Is anyone using a ULN8 as an active crossover ?

[twist222]

Rajupatel

 

Thanks for your thoughts on this subject.

 

I must agree that the errors bought about by room acoustics make the errors from every other part of our system look miniscule perhaps with the exception of the speakers themselves.

 

I have measured speakers, in both (a) as close as practical for me to an anechoic space and then (b) the same speakers speakers in room. Depending on the speakers and looking at something as simple as frequency response, the deviations from anything like a flat response is huge. (and that is not even considering things like phase, time alignment and impulse response. If you had an amplifier that measured that poorly I expect most people would throw it out. Would you buy and amplifier that had a measurable delay on all frequencies below 200Hz ? No but plenty of people happily by and use speakers that do exactly this. (I have only used 200Hz as an example. It will depend on the frequency of the crossover to the bass driver and some speakers do correct for this but many don't).

 

Of course that does not stop us (me included) using things like Amarra on a G5 rather then iTunes on a Mac mini. Whilst I have never tried it I doubt I could get a measurable difference between the two.

 

For me the biggest benefit of using the DEQX has come from (1) going to a fully active speaker setup (in my case 3 way) (2) speaker corrections (based on as close as practical to anechoic measurements). Yes I have also used room correction with the DEQX and it can help but it is not my primary focus.

 

With speaker correction I am currently working on building the best speakers I can, that need the least amount of correction from the DEQX.

 

Once I get the speakers right the longer term goal is to work on room acoustics, particularly focusing on the deficiencies identified by measuring the in room response of the speakers.

 

I did a little research on the treatment applied to the Amp/Speaker Monitoring System of Ole Lund Christensen and Paul Ladegaard System which is the subject on your link http://www.rayofsound.com/gamut.tas.html. It must sound awesome but at $250,00 for a studio treatment or $100,000 for a domestic room it is a little out of my price range.

 

I my additional research on this I did find a very interesting article from the San Diego Music and Audio Guild http://sdmag.org/articles/the-great-divide/ that helped me understand where Ole Lund Christensen and Paul Ladegaard were going. For anyone interested in speakers or room acoustics it is certainly worth a read.

 

I also had a look at HOLM Acoustics. I see that they have a similar, but different system to DEQX.

Do you have any idea what sort of price range the DSPre1 from HOLM is in ? It would be good if they could upgrade there USB input to accept better than 16 bit 44.1/48. I expect they will in time.

 

So thank you very much for some very interesting directions.

 

It looks like you have focused on the room correction via the way it is constructed. For me, I think you are correct but at this time my current focus and interest is on speaker measurement and correction.

 

Regards

 

Mark

 

 

[rajupatel]

Quote "I also had a look at HOLM Acoustics. I see that they have a similar, but different system to DEQX.

Do you have any idea what sort of price range the DSPre1 from HOLM is in ? It would be good if they could upgrade there USB input to accept better than 16 bit 44.1/48. I expect they will in time." unquote.

 

Yes the HOLM people are working to improve their input and also to accept higher sampling frequencies. They are priced at about $5000 retail if i am not wrong. It does all that DEQX can do.

 

"Once I get the speakers right the longer term goal is to work on room acoustics, particularly focusing on the deficiencies identified by measuring the in room response of the speakers."

 

I suggest you do the room first and then the speakers to match. Remember it is ultimately a box playing in a bigger box. If in the first place the bigger does not sound like a box, then a lot of problems are solved and need not be addressed (or rather tackled) in the first place.

 

Incidentally the FOCUS system does electronically (in the analogue domain) what you propose to do using the DEQX system. Please see http://www.holmacoustics.com/dspre1_introduction.php

 

Wish you all the luck in your endeavour. If you do need any help, then please do let me know. If you do visit Mumbai, INDIA, then you are welcome to come to my place and listen. Take care.

 

Raju Patel

 

 

 

 

[twist222]

Are you able to share some of the details about your room ? From what I understand Ole Chrisrensen and Paul Ladegraard designed and constructed the room so that the geometry resulted in all early reflections being directed away from the listening position.

 

I would love to understand a little better how they achieved this.

 

 

Regards

 

Mark

 

 

 

[Nyal Mellor]

Implementing FIR correction for each driver, then implementing crossovers, then FIR correction for the room is what the deqx does. I reckon (not based on experience) that it would be very difficult to achieve the same effect with the Metric Halo; you would be able to do the xo part ok, but implementing FIR correction for each driver and the room I think you would a) have to manually generate each FIR filter yourself (something the deqx does as part of its software) and b) run out of DSP (FIR filters being resource intensive).

