Incorporating MiniDSP With Audyssey XT32

[Pogre]

Uummmm you don't? I still have to catch up from a couple pages back. Unless I'm not in my right brain cell I'd still go from the avr to mini to subs. I think you'll use it with Audyssey, not in leu of. Unless I'm missing it, bypassing the avr will be convoluted and not work. Must eat...

So hook up the mini normal. Go hdmi from lappy to front of avr like taking sweeps.

What I'm trying to do is, use the mini to flatten my response as much as it can, THEN run Audyssey and give it less to fix.

I need to disable all filters and eq to get a raw sub sweep and see what I'm sealing with so I can apply corrections using the mini first.

 

[Pogre]

I'm following intructions here, but I need more detail. Like, how to set it so I can generate a signal while bypassing the avr...

I'm all set up and ready to go here. I need help!

 

[panteragstk]

I'm following intructions here, but I need more detail. Like, how to set it so I can generate a signal while bypassing the avr...

I'm all set up and ready to go here. I need help!

My bad, I thought you could use the minidsp as the output (I'm still going to look, I remember seeing this) but this link should get you going.

EDIT: Found this one too. It reads like you are using the mini to output the audio. This should be more clear.

 

[TheWarrior]

Copied from another thread, I think I've shared this with you before, but since the questions keep going in this thread I think some clarification is needed on why an omni mic in a room of unknown size can't 'know' what needs to be corrected:



The room is in control of bass. You can leave it to the magic of Audyssey and hope it's results are satisfactory (which your measurements, are not) or get out a tape measure and start measuring ALL of the parallel boundaries of your room.

1131(feet per sec) / measurement (in feet) = Room Mode - which will measure as a peak or null at the LP depending on it's +/- phase cycle - subwoofers are pressure sources, placement is everything

Frequency + Amplitude + Q = DSP Filter.

(center of peak or null) + (dB increase/decrease to reach relative level) + (bandwidth of resonance measured in hz) = DSP Filter

Those 3 parameters allow you to create a DSP filter to overcome the natural response of producing frequencies that are larger than the dimensions of your room. (20 hz = 56.5 feet for example)

Measurements MUST be taken as 1/24 octave so that nothing is masked. Windows and doors allow flex in boundaries which will lower the measured resonance frequency from predicted. But when you know which boundaries are influencing a frequency, the orientation of subwoofer(s) at that boundary allows you to constructively or destructively drive the standing waves.

If you're really serious and want to learn more, 'Sound Reproduction' by Floyd Toole is the textbook on the subject. He just released the 3rd edition so be mindful which one you order. This is a lifetime of research backed by double blind testing with only minimal mathematics being presented, such as what I've suggested.

 

[Pogre]

Well, here's what I came up with last night after wrestling with everything.

Both measurements were taken with Audyssey turned off, pure direct, crossover at 250hz, subs only. Dark green is after applying the filters to the mini. I see an improvement!

 

[TheWarrior]

Well, here's what I came up with last night after wrestling with everything.

View attachment 22014 View attachment 22015
Both measurements were taken with Audyssey turned off, pure direct, crossover at 250hz, subs only. Dark green is after applying the filters to the mini. I see an improvement!

1/24 octave resolution is needed to be certain nothing is missed. And as for what you did change, what filters did you apply? Why did you invert the responses? Your room is open to several others, yes? How did REW predict modal frequencies?

 

[Pogre]

Copied from another thread, I think I've shared this with you before, but since the questions keep going in this thread I think some clarification is needed on why an omni mic in a room of unknown size can't 'know' what needs to be corrected:



The room is in control of bass. You can leave it to the magic of Audyssey and hope it's results are satisfactory (which your measurements, are not) or get out a tape measure and start measuring ALL of the parallel boundaries of your room.

1131(feet per sec) / measurement (in feet) = Room Mode - which will measure as a peak or null at the LP depending on it's +/- phase cycle - subwoofers are pressure sources, placement is everything

Frequency + Amplitude + Q = DSP Filter.

