Is a higher directivity speaker 'coloring' the sound?

[tmtomh]

No Sophie's choice here. I kept both. The wider directivity Philharmonic Audio HT's are beautiful for home theater, acoustic recordings and multiple listening points at a 10-12 foot distance. The added reverb is pleasant and spacious. I use MiniDSP and DiracLive with a subwoofer. It's quite good and affordable.

For my music only, at a single listening place at midfield (4ft), for bass-heavy pop/rock - I significantly prefer the Genelec system including a Genelec subwoofer. The soundstage is smaller but so compellingly clearer, especially in the center vocals. If I could only keep one, it would be the Genelec system by a mile. It's also more than twice the cost, and I think significantly better.

This is so interesting. I have only one pair, Genelec 8351b’s, but I feel similarly to you.

Specifically, I find that I prefer a wider-dispersion, more “enveloping” speaker… for a little while. The problem is that after a few hours or even just a couple of tracks, I get tired of it and feel a nagging desire for the precision, clarity, and impact you describe your 8361s providing - and this is the same experience I have with my 8351s. (Yes, yes, I know someone will inevitably jump in saying, “I have both, no trade off necessary! That’s BS - you can certainly have good performance in both respects, but with regular stereo there’s still always some degree of tradeoff, for the reasons already discussed at length in this thread.)

In the distant past I’ve scratched this “envelopment” itch with that method where you wire up two rear speakers out of phase with the fronts to make a kind of stereo surround; and at another time with a Carver Sonic Holography unit my father gave me. But in both cases I took them out of my system after a few weeks because the novelty, while really fun, wore off pretty quickly. Even with my Genelecs, for a while I had them toed out considerably and slightly tilted upwards, and the result was a perceptually larger, slightly more diffuse soundstage. But after a while it just sounded wrong, as in it just started to sound unsatisfying. I undid the tilt and reduced the toe-out to the 10-15 degrees I’ve preferred with pretty much every speaker I’ve ever owned, and everything just snapped into place better.

Finally, I noted earlier in this thread that I like to listen in the dark, often with my eyes closed. I’ve noticed that when I do that, and I point my finger at what I’m hearing as the L or R edge of the soundstage, when I open my eyes my finger is usually pointing to a location about a foot (give or take) outside the speaker. But when I listen to the same recording with my eyes open, the illusion usually collapses and the soundstage edge moves slightly inward to the outside edge of the speaker. It’s an interesting psychoacoustic/psychovisual phenomenon. It seems that with my eyes closed my brain interprets the reverb/ambience of a hard-panned sound as distance, but with my eyes open, it interprets it as what it really is - a hard-panned sound with ambience/reverb applied, coming out of that speaker’s drivers.

 

[Heinrich]

You got a few things wrong about hearing ...

You also wrote: "Griesinger has shown that once you get to the subwoofer range, bass decorrelation is the most relevant measured quality, primarily phase difference over time between ears. Exactly what kind of decorrelation is not clear and is an open question."

In order to show-piece the benefit of new and more costly gear for more than ever valuable future recordings (of the good old "classics") --an A/B microphone setup is used - the "stereo" microphones left/right meters apart. What kind of a decorrelation is it? Maybe it is the Dunning-Kruger on my side, but can't get anything out of, nor get anything into it.

If these guys are into something, I would expect the muddy waters clarified and open questions closed. At least a plan to do so. You say, "primarily phase difference over time between ears". Is it a starting point? I would say, that is the signature of a large room, travelling, not yet settled reflected waves in contrast to a finally standing wave in a small room. This can be synthezised. Ahead we go.

sidenote: the industry dug its own grave with ultimately increasing cost for humble home stereo sets. Nothing was ever "good enough", except for the full budget (and beyond). People got tired of that game. Now they listen mono via bluetooth boxes. We don't need more.

 

[Curvature]

If these guys are into something, I would expect the muddy waters clarified and open questions closed. At least a plan to do so. You say, "primarily phase difference over time between ears". Is it a starting point? I would say, that is the signature of a large room, travelling, not yet settled reflected waves in contrast to a finally standing wave in a small room. This can be synthezised. Ahead we go.

Mono is fully correlated; there are a lot of simple manipulations to create deviation from mono and cause decorrelation. But which way is perceptually correct and readily causes envelopment? What is the method (or theory, if you will) for manipulating the sense of envelopment? It's certainly not as easy as wide vs. narrow radiation at higher frequencies.

Griesinger shows that it has something to do with modulation of the signal between the ears. His demonstration is an example of how to maximize decorrelation in room using a specific setup. But what's optimal? I don't know and I haven't seen research on the matter.

 

[kharan]

@tmtomh @Mort

Out of curiosity have either of you tried your speakers placed horizontally instead of vertically? You can get near constant directivity in the horizontal plane starting from 200 and 300 Hz instead of having a discontinuity in the middle of the medium range.