 

[sdevino]

The DEQX looks pretty interesting and probably would be a little more turn-key than the MH unit. It would be interesting to compare sound quality sometime though.

 

But I really wanted to comment on physical verse electronic room compensation. One advantage of physical room compensation is that it generally focuses on things like diffusion and bass trapping which are beneficial to the entire listening space (in general).

 

Electronic room compensation is usually only valid at the measurement point. To be effective beyond a single sweet spot the analysis needs to be done for all critical listening areas and then the best set of trade offs would be made. This is where a ULN8 can be advantageous. With 8 input channels the user can measure at 8 locations simultaneously and use MH DSP primitives such as band split and all pass filters and delays to make time based adjustments while using the transfer function of Spectra Foo to see full spectrum realtime phase and power response on all channels.

 

Electronic speaker/crossover compensation can of course be very effective as long as the speaker is measured in a well managed acoustic environment that considers the design of the driver.

 

Steve

 

[bdiament]

Hi Steve,

 

"...But I really wanted to comment on physical verse electronic room compensation..."

 

My approach is to address problems at the source, hence I prefer physical means for addressing the physical issues of the room.

 

While I know many folks who enjoy the results of electronic "compensation", my feeling is that room issues tend to be primarily time-based in nature (though they partially manifest themselves in the amplitude domain). Hence, attempts to address time-based issues with amplitude-based "remedies" are - to my ears- not successful. They certainly effect change but I can't help thinking of the examples I've heard over the years (from many manufacturers) as attempting to fix a broken arm by wearing a different hat.

 

Just my perspective.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[rajupatel]

send me an email on [email protected] and i can share what i did...

 

RajuP

 

[manisandher]

"Incidentally i have also tried and tested DEQX correction in my room and i felt the sound had lost it's life."

 

This was exactly my experience also. But I found it absolutley phenomenal for taking measurements. I'd actually be tempted to have it on standby just for that!

 

Mani.

 

[Nyal Mellor]

Something to consider - I always think you should use the right tool for the job. Deqx was designed explicitly for driver correction, XO and room correction. The metric halo unit was not. Trying to work around the limitations of a product always gives many headaches and much time wasted that could be better spent on other things. At least that is my experience from designing and building software...

 

Better to spend that money you were going to spend on the metric halo dealing with problems the deqx can't fix e.g. strong reflections, lack of low frequency absorption, etc.

 

[sdevino]

Its hard to argue with the advantages of the Deqx for speaker compensation, but one of the primary objectives for the original Metric Halo 2882 was multi-zone measurement for Spectra Foo. The 2882 and now the ULN8 are commonly used for large room (professional theatre) measurement because of the ability to compare multiple parts of the room.

 

That id meant to imply that it is better than the Deqx but it is a primary purpose for the tool. I think which is best depends somewhat on what you are doing and what your comfort level is with various tools. I like spectra Foo because I can clearly see what it is measuring and make my own decisions. Obviously the MH interfaces are better for Spectra Foo.

 

I imagine the Deqx would be far more convenient for speaker alignment.

 

Barry, I agree with your comment. So far I much prefer gear that needs no alignment, and I rarely if ever recall a cal procedure actually improving what was already a good design.

 

Steve

 

 

 

[Nyal Mellor]

Steve - agreed, DEQx software would not seem to be a replacement for professional quality measuring app e.g. spectrafoo, arta, smaart. That would be the issue with using a metric halo - without knowledge of what the measurements are telling you and how to develop filters to correct them getting good results out of the metric halo unit would be much harder. AFAIK (not having used DEQx), it seems like a much more 'wizard' based approach, where the user is clearly guided through the process, and told which measurements to take at which point in time, etc.

 

Interestingly I hadn't considered a metric halo units as external soundcards for sound system alignment / calibration. Do you know if there is a matrixing facility built into Spectrafoo i.e. the ability to take measurements from different inputs in turn over an extended period of time (often called a spatially averaged measurement in the theater alignment world). That could be interesting as a replacement to external offboard hardware matrix units.

 

[twist222]

I have the apple remote and Amarra but don't have a ULN 8 !!! I had hoped someone with the ULN 8 and Amarra had already tried it. Initial advice is that it only works with the Sonic Studio hardware.

