Re: my thoughts...
You can just get a Neutrik AES to SP/DIF adapter. It's a little transformer in a package the size of XLR connector.
Like this one (link) . Behringer also makes a device called Ultramatch. It will do the trick for a bit more money.
Then you need a 6 channel volume control, like the one they've been talking about on this board - Apox:
Apox thread
That will get you precision tracking between all 6 channels and remote control.
But my question still stands. Why go to the extend and get generic crossover curves, that are almost guaranteed not to integrate your drivers properly. Drivers roll off naturally. When you combine their response with that of the filter you don't get the slopes that you see on your computer screen while programming the crossover.
In order for textbook electrical crossover curves to work properly with real drivers, the drivers have to have a flat frequency response at least 1.5 to 2 octaves below and above the crossover point.
To illustrate this. If your tweeter has flat frequency response down to 1.5kHz, you should cross it over at 4.5kHz - 6kHz using a textbook electrical filter . Then your final crossover slope for the tweeter will be correct acoustically. But, what midrange driver goes flat to 15kHz - 18kHz ? Thats 1.5 - 2 octaves above the crossover point of 4.5kHz.
Same applies to the woofer/midrange crossover.
With a lot of the digital crossovers you can select different slopes on either side of the crossover point. This lets you come closer to the goal, but you are still limited to textbook filters like Butterworth, Bessel or L/R.
What if your driver's natural roll off is not complementary to any of these curves? You can use EQ to force it a bit, but without measurement capabilities you are shooting in the dark. What you most likely end up with is a generic sounding loudspeaker that cost you a lot of money to put together. Drivers, boxes, digital crossover, amps, volume control all cost good money. Is a mediocre, unexeptional speaker really worth the $2000.00 or more expense?
ralf said:
You can just get a Neutrik AES to SP/DIF adapter. It's a little transformer in a package the size of XLR connector.
Like this one (link) . Behringer also makes a device called Ultramatch. It will do the trick for a bit more money.
Then you need a 6 channel volume control, like the one they've been talking about on this board - Apox:
Apox thread
That will get you precision tracking between all 6 channels and remote control.
But my question still stands. Why go to the extend and get generic crossover curves, that are almost guaranteed not to integrate your drivers properly. Drivers roll off naturally. When you combine their response with that of the filter you don't get the slopes that you see on your computer screen while programming the crossover.
In order for textbook electrical crossover curves to work properly with real drivers, the drivers have to have a flat frequency response at least 1.5 to 2 octaves below and above the crossover point.
To illustrate this. If your tweeter has flat frequency response down to 1.5kHz, you should cross it over at 4.5kHz - 6kHz using a textbook electrical filter . Then your final crossover slope for the tweeter will be correct acoustically. But, what midrange driver goes flat to 15kHz - 18kHz ? Thats 1.5 - 2 octaves above the crossover point of 4.5kHz.
Same applies to the woofer/midrange crossover.
With a lot of the digital crossovers you can select different slopes on either side of the crossover point. This lets you come closer to the goal, but you are still limited to textbook filters like Butterworth, Bessel or L/R.
What if your driver's natural roll off is not complementary to any of these curves? You can use EQ to force it a bit, but without measurement capabilities you are shooting in the dark. What you most likely end up with is a generic sounding loudspeaker that cost you a lot of money to put together. Drivers, boxes, digital crossover, amps, volume control all cost good money. Is a mediocre, unexeptional speaker really worth the $2000.00 or more expense?
Moderation
Thunau
Sorry Jan,
Beeing still under moderation my posts are not directly visible to you, this can take time.
(Please moderators.........
)
I posted a reply to your preveous post already which was not visible to you when you posted your last reply.
Reading your last post I understood better what you where trying to say.
Have to let that sink in a little bit more though.......
Henry.
Thunau
Sorry Jan,
Beeing still under moderation my posts are not directly visible to you, this can take time.
(Please moderators.........