(center of peak or null) + (dB increase/decrease to reach relative level) + (bandwidth of resonance measured in hz) = DSP Filter

Those 3 parameters allow you to create a DSP filter to overcome the natural response of producing frequencies that are larger than the dimensions of your room. (20 hz = 56.5 feet for example)

Measurements MUST be taken as 1/24 octave so that nothing is masked. Windows and doors allow flex in boundaries which will lower the measured resonance frequency from predicted. But when you know which boundaries are influencing a frequency, the orientation of subwoofer(s) at that boundary allows you to constructively or destructively drive the standing waves.

If you're really serious and want to learn more, 'Sound Reproduction' by Floyd Toole is the textbook on the subject. He just released the 3rd edition so be mindful which one you order. This is a lifetime of research backed by double blind testing with only minimal mathematics being presented, such as what I've suggested.

Yeah, I know about Floyd's new book. I will eventually get a copy. I'm not sure how "serious" I am. I have time on my hands and the equipment to do it with, so I'm just wanting to improve my sound if I can. I don't know if I'm ready to dig into the formulas. It took me 2 days just to learn how to apply filters to my mini! lol

 

[Pogre]

1/24 octave resolution is needed to be certain nothing is missed. And as for what you did change, what filters did you apply? Why did you invert the responses? Your room is open to several others, yes? How did REW predict modal frequencies?

I don't know. I've been asking everyone how to do this and this is what I came up with on my own. I'll more than happily take some measurements again, right now, with all of the proper settings.

Does the 2nd graph not show an improvement?

 

[Pogre]

1/24 octave resolution is needed to be certain nothing is missed. And as for what you did change, what filters did you apply? Why did you invert the responses? Your room is open to several others, yes? How did REW predict modal frequencies?

I used REW to generate the filters based on the raw sub sweep and uploaded the settings to the MiniDSP HD. By 1/24 octave, you mean the smoothing, right?

I followed these instructions.

 

[TheWarrior]

Yeah, I know about Floyd's new book. I will eventually get a copy. I'm not sure how "serious" I am. I have time on my hands and the equipment to do it with, so I'm just wanting to improve my sound if I can. I don't know if I'm ready to dig into the formulas. It took me 2 days just to learn how to apply filters to my mini! lol

Well, you're throwing minidsp in to your signal chain, so I'd say you're serious enough to NEED to know what you are doing. There isn't much math to be found, actually. It's just an enormous amount of information, if you read the entire book. This latest, 3rd edition is so well laid out that you really can flip to what you need.

I don't know. I've been asking everyone how to do this and this is what I came up with on my own. I'll more than happily take some measurements again, right now, with all of the proper settings.

Does the 2nd graph not show an improvement?

You need to type out those 3 parameters to describe what filters you/REW added to minidsp. Look under PEQ. And you would want to compare those results with your own room mode predictions, that I outlined how to do. Otherwise you are making blind changes to your sound with no reference.

As for improvement, the only way to be sure is to use a higher resolution measurement so that you know no other peaks/nulls snuck in there as a result of your changes. Floyd goes in to great detail in the 3rd edition, as well as his public AES paper, 'Measurement and Calibration of Sound Reproducing Systems' that goes in to the exact reasons why an omni mic can't possibly know what the room is doing to the sound it hears.

 

[TheWarrior]

I used REW to generate the filters based on the raw sub sweep and uploaded the settings to the MiniDSP HD. By 1/24 octave, you mean the smoothing, right?

I followed these instructions.

You want 1/24 octave measurements with no smoothing

 

[Pogre]

I followed these instructions.

Well, you're throwing minidsp in to your signal chain, so I'd say you're serious enough to NEED to know what you are doing. There isn't much math to be found, actually. It's just an enormous amount of information, if you read the entire book. This latest, 3rd edition is so well laid out that you really can flip to what you need.



You need to type out those 3 parameters to describe what filters you/REW added to minidsp. Look under PEQ. And you would want to compare those results with your own room mode predictions, that I outlined how to do. Otherwise you are making blind changes to your sound with no reference.

