Speaker/Room calibration

[NorthStar]

I certainly can.

Just looked at the owners manual and they said it needs to be centered on the tweeter at 37"!

Bruce, I was very surprised at how high your ears are from the floor.
Normally people are between 30" to 42" (me I'm at 33").

I'd say; 42" (1/3rd of room's height) should be your max. And if you can, because of your speakers,
aim for 36", or 38.5". These last two are perfect for you in your room with your speakers.

 

[rbbert]

Your ear height should definitely be level with the speaker's tweeter height. For most speakers (with stands if needed) designers aim for 36" - 40".

The positions most of us are suggesting are based on theoretically minimizing reinforcement of room bass nodes, but of course actual in-room bass response may not agree with that predicted by those models.

IIRC, the Sumiko method is aimed at best treble response and imaging without paying that much attention to the bass, but I may not be remembering exactly.

 

[NorthStar]

Some speakers with a tweeter and midrange drivers would usually sound best balanced between these two drivers. Or even from the midrange driver.

Also, other speakers can have a tilt upward or downward to suit your listening height position.

So at the end the tweeter itself as a guide to your ear's height is just that, a guide.
And we're not talking about electrostats or horns or other special designed speakers.

In Bruce's case, the manufacturer for his particular loudspeakers recommend 36",
then he should be near that. And from my calculations, that is why I suggested 36", which is in perfect accordance to the manufacturer. Or 38.5", which is close too.
By the way, these two distances were calculated from his room's height.
And when I got 36", it just happened to be dead on; a lucky circumstance in Bruce's case.

Furthermore, depending of your sensitivity to high frequencies, plus the type of musical material you play, the amount of toe-in, sideways and up/down tilting or from the speakers themselves or from your ears, can vary accordingly. And for your best auditive comfort with the best balance in your room on sound propagation.

* I used 1/3, 1/5, 1,7, 1/9, 1/11, 1/13 and so on, and with their multiples for my calculations.
And that goes for all three dimensional distances (length, width and height of the room).

 

[rbbert]

Well, perhaps I should have been more specific about ear height; on tweeter axis in almost all cases. And that's only height (vertical axis) I refer to, not horizontal.

I'm pretty sure those fractions are for effects of room bass nodes, with little effect on other parameters of sound.

 

[jpv]

Bruce,
I see a lot of suggestions for speaker positions. Try them and post them here. I'm looking to see if we can fill in the part between 80 to 200.
be sure to link to your MDAT. Also do the sweep to 1K.
John

 

[treitz3]

I'll have to wait until I have a "strong" buddy. Speakers weigh ~530lbs ea.!

You say that like they are heavy or something...

A towel slid underneath a heavy speaker usually works but in your case, you might need a moving blanket...but I would still have doubts of being able to easily move them.

 

[microstrip]

Your ear height should definitely be level with the speaker's tweeter height. (...)

I hope you are not addressing the Dynaudio Consequence ...

 

[RBFC]

No, you listen to those after fainting when you get the bill.

Lee

 

[NorthStar]

That tweeter seems to be approximatively 12 inches from the floor!

See, the tweeter axis means nothing; it is a wide guide. :b

 

[rbbert]

So Bruce have you decided to leave the speakers and setttings where they are for awhile, and see how you like the sound?

 

[Mitchco]

Hey Bruce,

Looks good! Have you ever considered Digital Room Correction? I use Audiolense myself, but there are several others. Given the 64bit processing paths, I have never heard any artifacts generated by the digital filters.

Here is an example of my Bob Crite's Cornscalas at the listening position. Purple is the uncorrected and the blue curve is the corrected. 1/6 octave smothing.

For translation, I find the B&K house curve http://www.bksv.com/doc/17-197.pdf to provide the most natural tonal balance. Meaning -0.5Hz dB at 200Hz, -3dB @ 2KHz and -6 dB @ 20Khz. See Fig 5 in the doc.