)I posted a reply to your preveous post already which was not visible to you when you posted your last reply.
Reading your last post I understood better what you where trying to say.
Have to let that sink in a little bit more though.......
Henry.
Understanding the limitations and upgrading Pro Gear....
Koinichiwa,
Just some general thoughts here.
1) The AKM A2D and D2A Converters arelikely to be at the very least "very okay". AKM and Cirrus Logic had fo a long time a cooperation, I'd guess that the best AKM Chips are no worse than the best Cirrus Logic (Crystal) ones and either set of chips tend to be "industry standard". Yet, you CAN do better, but only at very high cost (say 6 Channels of PCM1704 with suitable Digital filter etc).
2) The key weaknesses in almost all pro-gear, regardless of price are the analogue stages and powersupplies, especially decoupling. Replacing the PSU Decoupling Cap's with Os-Con's for Digital, Elna Silmic or Black Gate (according to taste) for ananlogue decoupling and analogue coupling duties will help. Going for full "tripplets" or some of the more recently suggested schemes will go further. Changing the Analogue Stage chips for any number of the better quality (especially faster) types will again help, there are plenty of discussions here and at the Audio Asylum.
3) Clocking - the internal clock's only operate when used via analogue Inputs. The on-board clock becomes the main problem only if you use the Unit as A2D with digital outputs, something that is not an option for the Digital X-Over. Still, clean up the powersupply as well as you can for the clock and check if better grade Oscillator chips can be applied.
When using the Digital input the Unit simply slaves onto the source clock. In that case adding a secondary PLL or any other re-clocking scheme that eliminates source and connection jitter will be a good idea. The very least should be a very well cleaned up power supply.
4) For the more adventerous, the analogue input circuitry can be completely converted to use a simple, single transformer plus a pair of resistors and a capacitor. The outputstages require some gain, in many cases again a transformer directly off the DA Chip will work well enough.
I think given the Behringers cost changing the analogue stage Op-Amp's and replacing all 'lytic capacitors plus some work in the Digital I/O and/or Clock section will be reasonably cost effective and give very material improvements. Also, there are many tricks to de-noising Switched more supplies.
Sayonara
Koinichiwa,
Just some general thoughts here.
1) The AKM A2D and D2A Converters arelikely to be at the very least "very okay". AKM and Cirrus Logic had fo a long time a cooperation, I'd guess that the best AKM Chips are no worse than the best Cirrus Logic (Crystal) ones and either set of chips tend to be "industry standard". Yet, you CAN do better, but only at very high cost (say 6 Channels of PCM1704 with suitable Digital filter etc).
2) The key weaknesses in almost all pro-gear, regardless of price are the analogue stages and powersupplies, especially decoupling. Replacing the PSU Decoupling Cap's with Os-Con's for Digital, Elna Silmic or Black Gate (according to taste) for ananlogue decoupling and analogue coupling duties will help. Going for full "tripplets" or some of the more recently suggested schemes will go further. Changing the Analogue Stage chips for any number of the better quality (especially faster) types will again help, there are plenty of discussions here and at the Audio Asylum.
3) Clocking - the internal clock's only operate when used via analogue Inputs. The on-board clock becomes the main problem only if you use the Unit as A2D with digital outputs, something that is not an option for the Digital X-Over. Still, clean up the powersupply as well as you can for the clock and check if better grade Oscillator chips can be applied.
When using the Digital input the Unit simply slaves onto the source clock. In that case adding a secondary PLL or any other re-clocking scheme that eliminates source and connection jitter will be a good idea. The very least should be a very well cleaned up power supply.
4) For the more adventerous, the analogue input circuitry can be completely converted to use a simple, single transformer plus a pair of resistors and a capacitor. The outputstages require some gain, in many cases again a transformer directly off the DA Chip will work well enough.
I think given the Behringers cost changing the analogue stage Op-Amp's and replacing all 'lytic capacitors plus some work in the Digital I/O and/or Clock section will be reasonably cost effective and give very material improvements. Also, there are many tricks to de-noising Switched more supplies.