 

[twist222]

Sorry if this reply is a bit out of order. It was supposed to be a reply to Mani Sandher's comment which is now about 4 or 5 posts above.

 

Mani,

 

Your experience that "i felt the sound had lost its life" is different to mine. I am not sure at what level you were using the DEQX ? From your post I expect it was for room correction on a set of speakers that utilised a passive crossover. A valid use for the DEQX but perhaps different to my preferred use for the DEQX.

 

I have now used a DEQX for almost 6 years and have a few friends who also use them in their systems.

 

We have found the DEQX gives fantastic results when used to (a) run your speakers fully active (b) take good measurements of the speaker, measuring each driver individually (as close to anechoic as practical) © correct the individual drivers for frequency, phase, time alignment based on your anechoic measurements (d) put the speakers back in the listening room and manually apply room correction.

The dynamics and speed you achieve with a fully active speaker never ceases to surprise me. Using speaker corrections with the DEQX you are then able to get the 3 separate drives to integrate into a single whole speaker. I use the room correction function as a bit of icing on the top. For me room corrections is not where the real magic of the system happens.

 

Others may disagree but from my experiences one of the keys to achieving good speaker corrections with a DEQX is in obtaining good results. Remember we are trying to measure and correct the individual drivers. Not measure room effects. Room correction is a separate operation carried out after you have corrected the speakers. I have only ever achieved good speaker correction measurements (remember as close to anechoic as practical) by moving the speakers out of the listening room. In my case taking DEQX, amps, speakers, leads etc to a large hall with a high ceiling and getting the speaker a long way off the ground. The purpose being to get the first reflection out of the zone where we are trying to measure.

 

The DEQX speaker corrections are based on the measurements you take. Give it poor measurements and you probably know what sort of result to expect.

 

Sorry if I have got a little off track here but I suppose I sometimes get a little frustrated when people who are trying to use a DEQX say it is too hard to move their speakers etc. Yes there are measuremnt methods with the DEQX that allow you to measure in room BUT ... You only have to measure (speaker measurements) them once. So yes it will take one day to pull your system apart, take them to a suitable location and measure them. But how long are you going to spend after that listening to the speakers ? Once you have a good set of speaker measurements you can come home and spend as long as you like building and trying out different filters for speaker correction.

 

So this is not actually a comment on Raju's and Mani's experience but hopefully I have provided a little more background on how I have been using a DEQX and what has worked for me.

 

So why am looking at a ULN 8 / Model 305 ? Well after 6 years with a DEQX (2.6 then 3.0) I am planning in keeping the DEQX but would like to try another option. Preferably in a more open environment with even more control over how you achieve the end result.

 

Regards

 

Mark

 

[manisandher]

Hi Mark,

 

Thanks for sharing your experiences.

 

You said, "I am not sure at what level you were using the DEQX ? From your post I expect it was for room correction on a set of speakers that utilised a passive crossover. A valid use for the DEQX but perhaps different to my preferred use for the DEQX."

 

No, absolutely not. I was using it in a fully active system - all passive crossovers removed from speakers. I was following your a) to d) approach to the letter... and not a pleasant experience with heavy speakers! Like you, room correction was an added bonus, but certainly not the reason why I went for a DEQX.

 

As I said, I loved the measurement capabilities (speaker not room) of the DEQX, and would seriously consider getting another one just for that. But I would use these measurements to help me optimise my Pass Labs XVR1 crossovers. (At the moment, I do everything by ear, voicing the speakers to match a pair of AKG K-1000 headphones.)

 

I think the DEQX is a great product... just not up to performance of a good DAC and a good electronic crossover. This is not a criticism, by the way. In pure cost terms, my current DAC/x-overs cost 5-10 times more than a DEQX... and you get a damn site less flexibility for the money.

 

Mani.

 

[bdiament]

Hi Mark,

 

I believe the "initial advice" is mistaken. (Unless perhaps it is specifically with reference to Amarra.) While I don't have an Apple remote, I know folks who have used them with ULN-8s.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[sdevino]

Hi Barry,

FYI the Apple Remote support for the ULN8 was disabled during the beta cycle. It worked great so I hope it makes a reappearance someday.

 

Steve

 

[bdiament]

Hi Steve,

 

Thank you.

Guess I missed the "opportunity" as I never got an Apple remote.

