[Rick Craig]

Any thoughts from DEQX and Behringer (or similar units) owners?

[JoshK]

I regards to what exactly?  How to measure, or whether DSP xo's work well for dipole speakers?

[Rick Craig]

I regards to what exactly?  How to measure, or whether DSP xo's work well for dipole speakers?

I guess what I'm looking for is what kind of results have you had. I'm in the process of coming up with a 3-way design that will be an alternative to the Orion. I don't want to clone the Orion design although it can be done with DSP despite what I've read elsewhere. There are tradeoffs in that design that I don't like and the DEQX will allow me to use other drivers for a different format.

[JoshK]

I have seen a few others do this but not on this site.  As you know, this is the avenue I am going down but haven't had time to make much forward progress on my project.  I think Davey, come to think of it, is using the Behr w/ a pheonix like dipole system.  Then there is Monte Kay.

[ctviggen]

Are they using adjustable filters in the Deqx or are they trying to do speaker/room correction?  I could see some crossover filters and maybe some speaker correction (using one pole of the dipole), but I think room correction would be rough  -- how could you adjust phase on a driver (a pole) when another driver (the dipole) is producing something that will hit the microphone at a completely different phase, based solely on position in the room?

[Rick Craig]

Are they using adjustable filters in the Deqx or are they trying to do speaker/room correction?  I could see some crossover filters and maybe some speaker correction (using one pole of the dipole), but I think room correction would be rough  -- how could you adjust phase on a driver (a pole) when another driver (the dipole) is producing something that will hit the microphone at a completely different phase, based solely on position in the room?

With the DEQX it's up to the user or designer as far as what you decide about filters. As far as the room correction you can alter the room EQ and adjust some of the time domain parameters. Because I have an additional measurement system I can go back and verify the results after the room correction is applied.

As you suggest it can be rather complex, especially for audiophiles who don't have experience in designing and measuring speakers. The good thing is that you can adjust your system to your listening room and personal preferences.

[Davey]

Well, I'm glad Rick is going to apply his efforts to DSP/dipoles and show the rest of us idiots how to do it.  

I don't think anyone said Orion's or clones couldn't be done with DSP, but it does represent a challange and a compromise with regard to dynamic range in the woofer portion.  Especially with commercial units like the DCX.  The DEQX is better suited, but it also has some concerns that can't be overlooked with regard to large EQ's.  Keep in mind that the low Q value of the Orion woofer drivers require a large EQ correction in addition to the nominal dipole roll-off correction.  The addition of these two require boosts in the neighborhood of 40db.  An analog active circuit is by far the best way to accomplish this.

Good luck on the efforts, but many of us have been down this road quite a few years ago.

Cheers,

Davey.

[Rick Craig]

Well, I'm glad Rick is going to apply his efforts to DSP/dipoles and show the rest of us idiots how to do it.  

I don't think anyone said Orion's or clones couldn't be done with DSP, but it does represent a challange and a compromise with regard to dynamic range in the woofer portion.  Especially with commercial units like the DCX.  The DEQX is better suited, but it also has some concerns that can't be overlooked with regard to large EQ's.  Keep in mind that the low Q value of the Orion woofer drivers  ...

No, I wouldn't want to take away anything from what you've done and respect your design capability. The good news is that I know the guys at Audio Artistry and their shop is only five minutes from me    and yes they still are in business.

With the DEQX I won't need all of the EQ since I'm not going to use dipole woofers. As Linkwitz notes on his site room correction via DSP can be quite effective.

[JohnR]

Keep in mind that the low Q value of the Orion woofer drivers require a large EQ correction in addition to the nominal dipole roll-off correction. The addition of these two require boosts in the neighborhood of 40db. An analog active circuit is by far the best way to accomplish this.

That's an interesting point. A DIYer could presumably simply put a 40dB op-amp gain stage in front of the woofer amps to work around this... ?

[Rick Craig]

Keep in mind that the low Q value of the Orion woofer drivers require a large EQ correction in addition to the nominal dipole roll-off correction. The addition of these two require boosts in the neighborhood of 40db. An analog active circuit is by far the best way to accomplish this.

That's an interesting point. A DIYer could presumably simply put a 40dB op-amp gain stage in front of the woofer amps to work around this... ?

Maybe I misunderstood Davey but 40db sounds unusual to me, especially for a system that's sold with 60-watts/channel amps. Open baffle drivers exhibit a 6db/octave rolloff with the frequency shelf depending on the baffle width. I have a friend using the 12" XLS woofers in dipole with the DEQX and I think his EQ was around +12db, maybe a little more.