Sayonara
Audiophile version
Hi,
I am looking into creating an audiophile digital crossover. As co-developer of the APOX volume control, I think there may be merit in a unit designed for best audio performance.
Instead of just allowing Nth order slopes, I am hoping to allow arbitrary mag/phase plots to be downloaded and the appropriate filter coefficients calculated. Of course, standard butterworth and other types would be supported.
I am not sure that a DSP is really needed for the calculations. I would rather use a couple of high speed microcontrollers or even a low end X86 type chip (built in FPU).
Could people list a few high level specs that they would like to be included...
In general, I was anticipating:
SPDIF or I2S inputs
Analog Inputs: 24/96 ADC
six or eight channels 24/96 DAC (3 or 4 way)
Digital output?
Volume control via PGA43XX or PGA23XX in analog domain
Etc...
Dale
Hi,
I am looking into creating an audiophile digital crossover. As co-developer of the APOX volume control, I think there may be merit in a unit designed for best audio performance.
Instead of just allowing Nth order slopes, I am hoping to allow arbitrary mag/phase plots to be downloaded and the appropriate filter coefficients calculated. Of course, standard butterworth and other types would be supported.
I am not sure that a DSP is really needed for the calculations. I would rather use a couple of high speed microcontrollers or even a low end X86 type chip (built in FPU).
Could people list a few high level specs that they would like to be included...
In general, I was anticipating:
SPDIF or I2S inputs
Analog Inputs: 24/96 ADC
six or eight channels 24/96 DAC (3 or 4 way)
Digital output?
Volume control via PGA43XX or PGA23XX in analog domain
Etc...
Dale
Re: Moderation
I can see your posts. I figured it wasn't necessary to respond to you directly, since I addressed your concerns in my other post (like you noted). I suggest that you look at some loudspeaker projects that experienced DIY'ers have posted on the Internet. A good source of information about designing and constructing crossovers is the madisound.com board and partsexpress.com forum. You can learn a lot by just lurking and reading posts.
See
JPO links page for links to a lot of audio related information. Sometimes its better to educate yourself a bit before spending money.
byteboy said:Thunau
Sorry Jan,
Beeing still under moderation my posts are not directly visible to you, this can take time.
(Please moderators.........
I posted a reply to your preveous post already which was not visible to you when you posted your last reply.
Reading your last post I understood better what you where trying to say.
Have to let that sink in a little bit more though.......
Henry.
I can see your posts. I figured it wasn't necessary to respond to you directly, since I addressed your concerns in my other post (like you noted). I suggest that you look at some loudspeaker projects that experienced DIY'ers have posted on the Internet. A good source of information about designing and constructing crossovers is the madisound.com board and partsexpress.com forum. You can learn a lot by just lurking and reading posts.
See
JPO links page for links to a lot of audio related information. Sometimes its better to educate yourself a bit before spending money.
Dale
This is a great idea, and I am up for it.
However, I think it is worthy of its own thread, so I will copy your post and open up a new thread on this topic, and comment there
(edit)
New thread is at-
http://www.diyaudio.com/forums/showthread.php?s=&threadid=16176
This is a great idea, and I am up for it.
However, I think it is worthy of its own thread, so I will copy your post and open up a new thread on this topic, and comment there
(edit)
New thread is at-
http://www.diyaudio.com/forums/showthread.php?s=&threadid=16176
Re: Re: my thoughts...
Jan, I get your point, but as far as I know, the behringer also let you implement a kind of parametric equalisation. With that in mind, you are VERY flexible, and although you might be able to do the same in analog, it would be much more difficult because there will be a lot of interaction between parameters. With the behringer on the laptop you can go through trials in an hour that would take days with conventional components.
I decided to see for myself and ordered one.
For those of you in Belgium: Jacky Claes sound in Hasselt, euro 435 incl BTW.