(I'm still running 10.5 my PowerBook G4 too. Will go for a MacBook Pro this Spring - the old machine is getting slow with 192k.)

 

I too hope use of the remote makes that reappearance some day.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

 

[twist222]

Barry,

I am actualy replying to your advice about using an apple remote with the ULN 8 I have moved my reply to the end of the post as I almost missed your post on this.

 

You wrote "I believe the "initial advice" is mistaken. (Unless perhaps it is specifically with reference to Amarra.) While I don't have an Apple remote, I know folks who have used them with ULN-8s."

 

I know you can use the apple remote to control the volume in Amarra and if you have the Model 4 Amarra will control the analoge volume control in the Modell 4 rather then Amarra just applying digital attenuation.

 

What are the people with apple remote and ULN 8's controling with the remote ?

 

Is there a way to remotly control the analoge volume in the ULN 8 with an apple remote without using Amarra ?

 

Regards

 

Mark

 

[Nyal Mellor]

@Mark

 

How would you propose using ULN 8 to do driver correction?

 

[bdiament]

Hi Mark,

 

"What are the people with apple remote and ULN 8's controling with the remote ?"

 

Steve could answer this better than I but apparently, for now at least, it will be in the past tense. (Not having an Apple remote myself, I didn't know until Steve told me -see his post- that use of the remote on the ULN-8 was disabled during the beta run.)

 

I believe it functioned as more than a simple volume control. But again, I think Steve actually used it, I didn't.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[sdevino]

The ULN8 supported the use of the Apple Remote during the beta stage. For various reasons that capability was withdrawn prior to release. I am under NDA so I will not discuss this in further detail.

 

That being said, Since my shuttle Express is the primary control I use for operating the ULN8's analog console volume it should be possible to use any device or application which provides detectable key commands.

 

The key sequence for volume control is "cmd-opt-ctrl-" followed by up or down arrow.

 

I would test my remote with Amarra but I decided not to purchase a license after my demo ended.

 

Steve

 

[twist222]

Thanks to Steve and Barry for your follow up on the a remote volume control for the ULN 8. I am very glad I asked further.

 

Steve's thanks for letting me know about the shuttle express. I have not had time to research it well but it looks like the shuttle express mini will easily provide remote volume control and (without checking) transport controls for iTunes. Alternatively it looks like I could use any device that will learn key commands.

 

Whilst remote volume may seem a small item, having this option with the ULN 8 means that I freely chose between MH or SS without feeling forced to go with one to get remote analogue volume control.

 

That said I have found both companies responsive and good to deal with. So I can now look at any differences in their software set up that may effect me.

 

Regards

 

Mark

 

 

 

[twist222]

Thank you to everyone for posting your advice on the use of a ULN8/ Model 305. I have been reading everybody's comments with great interest. Unfortunately I have not been available to reply to them all. So I thought I might summarize some of my current thoughts after reading your posts.

 

Let me also say thanks to all for the positive way you have responded.

 

My current thoughts are

 

(1) As a starter I should reiterate that I have had a DEQX for almost 6 years although for most of this time I have not used it in an optimal configuration.

 

(2) I am planning on retaining the DEQX but am interested in exploring other ways of applying an active crossover, speaker correction and room correction. It will actually be good to be able to compare it with the DEQX from both a measurement and listening perspective.

 

(3) A couple of people here have found other options that for them sound better sonically than the DEQX

 

(4) I am only doing this as a hobby so time spent learning new ways of doing things and trying different ways is not a financial issue

 

(4) I have heard very few negative comments about the hardware side of the ULN8/Model 305

 

(5) The ULN8/Model 305 seems to have the hardware side of what I am looking for well covered

 

(6) There are few, if any, other interfaces with similar capabilities including this level of on board DSP that are not part of a propriety system eg DEQX, Holm Acoustics - Yes I realize there are also restrictions with the ULN 8/Model 305.

 

(7) For most people a proprietary system is probably the best option but I am keen to experiment and play with different options and live with the required learning curve.

 

(6) After it has been set up with a computer the onboard DSP allows the unit to be run in stand alone mode.

 

(7) Spectra Foo and Fuzzmeasure should provide me with a solid base for taking measurements.

 

(8) So hardware and measurement software is looking good BUT !!

 

(9) Software for applying speaker and room correction ?? Yes there is software with the system that will work and may work very well but how this compares with proprietary software written specifically to do what I am trying to do - only time will tell.

