Speaker shootout question -- do you position the same or differently, depending?


If you're comparing two speakers at home, do you position each the same or do you position each as (roughly) optimal for that speaker in your room?

I'm comparing a tower and a bookshelf now, and their design is different. It would seem that the best way to compare would be to figure out what is optimal for each and then compare them in (likely) different positions. 

What kind of process do you use for comparing two differently designed speakers?

128x128hilde45

@snilf @snilf Good suggestions. I have a backup Adcom amp with A & B speaker sets. I could quick switch with those. Of course, the speakers would be coming through the Adcom rather than my tube amps, and couldn’t use my better speaker cables, but at least both speakers would have the same cables and the same power amp.

@jjss49 @djones51 @jjss49 @djones51 Thanks for saying so. The back and forth between listening and measuring strikes some as over the top but this is exactly how many hobbyists do it, and of course speaker companies, too (such as Harman). It’s not just measuring but an iterative process. Paul McGowan talks about this all the time — design, measure, listen, design, measure, listen.

These topics seem so simple to some (who have likely just done things the same way forever) but once the suggestions start coming in, it’s clear that the complexity of both acoustics and how people decide to test their own perception makes the experimental procedures involved quite varied, indeed.


@bjesien @bjesien

"In my experience different speakers with behave quite differently depending on placement to boundaries and distance to listing position (reflection points, nodes, etc.) The position of the drivers in the cabinet will give a wildly different presentation, as will the width and depth of the cabinets."

This is so true. In fact, just playing with the height of my seating position the other day, I realized how much different the new speakers were with a 2" lower seating position -- a change which did not affect the older pair at all.

Also +1 about the comment regarding the need for a "symmetrical space." While my room IS a rectangle with (somewhat different kinds of bookshelves on both sides), one is easily fooled into thinking that it is symmetrical. This is where REW is so helpful. Do a sweep with just left and just right and the differences are *immediately* apparent.

@astolfor @astolfor
"Your ear will take you 60-70% there depending how difficult your room is, but to do it right you need to use some application like REW and decent calibrated microphones to get past that 70%….I know that what I am going to say is subjective, but using my ears and REW, the transformation of the sound was/is nothing short incredible."

I totally agree — after positioning things in places they "sound good," I measure and adjust and then listen again. What the anti-measure people don’t realize is that when there are big dips in a curve, they could be missing even an octave of notes but because they brain tends to create a "gestalt" they don’t notice anything is missing, often. But once those dips are removed, one can perceive additional notes — manifesting, often, as voices or instruments playing those parts — and realize they were missing important parts of the music.

Here’s the curve I achieved using REW *after* listening:

 

And here is the impulse graph, above -20db, where things start to get smeared. I went from a forest of reflections to a few scattered trees, and the definition on images throughout the soundstage got startlingly firm. 

 

What speaker is that measurement of? If I had something measured like that I wouldn’t bother with it but to each their own. Is it a high efficiency for low watt amp speaker?

@djones51 That’s a measurement of an Ascend Sierra Tower with RAAL tweeter along with 2 Rythmic subs and a REL 328 sub, in well treated room. The speaker measured much differently in a different position and without subs or treatment.

Can you say what you don’t like about the graph? Here’s how it measured before subs and treatment (initial tone was 80 dB):

 

The latest red plot Betweeen 1500hz and 10,000hz is a 10db drop then a 10db rise to 20,000hz. I prefer a smooth fall of 5db from 1000hz to 20,000hz. I would get rid of the rise from 10k to 20k.

The blue first plot I would slope it from 4k to 20k by a few db or flatten it. I also don’t like that much bass boost.

Of course this is my preference not yours but it look pretty hot in the upper frequencies.

Looking at their website those speakers have a pretty flat response what to you get at about 3 feet from each speaker? 

As noted earlier in this thread, your ROOM is the most important variable in ANY sound system.

Why do large venues hire audio experts to help design the room for them?