Jan Didden
Thunau said:
Jan, I get your point, but as far as I know, the behringer also let you implement a kind of parametric equalisation. With that in mind, you are VERY flexible, and although you might be able to do the same in analog, it would be much more difficult because there will be a lot of interaction between parameters. With the behringer on the laptop you can go through trials in an hour that would take days with conventional components.
I decided to see for myself and ordered one.
For those of you in Belgium: Jacky Claes sound in Hasselt, euro 435 incl BTW.
Jan Didden
Re: Audiophile version
I have almost the same plan. I want to make a EQ in a DSP or microcontroller. I hope I'll ever finish that project. It would be very cool to have a chip that interfaces to I2S and can be filled by filter algorithms in the computer. I would like to have parametric EQ implemented. Problem is that I don't know much about microcontrollers and DSPs. But at least, I can program assembly
Fedde
harvardian said:
I have almost the same plan. I want to make a EQ in a DSP or microcontroller. I hope I'll ever finish that project. It would be very cool to have a chip that interfaces to I2S and can be filled by filter algorithms in the computer. I would like to have parametric EQ implemented. Problem is that I don't know much about microcontrollers and DSPs. But at least, I can program assembly
Fedde
Here is the eval board that I just purchased.
http://www.analog.com/UploadedFiles/Product_Briefs/58347282NewADSP-21161N.pdf
Has everything that I need to do development:
DSP, SPDIF (CS8414) recvr, 24 bit ADC, 8 channels of DACS, flash, ram, USB interface...
http://www.analog.com/UploadedFiles/Product_Briefs/58347282NewADSP-21161N.pdf
Has everything that I need to do development:
DSP, SPDIF (CS8414) recvr, 24 bit ADC, 8 channels of DACS, flash, ram, USB interface...
Re: Re: Re: my thoughts...
Yes, I realize that there is (multiple) parametric EQ in the box. With it you can sort of accelerate or slow down a filter slope. But how will you know how this slope adds to the response of the driver? Do you use a measuring setup like Clio, MLSSA, Speaker Workshop, JustMLS etc? Or do you just do it all by ear? Nothing wrong with that, except it's a bit harder to get really good results.
See my Helios project pages .
Can you do something like that (and I'm not claiming this is anything exceptional) with just the DCX2496? It does sound very good through lspCAD emulator.
janneman said:
Yes, I realize that there is (multiple) parametric EQ in the box. With it you can sort of accelerate or slow down a filter slope. But how will you know how this slope adds to the response of the driver? Do you use a measuring setup like Clio, MLSSA, Speaker Workshop, JustMLS etc? Or do you just do it all by ear? Nothing wrong with that, except it's a bit harder to get really good results.
See my Helios project pages .
Can you do something like that (and I'm not claiming this is anything exceptional) with just the DCX2496? It does sound very good through lspCAD emulator.
An externally hosted image should be here but it was not working when we last tested it.
Filterslopes dcx2496
Janneman
That for me too was exactly the reason why I chose something like the Behringer. (After I DID (and still do) read on different speaker forums on the internet......)
I disagree with Jan/Thunau that not fully "perfect" filterslopes do away with all the other advantages of an active loudspeaker system and reduces it to a mediocore speakersystem.
And like you said, there are still equalisation options in the DCX2496.
Furthermore, if you compare the price of the DCX2496 with other high-end active analog crossover kits, it is not bad value for money at all.
But I agree, he has got a point.
Without good knowledge and measuring tools it is a bit difficult to set it up correctly and get the most out of your invested money.
That is why I was looking already for a good real time analyser with measuring microphone etc. and was reading forum threads on how to use such tools (on the Internet... ) to be able to evaluate the adjustments.
Henry.
Janneman
That for me too was exactly the reason why I chose something like the Behringer. (After I DID (and still do) read on different speaker forums on the internet......)
I disagree with Jan/Thunau that not fully "perfect" filterslopes do away with all the other advantages of an active loudspeaker system and reduces it to a mediocore speakersystem.
