What am I missing from this test?

[soundrise]

I think I captured the difference in sound from speakers placed directly on your desk compared to decoupled speakers on stands.

You can clearly hear the difference because speakers on desk are actually LOUDER, but that's the desk vibrations adding to the volume and the bass.

How can I improve this test? Is this an accurate reading of the decoupling effect?

 

[LTig]

I think I captured the difference in sound from speakers placed directly on your desk compared to decoupled speakers on stands.

You can clearly hear the difference because speakers on desk are actually LOUDER, but that's the desk vibrations adding to the volume and the bass.

How can I improve this test? Is this an accurate reading of the decoupling effect?

I didn't listen to the comparison since the differences almost exclusively result from the different positions and the resulting reflections on the desktop. IME placing a desktop speakers such that its tweeters is at ear height sounds much better than placing it on the desk since the reflections on the desk won't reach your ears.

 

[Weeb Labs]

I didn't listen to the comparison since the differences almost exclusively result from the different positions and the resulting reflections on the desktop. IME placing a desktop speakers such that its tweeters is at ear height sounds much better than placing it on the desk since the reflections on the desk won't reach your ears.

An actual measurement would have been far more informative than placing a phone on a table and then effectively moving vertically off axis. As you mentioned, the differences simply result from a combination of desk reflections and the changing listening axis; "decoupling" has little to do with it.

 

[kemmler3D]

I don't think there's a single "decoupling effect" that applies to all situations, since it's going to depend on how much the cabinet resonates, whether the desk resonates, and the size of the speaker and how that affects desk reflections.

Some speaker cabinets will transmit vibrations into the desk that might end up being audible. The desk itself might have resonances, too. Then you have the different reflections from the desk depending on the height of the speaker vs. the desk.

As @LTig said I think the biggest benefit is getting the listening axis and your ears lined up. Reflections I would place second. Vibrations transmitted into the desk, HOPEFULLY third, as the cabinet shouldn't be vibrating that much on a good speaker.

 

[solderdude]

When you want to prove it is vibrations making the difference you should decouple the speaker from the desk (lifting it or using sorbothane or some foam pads)
Then replace the decoupled space with something solid of the same height.
This way the speaker remains in the same position and acoustics will be the same either. Only the coupling will be gone.
Requires someone doing that while the listener stays in position.

What you are missing is controls and you have more than 1 variable in play.

 

[LTig]

As @LTig said I think the biggest benefit is getting the listening axis and your ears lined up. Reflections I would place second.

Not in my experience. Placing the speakers on the desk and tilting it backwards so the tweeter axis points to the ears sounds much worse. Upper bass and lower mids are boomy.

 

[kemmler3D]

Not in my experience. Placing the speakers on the desk and tilting it backwards so the tweeter axis points to the ears sounds much worse. Upper bass and lower mids are boomy.

Fair point, this matches my recent experience of putting my 8030s on stands at ear height (about 30cm off the desk).

 

[AnalogSteph]

I've never had particularly good results placing speakers directly on a desk either... generally it's a phasey mess. Turn on some pink noise and move around the place... dreadful. JBL 104s, being 4" coaxes and as such fairly low to the ground, were quite bad in this regard. I wonder how the IN-UNF avoids this (I mean, the tilt and relatively high XO would no doubt help, but...).