 

(10) Most, if not, all of the proprietary systems use FIR filters for speaker correction. Does this say something ?

 

(11) As a pro based system the plugins for for the ULN8/Model 305 tend to be based on IIR filters that have lower latencies but, if I understand matters correctly, can cause more phase issues.

 

(12) FIR filters could be constructed for the ULN8/ Model 305 but would involve setting them up at would I would call a low level.

 

(13) The whole console/ plugin system would be a large learning curve for me. Particularly if I chuck in a DAW interface so I can use additional plugins. That said it might help me understand the pro side of things just a little better.

 

(14) The ULN 8/Model 305 is Mac OS only based. This further restricts available software for the speaker correction. ie I have seen a few FIR based correction systems that are windows based but to date nothing similar on the Mac platform.

 

(15) No windows base firewire drivers also mean that even after you have configured the ULN 8 Model 305 with a Mac I still cannot use the firewire connection to a windows based computer for music playback

 

(16) That said I am very happy with Amarra as a playback engine on my G5 powermac. No if only the powermac had less fan noise on a hot day !

 

(17) The pro blokes seem to concentrate on measuring the speaker in the venue and then applying correction to the whole of system in room response. I am used to correcting the speaker separately (part of building the crossover / correction filters and then correcting the in room response of the corrected speaker.

 

(18) I agree with the comments about acoustically treating the room rather than applying electronic room correction.

 

(19) The measurement tools - Spectra foo and Fuzzmeasure should also be great in identifying room problems.

 

(20) Chicken or the Egg ? Do you correct the speaker and then see how it measures in the room or do you attack the room first ?

 

(21) We haven't been overrun by people replying and saying they have done something similar. Does this say something or is it more related to the choice of forum

 

So I think I may now have more questions than when I started but at least after reading your comments and advice I am sure I also know a little more.

 

Now Nyal if you are still with us I m very keen to have a go at answering your question "How would you propose using ULN 8 to do driver correction?" Not because I believe I know the 100% correct answer ( particularly as I have never even used a ULN8) but I will be keen on receiving comments on how I would currently do it.

 

Unfortunately because of time I will have to leave it to my next typing session. ( I am not a fast at typing)

 

If anyone has managed to get to the end of this I hop[e I haven't bored you too much.

 

Till next time

 

Mark

 

 

[sdevino]

I think many pros (in my case I am a sound designer for theatre) actually develop speaker correction separately from room correction. Speakers are typically measured offline in some type of near anechoic environment (such as outside). In live settings room corrections are usually focused on time alignment from multiple sources (i.e. aligning arrival from the fill speaker with arrival from the center cluster or line array.

 

Room issues are managed by speaker selection and focus by selecting the most appropriate coverage for a given space requirement. The idea is to keep the sound off the walls and on the audience. Minor EQ can be done but I mostly focus on using EQ to avoid bass build up caused by having multiple directional speakers in a given area all of which are omnidirectional at lower frequencies.

 

Lastly for high end active speakers such as Meyer, d&B, ADR Audio and others, the speaker comes with built in DSP compensation for driver and crossover alignment. It is very hard to improve on what these manufacturers provide.

 

Just my perspective which is probably different than the typical live rock show provider.

 

Steve

 

[Nyal Mellor]

Hi Mark

 

Hats off to you for taking on this challenge. I have been thinking about doing digital XO, driver and room correction on a computer for a while since it is a logical extension of the music server concept. However the actual complexity of implementation has always put me off. Even something you would think would be simple like getting audio routed between different applications has been a challenge.

 

The MIO console is not designed as a XO, EQ and driver/speaker correction utility, but there are other options.

 

With your aim in mind I wonder whether you should also look into a few of these (you may have already) - note all are PC only:

 

- Thuneau (www.thuneau.com)

- Acourate (www.acourate.com)

- Audiolense (www.juicehifi.com)

 

Does MIO console support AU or VST plugins? If not then you are definitely limiting your options to those DSP plugins that Metric Halo provides.

 

Other vendors produce linear phase EQ products for the OSX platform and for the PC there are even more (Flux Epure, DDMF LP10, AP EQ, Refined Audiometrics Labs). You will probably also need to look into FIR based convolvers (e.g. Voxengo Pristinespace) for room correction.

 

You could then use the Metric Halo for its multichannel DAC capability and use other designed for purpose software products.

 

Note: I have no commercial affiliations to any of the above companies.