And like you said, there are still equalisation options in the DCX2496.
Furthermore, if you compare the price of the DCX2496 with other high-end active analog crossover kits, it is not bad value for money at all.
But I agree, he has got a point.
Without good knowledge and measuring tools it is a bit difficult to set it up correctly and get the most out of your invested money.
That is why I was looking already for a good real time analyser with measuring microphone etc. and was reading forum threads on how to use such tools (on the Internet...
) to be able to evaluate the adjustments.Henry.
Re: Filterslopes dcx2496
I said exactly that a few posts back. Once you aqcuire the tools and knowledge of what makes a good loudspeaker, the generic stuff like Behringer crossover, doesn't look as attractive anymore. I spent $25.00 on circuit boards from ESP and maybe $35.00 on very good quality parts to construct a 3 way crossover that I designed in lspCAD pro. Granted, the required software , multichannel soundcard, preamp and microphone are not cheap. But in my experience, they will give you better results in the long run.
The type of processing you get in the Behringer was sucessfully used in many pro-audio applications for last 10 years or so. It used to cost about 10 times as much too. If you use RTA ( hardware or software) and calibrated microphone you can arrive at a decent sounding system. But the last few percent of performance will be achievable only with dedicated analysis tools and custom electronics or DSP algorithms. Remember that RTA will include the effects of the room in the measurements. And those vary from one location to another. So, when you move your speakers and listening position, you will have to adjust the crossover accordingly. You should use semi-anechoic measurements where possible. Those are immune to ambiant noise and room effects (to a large degree).
There is a lot of solid knowledge floating out there on the Internet. Take advantage of it.
My recommendation for you would be to download demo version of lspCAD and play with it. Then download Speaker Workshop (freebie) and learn how to measure loudspeakers with your sound card and a decent microphone. The Behringer ECM8000 could do, if you luck out and pick up one that is to spec. At FRD Consortium you can get a very nice utility called d-player. It will simulate crossovers that you design in other software. If you have a multichannel sound card it will do for you what the Behringer DCX2496 does, only better customized to your drivers (provided you designed a good crossover and entered it in the d-player correctly).
LspCAD pro and SoundEasy will do all that (measure, design, simulate) in one package for about the same price as DCX2496 (lspCAD pro) or much less (SoundEasy).
byteboy said:But I agree, he has got a point.
Without good knowledge and measuring tools it is a bit difficult to set it up correctly and get the most out of your invested money.
That is why I was looking already for a good real time analyser with measuring microphone etc. and was reading forum threads on how to use such tools (on the Internet...
Henry. [/B]
I said exactly that a few posts back. Once you aqcuire the tools and knowledge of what makes a good loudspeaker, the generic stuff like Behringer crossover, doesn't look as attractive anymore. I spent $25.00 on circuit boards from ESP and maybe $35.00 on very good quality parts to construct a 3 way crossover that I designed in lspCAD pro. Granted, the required software , multichannel soundcard, preamp and microphone are not cheap. But in my experience, they will give you better results in the long run.
The type of processing you get in the Behringer was sucessfully used in many pro-audio applications for last 10 years or so. It used to cost about 10 times as much too. If you use RTA ( hardware or software) and calibrated microphone you can arrive at a decent sounding system. But the last few percent of performance will be achievable only with dedicated analysis tools and custom electronics or DSP algorithms. Remember that RTA will include the effects of the room in the measurements. And those vary from one location to another. So, when you move your speakers and listening position, you will have to adjust the crossover accordingly. You should use semi-anechoic measurements where possible. Those are immune to ambiant noise and room effects (to a large degree).
There is a lot of solid knowledge floating out there on the Internet. Take advantage of it.