As for improvement, the only way to be sure is to use a higher resolution measurement so that you know no other peaks/nulls snuck in there as a result of your changes. Floyd goes in to great detail in the 3rd edition, as well as his public AES paper, 'Measurement and Calibration of Sound Reproducing Systems' that goes in to the exact reasons why an omni mic can't possibly know what the room is doing to the sound it hears.

Okay, I'll do some measurements right now with 1/24 smoothing. I have everything set up from last not still. My thinking is, if I can get some accurate measurements and apply the filters to my mini, then run Audyssey to dial it in even better.

*Edit: What settings should I have in my avr for a "raw" sub sweep? Should I be using my avr to do sweeps or is there a way to bypass it and go direct to the subs?

 

[TheWarrior]

I followed these instructions.

Okay, I'll do some measurements right now with 1/24 smoothing. I have everything set up from last not still. My thinking is, if I can get some accurate measurements and apply the filters to my mini, then run Audyssey to dial it in even better.

*Edit: What settings should I have in my avr for a "raw" sub sweep? Should I be using my avr to do sweeps or is there a way to bypass it and go direct to the subs?

REW lets you sweep whatever frequency range you want, you can type it in manually. It gets a bit more complicated with higher order resonances (modal frequency x2, x3, x4, and so on, these are reflections, sound doesn't bounce once and lose all its energy) as thats when placement really comes in to play due to peaks/dips being measured above the subwoofer crossover. You *hope* correcting the fundamental resonance will solve it, but the room (and all that it is open to) determine that.

 

[TheWarrior]

But again, without knowing all your room dimensions, and their resonant frequencies, there is only so much I could say about your measurements without knowing what the problems were, and what exactly was done to correct them.

 

[Pogre]

REW lets you sweep whatever frequency range you want, you can type it in manually. It gets a bit more complicated with higher order resonances (modal frequency x2, x3, x4, and so on, these are reflections, sound doesn't bounce once and lose all its energy) as thats when placement really comes in to play due to peaks/dips being measured above the subwoofer crossover. You *hope* correcting the fundamental resonance will solve it, but the room (and all that it is open to) determine that.

I thought I could take measurements with REW, generate a filter for my mini, apply that filter, then measure again to check my response. I have a before and after chart above that shows (to me) a very clear and visible improvement. You're not seeing that? Ami reading it wrong? It measured a lot flatter and I thought that was the goal? Smooth the response out with my mini, then run Audyssey. It should have an easier time working its magic if it has a better response to work with, right?

I keep repeating that I'm new to this and am so far, self taught on how to use REW and now the mini.

 

[Pogre]

All I want to do is, defeat any and all dsp processing being applied to my subs, get a raw, unfiltered response graph, generate the filter file for my mini in REW, then upload that to the mini. What more do I need to do than that?

*Edit: Where do I change resolution? Is that different from smoothing?

 

[Pogre]

You need to type out those 3 parameters to describe what filters you/REW added to minidsp. Look under PEQ. And you would want to compare those results with your own room mode predictions, that I outlined how to do. Otherwise you are making blind changes to your sound with no reference.

As for improvement, the only way to be sure is to use a higher resolution measurement so that you know no other peaks/nulls snuck in there as a result of your changes.

Higher resolution? Where do I find that setting in rew?

 

[William Lemmerhirt]

View attachment 22017


Higher resolution? Where do I find that setting in rew?

In the measurement page, top left. Graph tab. Iirc.

 

[William Lemmerhirt]

In the measurement page, top left. Graph tab. Iirc.

Your pic is from the minidsp.

 

[shadyJ]

All I want to do is, defeat any and all dsp processing being applied to my subs, get a raw, unfiltered response graph, generate the filter file for my mini in REW, then upload that to the mini. What more do I need to do than that?

*Edit: Where do I change resolution? Is that different from smoothing?

I haven't been keeping up with this thread very closely, but to get a raw response of the subs, why not just connect them to your computer directly (assuming your computer has analog outputs) and then measure the response?