[_scotty_]

This may be stating the obvious but a level increase of 40dB on the signal going to the amp on the bass section of a dipole relative to the signal level used on the mid and high frequency amps should result in clipping any amplifier that the signal is fed into.  Or if the amp is big enough, the physical distruction of the bass driver.
Scotty

[Davey]

Scotty,

The 40db consists of two shelve filters.....20-110Hz (Q correction of the Peerless drivers to 0.5) 14.8db....and 20-305Hz (6db dipole correction)  23.7db...for a total boost of 38.5db.  However, there is a subseqent padding network that reduces the overall level.

Once the crossover filter is applied the final transfer function result is actually closer to vertical than flat (horizontal)    It is very steep and occupies a full four octaves.  The woofer outputs of the Orion ASP will actually clip with approximately 0.85 volts input at 20Hz.  In real usage you won't drive the crossover with this high a voltage.

DSP crossovers can certainly be programmed (different manufacturers/types in different ways) to mimic this curve, but when programming DSP units with net boosts above the nominal level the input/output levels have to be monitored to prevent clipping in all possible cases.  This "prevention" usually forces the user to lower the input level to the DSP unit far below what would optimal for dynamic range purposes and thus real music levels get very close to the "noise floor" of the unit.  It's a big compromise IMHO.  Many people have noted that these units work "great" or "have no problems" or etc...They haven't looked at the situation as carefully as I have and don't fully understand what the gain structure really is.

Rick,

An Orion clone not using dipole woofers is not really an Orion clone, and Siegfried would most likely take offense at the usage.  In fact the whole concept of an Orion "clone" is rather ridiculous....IMHO.

Cheers,

Davey.

[Rick Craig]

Scotty,

The 40db consists of two shelve filters.....20-110Hz (Q correction of the Peerless drivers to 0.5) 14.8db....and 20-305Hz (6db dipole correction)  23.7db...for a total boost of 38.5db.  However, there is a subseqent padding network that reduces the overall level.

Once the crossover filter is applied the final transfer function result is actually closer to vertical than flat (horizontal)    It is very steep and occupies a full four octaves.  The woofer outputs of the Orion ASP will actually c ...

Thanks for the clarification - now I know why Mac sold his DEQX. No attempt to clone anything here - different drivers, format, crossover points / slopes, and DEQX  

[Davey]

Rick,

I'm not sure that's the (whole) reason why Mac sold the DEQX.  Anyway, he ultimately worked around the EQ situation by adding an outboard, analog shelving filter (that I built for him) to the woofer outputs of his Orion/DEQX setup.  This relieved the DEQX of providing the large EQ requirement on the woofer leg.  I believe that worked really well for him.

There were other "issues" with the DEQX that prompted the sale, but I don't know the specifics....you'd have to ask him about that.

The "problem" I saw with the DEQX...with my limited experience with Mac's unit....was the lack of enough (and this will sound funny) manual control of the curves.  The (automatic) closed-loop operation of the DEQX taking away much of the methodical design and manual control that a speaker designer like Siegfried Linkwitz would like.  However, those more familiar can probably correct me on that.

The biggest problem with the DEQX is the un-godly price.  That's an immediate show-stopper for me and many others.  If they move the decimal point in the price one place to the left then I'd be interested.  As it is, I have no interest.

Why the sudden interest in dipole speakers that the rest of us have been playing with for years?  Thinking outside the box now?  

Cheers,

Davey.

[Rick Craig]

Rick,

I'm not sure that's the (whole) reason why Mac sold the DEQX.  Anyway, he ultimately worked around the EQ situation by adding an outboard, analog shelving filter (that I built for him) to the woofer outputs of his Orion/DEQX setup.  This relieved the DEQX of providing the large EQ requirement on the woofer leg.  I believe that worked really well for him.

There were other "issues" with the DEQX that prompted the sale, but I don't know the specifics....you'd have to ask him about that.

The "pro ...

I think there was a software and / or firmware change that makes the EQ function more versatile and user friendly. There's quite a bit of manual control available that should be sufficient for what I want to do.

I agree that the cost is high but there's really nothing else available for home audio that has the total package. If Behringer or another large company were to make a combination crossover / EQ unit for home use then I think that would be ideal. A unit for well under $1,000 is possible and I'm surprised nobody is doing it.