My recommendation for you would be to download demo version of lspCAD and play with it. Then download Speaker Workshop (freebie) and learn how to measure loudspeakers with your sound card and a decent microphone. The Behringer ECM8000 could do, if you luck out and pick up one that is to spec. At FRD Consortium you can get a very nice utility called d-player. It will simulate crossovers that you design in other software. If you have a multichannel sound card it will do for you what the Behringer DCX2496 does, only better customized to your drivers (provided you designed a good crossover and entered it in the d-player correctly).
LspCAD pro and SoundEasy will do all that (measure, design, simulate) in one package for about the same price as DCX2496 (lspCAD pro) or much less (SoundEasy).
Jan,
For the record, I disagree for a number of reasons.
Most importantly is the effect that a passive crossover has on the bass element, as well as the flexibility to change things after the fact.
Then again, it is of course possible to make good quality equipment the old fashioned way as well
Petter
For the record, I disagree for a number of reasons.
Most importantly is the effect that a passive crossover has on the bass element, as well as the flexibility to change things after the fact.
Then again, it is of course possible to make good quality equipment the old fashioned way as well
Petter
Petter said:Jan,
For the record, I disagree for a number of reasons.
Most importantly is the effect that a passive crossover has on the bass element, as well as the flexibility to change things after the fact.
Then again, it is of course possible to make good quality equipment the old fashioned way as well
Petter
I'm all for digital equipment Petter. Don't get me wrong. What I have a problem with is the limitations imposed on the user by makers of the boxes that are available. A textbook LR4 ELECTRICAL crossover with real world drivers will not result in a LR4 ACOUSTIC crossover ( and ACOUSTIC response is what matters in the end result) . It will be something between 4th and 6th or even 7th order with all the resulting frequency, transient, phase, polar response consequences.
The user of a box like the Behringer has no feedback as to what his system is doing acoustically, unless he measures every time he makes an adjustment, or has something like JBL SMAART running along the side of loudspeaker CAD software.
One manufacturer that has sort of cought on is Electrovoice. Their digital crossover lets you import individual driver's frequency responses (but only EV drivers saved in some proprietary file format - what a rocket) and see final acoustic response of the system. Their electrical curves are also only textbook filters, but at least you see what you're doing with them.
A good digital crossover will have the ability to import frequency response files of any driver (maybe provide a utility to measure your drivers from within the opearting system of the box), combine these files with the filters generated by the box and display the predicted final combined acoustic response of the system. The crossover filters will include the possiblity of stacking multiple 1st order filters at different frequencies on the same output as well as multiple variable-Q second order filters. This combined with delays and all-pass filters would let you achieve what most analog crossovers could do for the last 50 years.
In addition the outputs would be followed by high quality VCAs, so the user can control the output levels with one 20k pot or IR commands without decimating his digital resolution.
As to flexibility of the Behringer box. Yes, you can adjust everyday. So what. Would you rather drive a different Hundai every day or have one Mercedes Benz for a long time.
Hi all
A digital crossover would be interesting for me because
it would then be easier to time-align a 4-way horn system,
that's not so easy otherwise, when the treble-horn is maybe
10 cm long and the bass-horn is maybe 360 cm long.
It's also easier to try out several different crossover configurations
more or less simultaniously , let's face it ,
in my case I've been trying to build good loudspeaker-systems for 15-20 years, but crossovers are bl**dy difficult to get right,
and yes, I have a couple of PC based measuring systems at home,
so I can measure what I build.
So , a digital crossover =
one of many tools one can use to try to build a better speaker
cheers
A digital crossover would be interesting for me because
it would then be easier to time-align a 4-way horn system,
that's not so easy otherwise, when the treble-horn is maybe
10 cm long and the bass-horn is maybe 360 cm long.
It's also easier to try out several different crossover configurations
more or less simultaniously , let's face it ,
in my case I've been trying to build good loudspeaker-systems for 15-20 years, but crossovers are bl**dy difficult to get right,
and yes, I have a couple of PC based measuring systems at home,
so I can measure what I build.
