system for a
live music venue.

--
Anahata
-+- http://www.treewind.co.uk
Home: 01638 720444 Mob: 07976 263827

A similar technique is also often used for equalizing a PA system for a
live music venue.

I've played with the Behringer Ultracurve in that mode. Have it output
noise, run a measurement mike back in, let it examine the frequency
spread and adjust. Ideally, repeat at multiple locations within the hall
and average the results.

In article ,
wrote:

"James Lehman" wrote in message
news
Graphic EQs have a set of fixed frequencies that they can either cut or
boost. Parametric EQ's allow you to move the frequency and, in some cases,
the bandwidth around the frequency, while still providing the same idea of
cutting or boosting.

It would seem to me that it should be possible to position a microphone at
the listening position and then feed a sweep frequency into the system. The
parametric equalizer could then adjust the response of the system so that it
was flat at the listening position. Once the system is flat, it could then
be adjusted to taste by the listener. Has anyone done this?

Yes, but if you move the mike six inches, the response totally changes.

Also, most room problems aren't exactly a third-octave wide. If you make
the third-octave response flat, you may well be overcompensating for narrower
problems.

Doing this will make problems much worse, not better.
--scott

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