Several years ago I heard the Audio Artistry speakers but wasn't totally convinced by the dipole concept. Then about 1 1/2 years ago I heard JohnK's dipole back-to-back with the Salk HT3. The dipole was far more convincing in terms of spatial cues - a DEEP soundstage, very open and uncolored sounding. In comparison the HT3 was flat and two-dimensional, not "boxy" sounding, but much less realistic and involving. That's not a knock on the HT3 but a example of the limitations of the conventional box speaker format in terms of soundstage capability. If budget and room placement are not issues then in my opinion it's a no-brainer. Peter Aczel's review of the Orion puts it in words better than I can.

[JoshK]

I agree that the cost is high but there's really nothing else available for home audio that has the total package. If Behringer or another large company were to make a combination crossover / EQ unit for home use then I think that would be ideal. A unit for well under $1,000 is possible and I'm surprised nobody is doing it.

My understanding is the difficulty is more in the software, UI & algorithms. This is ultimately a good portion of the cost.  Behringer has the advantage of scale and cheap production vis-a-vis China.  DSP chips are finally getting to the point where they can handle real time complex computational problems such as convolution.   But with as with any rapidly advancing technology, there are little standards so any software you develop will have to be ported to another chip, which isn't a small task.  

I think ClarityEQ is one of the brave pioneers that is willing to take on doing it for one particular hardware piece without really knowing how large the demand is, at least in the home audio arena. I think ultimately as more capable DSP chips come to fruition, companies like ClarityEQ will divulge themselves of the hardware game (maybe ultimately selling the end product but subcontracting the production) and concentrate on the software sales and getting the UI down.

[gonefishin]

Hi Rick (others)

   I was just wondering if you have had a chance to download the latest firmware and software versions (which have 96k filters among with some other nice improvements)  Software 2.6 and firmware V60.8 (I believe).  It's supposed to be some nice improvements...but I haven't been able to succesfully measure my speakers (due to dropped communications).

    To date...I have got some communication problems which have left me unable to perform the upgrade.  But others seem to talk well about it.

    have fun!

  dan

[JoshK]

I was unaware of SACD support, how do they do that?

[gonefishin]

I was mentioning the 96k filters that they added...below are the upgrades in the 2.2 software (sorry about the mis-type)

 

v2.2
- 96k filtering
- 96k measurements ability
- 96k import ability
- multiple sample-rate support of measurement, viewer, calibrations and
configurations
- copy&paste/drag&drop support for PDC Configuration speaker parameters
- open calibration templates and pdc configuration windows are now
remembered in the project file
and restored when the file is next opened
- improved 'autocomplete' combo boxes
- removed unnecessary measurement window page when adding new data
- added DEQXCalibrated logo to all graphs
- added legends to all printed graphs
- added context menu around status bar link icon for quicker configuration
- revised plot context menu
- added plot toolbar for quick change to frequency/time/group delay and
phase
- can set title of graphs
- revised calibration process
- made calibration templates simpler and more intuitive to use
- can view results of a correction design
- can view design parameters of a correction filter
- support for 96khz calibration templates added
- reduced overall delay when using multiple crossovers / limit filters
- improved responsiveness of 'cancel' button during filter generation
- reduced amount of redraws when using PDC Configurations
- equalizer plot changed to show 'true' eq line, and to make control
points actually sit on the line
- clip lights latch
- Improved confidence metric by only using signal between 20Hz and 20kHz
- Fix bug relating to scaling between measurement channels
- Added ability to synchronise multiple PDCs.

  Like I said...I haven't been able to try it yet, even though it's been out for a while now

   

  (sidenote:  Josh...is the sb2 or sb3 that much of an improvement?  I've been on the fence for sometime...and it's getting difficult not to make the jump on the SB bandwagon.  I wanted to start a #45 tube amp project next...but maybe I should go with the SB instead???)

   thanks,
  dan

[JoshK]

well my foray into Squeeze-o-land is a love/hate affair.  It isn't the SB's fault but the amount of computer stuff I have had to deal with trying to get a file server set up and networked correctly.   I HATE networking.  It is still just so clunky unless you really understand how it all works, which I don't.  Unfortunatley, I don't care to either.

I just had got my SB up and running and then I had another network fall out and lost internet.  An hour later after a tech support call, I had my internet back but no network.   The SB is uber simple to setup if the network is good to go, but there is my problem.

My very brief audition of the SB showed promise, but in stock form with stock wall wart it doesn't beat my Sony XA777.  I do think it has the potential though, just not off the shelf with analog output.  I didn't try a comparison of the SP/DIF output into my DEQX.