So , a digital crossover =
one of many tools one can use to try to build a better speaker
cheers
Re: Filterslopes dcx2496
It also depends on how you look at it. For me, the fact that every change to the passive xover after the poweramp interacts with the speaker parameters is defenitely a disadvantage. Try to tweak the passive xover for flat freq response AND flat delay, impossible 99 out of 100 cases. With the DSP its just a piece of cake. The only unknown for me is the quality of the DACs in the behringer (I would use the digital input straight from the CD transport.
And as far as measuring is concerned, I'll try to do it by ear to start with. But, you see, I have this Audio Precision System One with DSP and MLS analyzer...
Now who has a good measuring mike for sale?
Jan Didden
byteboy said:Janneman
That for me too was exactly the reason why I chose something like the Behringer. (After I DID (and still do) read on different speaker forums on the internet......)
I disagree with Jan/Thunau that not fully "perfect" filterslopes do away with all the other advantages of an active loudspeaker system and reduces it to a mediocore speakersystem.
And like you said, there are still equalisation options in the DCX2496.
Furthermore, if you compare the price of the DCX2496 with other high-end active analog crossover kits, it is not bad value for money at all.
But I agree, he has got a point.
Without good knowledge and measuring tools it is a bit difficult to set it up correctly and get the most out of your invested money.
That is why I was looking already for a good real time analyser with measuring microphone etc. and was reading forum threads on how to use such tools (on the Internet...
Henry.
It also depends on how you look at it. For me, the fact that every change to the passive xover after the poweramp interacts with the speaker parameters is defenitely a disadvantage. Try to tweak the passive xover for flat freq response AND flat delay, impossible 99 out of 100 cases. With the DSP its just a piece of cake. The only unknown for me is the quality of the DACs in the behringer (I would use the digital input straight from the CD transport.
And as far as measuring is concerned, I'll try to do it by ear to start with. But, you see, I have this Audio Precision System One with DSP and MLS analyzer...
Now who has a good measuring mike for sale?
Jan Didden
Koinichiwa,
Well, I'll ignore at the moment the comment about "limitations", as with any of the modern Digital X-Overs known to me you can choose the slope of each filter (both HP and LP for Bandpass) and add further EQ's. The DCX2496 appears especially flexible as you can combine any given order of HP and LP slope with further parametric shelving filters untill you run out of processing power to give any funloving phase and amplitude response.
Now as for the subject of the Behringer box not calculating your X-Over for you, hell, not much out there will both provide comprehensive measurement facilities and crossover production.
But as old pro-sound hand I would argue that if you combine a suitable digital X-Over with a reasonable spectrum analyser and 'scope you can adjust the unit very well to give excellent results. With 'scope and Mike (or for the behringer automatically) you use a single repetetive spike to timealign all drivers. Then you use suitable basic slopes to give some form of X-Over and use the analyser to first set the levels of each way and thenuse the parametrics and x-over slopes to compensate driver peculiarities. Using the "mute way" buttons or soft functions allows you to see and hear with pink noise each way on it's own. The result of doing such setup reasonably diligent is very good.
In the old days we had to manually mechanically "timealign" the speakers and then using much less comprehensively equipped electronic X-Overs and added EQ's to achieve the same and it was more difficult.
I would argue that this is better handled as stand alone PC-Software which can be used to simulate. If you use Calsod you can add any numbers of filters (such as minimum phase parametrics, highpass, lowpass & bandpass of all sorts of orders and classes) in series with your driver and simulate.
Get Calsod to adjust the Filters and read of the settings and apply to a generic, off the shelf, digital X-Over selling for around 400 Bucks. What's wrong with that? Why add a huge overhead to a simple device? If you go that far, the X-Over should really measure the driver response by itself, measure the room effects as well and simply adjust itself to provide a given desired traget response at the listening position. Because to import driver files etc. still needs loads of external gear.
The terms "high quality" and "VCA" are mutually exclusive.
For a minumum switched resistor networks are needed, but with DAC's now easily pushing past -120db S/N I think an analogue preset in 3 or 6db setps plus Master volume control and individual channel level control should be handeled in the digital domain.
