Measurement question

[Beanman]

Looking to pick up a new pair of speakers. I read the reviews and look at the measurements. It seems many times the objective and subjective reviews are rather mediocre but then with a subwoofer the preference score is high. Not sure I see the correlation here. Not looking for in depth analysis, maybe just a cliffs notes explanation if it can be done. Thanks.

 

[Purité Audio]

The sub woofer makes a good speaker albeit with limited bass extension into a full range loudspeaker, if you integrate it properly.
Keith

 

[staticV3]

It seems many times the objective and subjective reviews are rather mediocre but then with a subwoofer the preference score is high. Not sure I see the correlation here.

Poor bass extension results in lower preference score. Add a Sub->increase LF extension->improve preference score.

 

[Beanman]

Is there information out there on how the preference score is determined? Is it from the spinorama data? Just wondering how a quantitative measurement is compiled into a qualitative one.

 

[mj30250]

It's calculated via formula. Bass extension is a component of the formula, and as most speakers (particularly stand-mount) cannot out-extend a "perfect" sub, a good sub will almost always increase the score.

Flatness on-axis, listening window, and PIR are other data points from the spinorama that feed into the formula. If you're attempting to rank or otherwise categorize speakers by preference score, then I'd focus on the standalone speaker score, and not so much the "with sub" score, unless you have absolutely no plans to integrate subs.

 

[Beanman]

It's calculated via formula. Bass extension is a component of the formula, and as most speakers (particularly stand-mount) cannot out-extend a "perfect" sub, a good sub will almost always increase the score.

Flatness on-axis, listening window, and PIR are other data points from the spinorama that feed into the formula. If you're attempting to rank or otherwise categorize speakers by preference score, then I'd focus on the standalone speaker score, and not so much the "with sub" score, unless you have absolutely no plans to integrate subs.

Thank you. That helps. As I haven't been through the reviews completely. Are there any examples of speakers with so-so objective measurements but high preference score?

 

[HarmonicTHD]

Thank you. That helps. As I haven't been through the reviews completely. Are there any examples of speakers with so-so objective measurements but high preference score?

Most likely not as that would be somewhat of a contradiction.

As said above the preference score is based on a formula which tries to determine how close the objective measurements are to the actual preference curve.

This preference curve is based on the work of Floyd Toole and Sean Olive who over some 20 odd years conducted many many hundreds of controlled listener studies with the objective to determine what speaker the majority of the test listeners would prefer. (Search the forum for papers from Sean Olive or for more in depth reading the book “Sound Reproduction” from Floyd Toole). This curve is sometimes also called the Harman preference curve and it is the result of an anechoic flat speaker playing in a room. It exhibits a somewhat elevated bass and a slight roll off to 20kHz.

Now, as by these studies, bass was found to contribute about a whopping 30% to a preferred speaker, therefore the addition of subwoofers drives up the preference score strongly for an otherwise bass limited speaker.

Hope that helped and wasn’t too long.

 

[Beanman]

Most likely not as that would be somewhat of a contradiction.

As said above the preference score is based on a formula which tries to determine how close the objective measurements are to the actual preference curve.

This preference curve is based on the work of Floyd Toole and Sean Olive who over some 20 odd years conducted many many hundreds of controlled listener studies with the objective to determine what speaker the majority of the test listeners would prefer. (Search the forum for papers from Sean Olive or for more in depth reading the book “Sound Reproduction” from Floyd Toole). This curve is sometimes also called the Harman preference curve and it is the result of an anechoic flat speaker playing in a room. It exhibits a somewhat elevated bass and a slight roll off to 20kHz.

Now, as by these studies, bass was found to contribute about a whopping 30% to a preferred speaker, therefore the addition of subwoofers drives up the preference score strongly for an otherwise bass limited speaker.

Hope that helped and wasn’t too long.

That was gunna be my next question if it was based on the Harman curve

 

[HarmonicTHD]

That was gunna be my next question if it was based on the Harman curve

There you go.

F Toole and S Floyd later worked for Harman where they refined their findings and conducted more listener studies. That’s where the name comes from.

 

[Beanman]

There you go.

F Toole and S Floyd later worked for Harman where they refined their findings and conducted more listener studies. That’s where the name comes from.

I have watched Toole online. Very interesting stuff.