Thoughts about single box stereo?

[pelanj]

The more I listen to that box, the more I like it. Electronic music can sound really nice with all the effects - and so does a jazz trio. Also satisfactory bass at comfortable listening levels. For the ebay price, I cannot complain.

 

[danb11]

That's pretty good going for the price.

I'm close to finishing my curved idea and was wondering how I would go about measuring the stereo coverage. I'm making it active 2.1, so can adjust the XO as I go. Is it simply pivoting the speaker over multiple measurements?

 

[pelanj]

This is a really good question - how to measure stereo coverage. I would think it is not too easy to measure - other than just moving around and judge by ear. I imagine the stereo effect to be sort of a psychoacoustic phenomenon and cannot imagine any other assessment than just individual subjective listening test over multiple locations.

 

[adason]

I am having the same dilemma. How to measure soundstage/imaging. When I finished some speakers, and wanted to compare the soundstage between 2-3 pairs, personal listening was the only option. But that makes it non quantifiable parameter, just subjective feeling. I moved from minimal baffle to no baffle for planars, and definitely heard improvement in imaging.
The only semi quantitative evaluation would be to make good recording in the listening position, ideally with dummy head (I can supply my head here), and then listen to the recordings with high quality headphones.

 

[danb11]

What is it one would need to measure? I wonder if one of those phase tests could be used. White noise in phase then out of phase. Compare around the room. That's just a hunch as I'm not scientific minded. Maybe there's a way REW could show that?
But I'm not sure soundstage is the same as demonstrating that there is a stereo field in different positions. I'm pretty sure soundstage on SSS is something that will never be as good as a stereo triangle.

 

[tmuikku]

Tried to search around and could not find measurement tests that would give complete picture of stereo performance. Linkwitz has comprehensive listening test list though, perhaps help if one would like to chart what the ears tell https://www.linkwitzlab.com/accurate stereo performance.htm

 

[danb11]

Is there a way to measure clearly whether 2 speakers are in phase/out of phase? If I could determine that in measurements, I could attempt to show whether my so-called "single stereo speaker" works in other "non-central"/"non-triangular" positioning. It won't show the actual performance stereo-soundstage wise. Thats surely always going to be subjective.
I can't honestly believe an idiot like me is the only one to wonder this ;-)

 

[mondogenerator]

Thanks, but of course I've read everything I could find here on the subject. The Nagaoka Tetsuo thread deals fairly specifically about one method of Mid/Side stereo, which I've found isn't applicable for most single box stereo needs.

I personally think there is room for more discussion than one thread that's gone largely untouched for the last 5 years.

There is a very good thread by a DIYA member that was quite prolific, I can't recall the username.

Typical.

Search Matrix stereo one box or something like that.

It was an intriguing triangular box or trapezoidal, with 2 outward facing speakers (maye a centre one too...my recall is bad)

That would be the path I'd follow, to start with, if I could find it here..

 

[adason]

He rejected that approach for he wanted to come up with something totally of his own, revolutionary.
Still waiting 🙂 not

 

[Pano]

A measurement technique for stereo images would be intriguing. I know that I have taken impulse measurements of good, bad and different systems with mics in my ears. What could be seen in impulse measurements like that? No idea, but perhaps it's a start.

 

[NareshBrd]

Fashion display dummy head shaped mic stand with two single or a stereo microphone?
Might work...

 

[Pano]

Yes, some sort of dummy head or similar shape. It might take machine learning to tease out what measurements correlate with "good" stereo.

 

[adason]

RightMark Audio does stereo separation test. Besides fr response, dynamic range, thd, im, and others.
And its free.

 

[adason]

I believe Siegfried Linkwitz (rip) used glasses with two mics tied to the frame where sound enters ear. One just has to sit still for few minutes while RightMark does sweeps and bursts and such.

 

[Pano]

RightMark Audio does stereo separation test.

Of speakers?

 

[adason]

From the signal from microphones, I suppose.
Its electrical signal like any other. RightMark compares signal out with signal in, and gives report.
Besides the value in stereo separation, you can see graph of stereo crosstalk vs frequency.
I never tried it using microphones, though.
Mic preamp is likely required, I know focusrite has amazing mic preamps.

 

[Pano]

I suppose that measuring crosstalk in a room wouldn't be too hard with a dummy head, but the amount of crosstalk will be huge. How would one interpret that?
Understanding what's going on with the "sound-field" in the room would not be a trivial task.

 

[adason]

I guess you are right, crosstalk will be huge. But the data are saved in files, and one can compare and overlay up to ten files.
So it still should be useful. For comparison. One would measure normal bookshelfs, save the data, and measure single box stereo, and overlay.
I am not going to do that right now, but it should be rather interesting.
Measuring total performance of the system.

 

[tmuikku]

Does dummy head recording sound like the real thing in headphones? In that case perhaps AI could be trained to rank the "perceived sound" by humans.

 

[Pano]

Dummy head does NOT sound like the real thing to me. But having a standard measuring technique like the KU100 head is important so that different systems could be measured equally.