Well, I think the issue is quite different. The Behringer (and similar units) can do a lot of things you either cannot do in the analogue domain or only can do with great difficulty. By the time you have daisychained the 4pcs of Op-Amp's needed for a 4th order LR bandpass and the added Op-Amp's for 2 - 3 parametric Equalisers to EQ the drivers in the analogue domain you have a lot of accumulated tolerances, loads of active and passive components and any number of other issues to contend with.
And if you use really high quality Op-Amp's and passives (and a good PSU) you can afford the Behringer XO PLUS all the money in parts to upgrade the analogue stages and you end upo with a unit that will be most likely still more transparent.
Add to that the ability to make iterative small changes to adjust the results to taste, to have multiple programs (like a "loud" one that uses steep slopes to protect drivers for parties and low order slopes for normal listening and many more) at the touch of a button I certainly can see the attraction.
Sayonara
Well, I'll ignore at the moment the comment about "limitations", as with any of the modern Digital X-Overs known to me you can choose the slope of each filter (both HP and LP for Bandpass) and add further EQ's. The DCX2496 appears especially flexible as you can combine any given order of HP and LP slope with further parametric shelving filters untill you run out of processing power to give any funloving phase and amplitude response.
Now as for the subject of the Behringer box not calculating your X-Over for you, hell, not much out there will both provide comprehensive measurement facilities and crossover production.
But as old pro-sound hand I would argue that if you combine a suitable digital X-Over with a reasonable spectrum analyser and 'scope you can adjust the unit very well to give excellent results. With 'scope and Mike (or for the behringer automatically) you use a single repetetive spike to timealign all drivers. Then you use suitable basic slopes to give some form of X-Over and use the analyser to first set the levels of each way and thenuse the parametrics and x-over slopes to compensate driver peculiarities. Using the "mute way" buttons or soft functions allows you to see and hear with pink noise each way on it's own. The result of doing such setup reasonably diligent is very good.
In the old days we had to manually mechanically "timealign" the speakers and then using much less comprehensively equipped electronic X-Overs and added EQ's to achieve the same and it was more difficult.
I would argue that this is better handled as stand alone PC-Software which can be used to simulate. If you use Calsod you can add any numbers of filters (such as minimum phase parametrics, highpass, lowpass & bandpass of all sorts of orders and classes) in series with your driver and simulate.
Get Calsod to adjust the Filters and read of the settings and apply to a generic, off the shelf, digital X-Over selling for around 400 Bucks. What's wrong with that? Why add a huge overhead to a simple device? If you go that far, the X-Over should really measure the driver response by itself, measure the room effects as well and simply adjust itself to provide a given desired traget response at the listening position. Because to import driver files etc. still needs loads of external gear.
The terms "high quality" and "VCA" are mutually exclusive.
For a minumum switched resistor networks are needed, but with DAC's now easily pushing past -120db S/N I think an analogue preset in 3 or 6db setps plus Master volume control and individual channel level control should be handeled in the digital domain.
Well, I think the issue is quite different. The Behringer (and similar units) can do a lot of things you either cannot do in the analogue domain or only can do with great difficulty. By the time you have daisychained the 4pcs of Op-Amp's needed for a 4th order LR bandpass and the added Op-Amp's for 2 - 3 parametric Equalisers to EQ the drivers in the analogue domain you have a lot of accumulated tolerances, loads of active and passive components and any number of other issues to contend with.
And if you use really high quality Op-Amp's and passives (and a good PSU) you can afford the Behringer XO PLUS all the money in parts to upgrade the analogue stages and you end upo with a unit that will be most likely still more transparent.
Add to that the ability to make iterative small changes to adjust the results to taste, to have multiple programs (like a "loud" one that uses steep slopes to protect drivers for parties and low order slopes for normal listening and many more) at the touch of a button I certainly can see the attraction.
Sayonara