There seems to be correlation to this curve as well as the folks ar Harmon found the same thing: https://docs.google.com/file/d/0B97...yLWEzZTAtMGJiODQ1ZTUxMGQ4/edit?hl=en_US&pli=1 See slide 24.

However, you can dial in any curve (i.e. flat) you want, regen the filters and be listening in under a minute.

Just for fun, you could give it a try and see what you think. There is a try before you buy aspect to these softwares. As you can see from my results, I will never go back

Happy listening!

Mitch

 

[NorthStar]

That blue line is pretty sweet indeed.

Mitch, how does it sound, to your own set of ears?

 

[Mitchco]

That blue line is pretty sweet indeed.

Mitch, how does it sound, to your own set of ears?

Hi NorthStar, thanks. I tried many curves, including flat, X curve, and others and always come back to the B&K target as it seems to have two properties that suit my ears (and seemingly others that were referenced in the B&K and Harmon articles). One is the tonal balance or timbre; sounds not too dark, not too bright, but just right.

The other aspect that surprised me was the depth of field of the soundstage. If my target curve was more towards flat frequency response (relative to the B&K curve) that pushed the depth of field too far forward. If it was rolled off more than the B&K target, the depth of soundstage was too far back. But the B&K curve, like the frequency response, is just right too.

The closest I can say it sounds like a good set of headphones with mimimal room resonances - very smooth response. Relative to headphones and the B&K curve, the frequency response at the listening position is +-3dB from 20Hz to 20KHz.

Back when I was a recording/mixing engineer, I worked as a house engineer in a few studios in Western Canada. I was lucky enough to have worked with Chips Davis and his LEDE rooms. In this article you can see some pictures of those rooms and the amount of acoustic treatments we used:
http://www.computeraudiophile.com/blogs/Hear-music-way-it-was-intended-be-reproduced-part-5

I have nothing against acoustic room treatment other than it is expensive to get the level of performance I can get with digital room correction. And this is not the old school correction where you are locked into one place. The whole room benefits. With the computing power we have today and the software sophistication of these DRC programs, it is amazing to me how good it sounds.

While I am just showing frequency response here, there is also time alignment available with these DRC software, which is another discussion altogther.

Btw, that is one super nice room that Bruce has there. Fantastic! I hope to see some ETC and waterfalls with REW.

Cheers, Mitch

 

[Bruce B]

So Bruce have you decided to leave the speakers and setttings where they are for awhile, and see how you like the sound?

Been pretty busy lately. Was going to make a graph overlay of my speakers at 1" increments, starting at where they are now and moving them back towards the front wall. I've settled on the side to side position.

After I choose the flatest response, I will then start making adjustments on the back of the speaker. I will then try analog and digital EQ at the very end.

 

[NorthStar]

Hi NorthStar, thanks. I tried many curves, including flat, X curve, and others and always come back to the B&K target as it seems to have two properties that suit my ears (and seemingly others that were referenced in the B&K and Harmon articles). One is the tonal balance or timbre; sounds not too dark, not too bright, but just right.

The other aspect that surprised me was the depth of field of the soundstage. If my target curve was more towards flat frequency response (relative to the B&K curve) that pushed the depth of field too far forward. If it was rolled off more than the B&K target, the depth of soundstage was too far back. But the B&K curve, like the frequency response, is just right too.

The closest I can say it sounds like a good set of headphones with mimimal room resonances - very smooth response. Relative to headphones and the B&K curve, the frequency response at the listening position is +-3dB from 20Hz to 20KHz.

Back when I was a recording/mixing engineer, I worked as a house engineer in a few studios in Western Canada. I was lucky enough to have worked with Chips Davis and his LEDE rooms. In this article you can see some pictures of those rooms and the amount of acoustic treatments we used:
http://www.computeraudiophile.com/blogs/Hear-music-way-it-was-intended-be-reproduced-part-5

I have nothing against acoustic room treatment other than it is expensive to get the level of performance I can get with digital room correction. And this is not the old school correction where you are locked into one place. The whole room benefits. With the computing power we have today and the software sophistication of these DRC programs, it is amazing to me how good it sounds.

