"Sense of depth" in loudspeaker soundstage ?

[Centvrion]

Everething has been said, IMO. My experience:
ultra-flat and extended widerange drivers in tapered-transmission line [speakers walls are 27mm thick]
class-d power amplifier [hifimediy T4]
semi-floating 10mm plywood panels as front wall with some absorbing element in the room [carpet, big shelf on back wall...]

These elements give me a carved, engraved, deep and profound, fisheye like sound stage.

 

[bzfcocon]

I'm getting back with some findings that are intriguiging to me.

In the attempt to fix the backwards radiation hole, I experimented also with absorbing the higher frequencies radiating to the back. It turned out that this has a significant effect on depth perception (again, I am talking about the depth that is present in recordings):
- with the more or less full range rear radiation - including the full upper range >5kHz - the depth perception is rather weak, practically no matter how well balanced is the response
- partially absorbing the upper range (with some damping material) results in a signifcantly improved perception of the original depth cues in the recording !


The above I found true in the context of dipole speakers relatively close(~1m) to a reflective, flat rear wall. A possible explanation would be that the reflection increases the apparent size of sources and depth sensations are also related to the fact that distant sources typically appear as very small in size, thus enlarging them "artificially" by reflection on the front wall is detrimental. It can be also related to pinna cues, given the effect of high frequencies. You would probably avoid the issue at best by having a larger distance to rear wall. Or maybe- and hopefully - by using diffusers on the back wall.


The test was a track on the Chesky Ultimate Demo Disk that is specifically recorded with a trumpet at some significant distance from the microfone (10 feet)

 

[mr_push_pull]

I think it all depends on room acoustics. why else would some manufacturers fit tweeters on the rear of the speakers it it were so bad?

 

[Face]

I think it all depends on room acoustics. why else would some manufacturers fit tweeters on the rear of the speakers it it were so bad?

All, no. Majority, yes.

 

[DBMandrake]

I'm getting back with some findings that are intriguiging to me.

In the attempt to fix the backwards radiation hole, I experimented also with absorbing the higher frequencies radiating to the back. It turned out that this has a significant effect on depth perception (again, I am talking about the depth that is present in recordings):
- with the more or less full range rear radiation - including the full upper range >5kHz - the depth perception is rather weak, practically no matter how well balanced is the response
- partially absorbing the upper range (with some damping material) results in a signifcantly improved perception of the original depth cues in the recording !

That's consistent with what I observe. Generally I've found reflections from the front wall (behind the speaker) are only ever detrimental to depth perception, and indeed imaging in general. Whether those reflections are reduced by absorption or by speaker directivity doesn't really matter, the improvement is similar.

If I had to guess I would say its a combination of comb filtering related notches forming in the response at low midrange frequencies (along the lines of my previous response to the thread) as well as the fact that you have a time delayed "image" of the signal which is arriving after the main signal which somehow confuses our sense of depth perception.

Another factor is tonal balance - too much high end treble will tend to cause a "light and airy" presentation which some people like, but which can sometimes lack in sense of depth. If you're applying absorption behind the speaker which will most likely be attenuating treble more than low frequencies, you will be altering that delicate balance between low and high frequencies. Even as little as a 1dB change in the balance between low midrange and high treble from your absorption could be enough to go from a slightly thin and airy sound to sounding more full bodied with greater depth.

Any time you're making changes which also inadvertently change the overall frequency balance, you need to bear in mind that some of the difference you hear may be due simply to the shift in tonal balance. (Sometimes making an opposite "correction" with the right EQ will help you identify whether its the tonal balance you're noticing rather than some other effect, or a combination of the two)

 

[bzfcocon]

I think it all depends on room acoustics. why else would some manufacturers fit tweeters on the rear of the speakers it it were so bad?

A dipole is different than a box with a fitted rear tweeter. Most dipoles would give you backwards radiation to about 2-3 kHz without a rear or dipole tweeter, while a box will not (it would probably drop off long before 1kHz or so).

In both cases, a rear tweeter (or dipole tweeter in this case) helps restoring the overall power response to some extent - which is a good thing per in the range to about 5kHz and you can't have that without a rear tweeter. However, power response is one thing and a strong primary reflection upon the front wall is another: it looks like the latter is disturbing if it contains high frequency energy. This is intriguing to me because others, like most notably Linkwitz, seem to prefer full range on the back.

While it depends on room acoustics, a typical setup does include a front wall behind the speakers and thus a reflection.

 

[DBMandrake]

However, power response is one thing and a strong primary reflection upon the front wall is another: it looks like the latter is disturbing if it contains high frequency energy. This is intriguing to me because others, like most notably Linkwitz, seem to prefer full range on the back.

You'll notice though that Linkwitz has a rather large listening room with a lot of (natural) diffusion, and has his speakers a LONG way from the walls. Nowhere near 1 metre from the front wall as in your case. (My speakers are even closer, about 0.7m, of necessity)

I think you'll find by the time you get down to 1 metre or less between speaker and front wall that high levels of rear radiation start to become detrimental to imaging.

The general placement advice for dipoles seems to be that they need to be further from the front wall than monopoles, this is partly because the bass output of a dipole reduces as they get close to the front wall (rather than increasing like a monopole would) but I think the strong reflection from the wall at higher frequencies is a large part of the issue too.