While I am just showing frequency response here, there is also time alignment available with these DRC software, which is another discussion altogther.

Btw, that is one super nice room that Bruce has there. Fantastic! I hope to see some ETC and waterfalls with REW.

Cheers, Mitch

Very nice reply Mitch, thanks!

* From your link, in one of the pictures; the Canadian Mirage M-1 loudspeaker. :b
I used to own a pair, way back then (just too big for my room, I returned them to my dealer).

Cheers,
Bob
_____________

P.S. Bruce, take all the time you need.

 

[Mitchco]

Been pretty busy lately. Was going to make a graph overlay of my speakers at 1" increments, starting at where they are now and moving them back towards the front wall. I've settled on the side to side position.

After I choose the flatest response, I will then start making adjustments on the back of the speaker. I will then try analog and digital EQ at the very end.

Bruce, agree with your approach 100%. I did similar, sans the analog eq, and really worked out well for me.

I highly recommend a digital laser measurer. I use this one: http://www.amazon.com/Bosch-DLR130K-Digital-Distance-Measurer/dp/B001U89QBU

A 20 KHz wavelength is just over 1/2", http://www.sengpielaudio.com/calculator-wavelength.htm I found setting up my equilateral triangle to be a critical factor in the reproduction of the soundstage. I got between 1/16" to 1/4" tolerance in mine and it made a huge difference in producing a rock solid phantom center image and 3D imaging. I was surprised at the level of improvement as the tolerance was tightened.

Looking forward to seeing your measurements at each stage. I am sure they will be educational and insightful!

Cheers, Mitch

 

[rbbert]

toe-in is so speaker and room dependent, though...

 

[NorthStar]

toe-in is so speaker and room dependent, though...

Recordings dependent too. But impractical.
& particularly with Bruce's speakers (500+ pounds each).

* Mitch brought one of the most important point IMO:
the triangulation formed between the main listener and the two main speakers.

** And Bruce is on the right track:
get the most linear curve first (from positioning), and then use EQ.

_______________________

Bruce's room is so fantastic, and what he's doing with all of us in this thread here is extremely generous. It is a course on the right approaches into Best sound by intelligent room treatments, measurements, smart positioning based on calculated acoustics from room's dimensions, and finally by using judicious room equalization with the right tools and programs (later on).
An excellent thread to everyone.

 

[rbbert]

...
* Mitch brought one of the most important point IMO:
the triangulation formed between the main listener and the two main speakers...

It was this point to which I was responding with my comment about toe-in. Other than in some near-field listening setups, I have rarely heard a system which sounded or imaged best in an equilateral triangle arrangement. I'm sure they exist, but I suspect they are rare, and others with far more experience setting up speakers in a variety of listening environments and getting them to sound very good rarely end up with an equilateral triangle, except in those near-field setups.

 

[NorthStar]

It was this point to which I was responding with my comment about toe-in. Other than in some near-field listening setups, I have rarely heard a system which sounded or imaged best in an equilateral triangle arrangement. I'm sure they exist, but I suspect they are rare, and others with far more experience setting up speakers in a variety of listening environments and getting them to sound very good rarely end up with an equilateral triangle, except in those near-field setups.

Yes I knew that this was what you were referring to.

And regarding the triangulation; I never mentioned "equilateral".
In some cases it works fantastically, and in others it don't.
...So many variables: speaker's sound dispersion, room's acoustics, music material, etc., etc., etc.
It's all about experimentation, measurements, and personal preference.
And also about compromises, because each recording is different.

But for most folks it's about what suits them decor wise;
so it's a big sound quality limitation.

But here, with Bruce, there is no limit in sound quality. :b