 

[krabapple]

This is so interesting. I have only one pair, Genelec 8351b’s, but I feel similarly to you.

Specifically, I find that I prefer a wider-dispersion, more “enveloping” speaker… for a little while. The problem is that after a few hours or even just a couple of tracks, I get tired of it and feel a nagging desire for the precision, clarity, and impact you describe your 8361s providing - and this is the same experience I have with my 8351s. (Yes, yes, I know someone will inevitably jump in saying, “I have both, no trade off necessary! That’s BS - you can certainly have good performance in both respects, but with regular stereo there’s still always some degree of tradeoff, for the reasons already discussed at length in this thread.)

If I were you I'd buy two more speakers as surrounds, and use Dolby Surround or some other upmixer when you feel like you want more envelopment. On/off is as simple as hitting a button on the remote, on most AVRs.

 

[tmtomh]

@tmtomh @Mort

Out of curiosity have either of you tried your speakers placed horizontally instead of vertically? You can get near constant directivity in the horizontal plane starting from 200 and 300 Hz instead of having a discontinuity in the middle of the medium range.

The Genelecs are coaxials.

 

[kharan]

The Genelecs are coaxials.

I know. That doesn't change anything with regards to my question. Vertical and horizontal directivity are different once you cross over to the woofers.

 

[tmtomh]

I know. That doesn't change anything with regards to my question. Vertical and horizontal directivity are different once you cross over to the woofers.

I hear you on that, but the Genelecs' symmetrical two-woofer setup seems to minimize the directivity differences between horizontal and vertical orientations, at least as far as I can tell.

 

[tmtomh]

If I were you I'd buy two more speakers as surrounds, and use Dolby Surround or some other upmixer when you feel like you want more envelopment. On/off is as simple as hitting a button on the remote, on most AVRs.

A perfectly reasonable suggestion, and essentially a more complex - and much more properly engineered - version of the Hafler circuit I used for a little while when I was younger (the two rear speakers wired out of phase with the mains scheme, or whatever it is, that I described in my prior comment).

For me personally, though, it's not worth the extra system complexity, the physical bulk of the rears (which would have to be on stands in my room), and so on - that kind of thing becomes like a special effect to me, and I'm 99% sure that once I'd cycled through most of my favorite music with it for a couple of weeks, I'd turn it off and never turn it back on.

Also, just to avoid giving everyone the wrong impression, my current setup does not produce a "small" soundstage. It's not huge, and it's not enveloping in the sense that I would feel it surrounding me, but it's quite wide and has a good deal of illusory depth. To @MattHooper 's point (I forget if here or in another thread), I do periodically get the illusion of reality in various ways: a spookily realistic lower-midrange texture on a dry, center-panned vocal; the startling transient pluck of a guitar string; that depth illusion I mentioned before when I listen to hard-panned sounds with my eyes closed; and in general with many recordings, the feeling that the group is performing right in front of me.

 

[pma]

the Genelecs' symmetrical two-woofer setup

D’Appolito configuration?

 

[Suono]

I listen like this, let's please judge this polar diagram.

 

[tmtomh]

D’Appolito configuration?

It would appear so.

Understanding the Genelec Ones design

There has been much discussion here (and elsewhere!) about the Genelec ones. They are extraordinary - indeed, I've just ordered a pair! Looking at them I can understand the logic and benefits for the coaxial mid and tweeter but what is the rationale for tucking the woofers behind the front...

www.audiosciencereview.com

 

[slaweks]

Agreed, but in my experience this is an answer that is in itself in need of an answer. Much of what I prefer to listen to does not exist in native multichannel, particularly in lossless formats that are readily accessible.

I've played around with upmixers several times, and each time I've found the results to be...mixed. Perhaps I'm just doing it wrong.

Try BACCH

 

[kharan]

The Genelecs are coaxials.

I know. That doesn't change anything. Vertical and horizontal directivity is different once you pass over to the woofers.

I hear you on that, but the Genelecs' symmetrical two-woofer setup seems to minimize the directivity differences between horizontal and vertical orientations, at least as far as I can tell.

You're not showing me anything new, rather I would like to point your attention towards how the directivity is mostly constant from 600 Hz and upwards on the horizontal plane, whereas the directivity is mostly constant from 200 Hz and upwards on the vertical plane. It's quite possible that the ear is not sensitive to that directivity variation in that 200-600 Hz range, but I think it would be something interesting to test.

 

[sfogg]

A perfectly reasonable suggestion, and essentially a more complex - and much more properly engineered - version of the Hafler circuit I used for a little while when I was younger (the two rear speakers wired out of phase with the mains scheme, or whatever it is, that I described in my prior comment).

For me personally, though, it's not worth the extra system complexity, the physical bulk of the rears (which would have to be on stands in my room), and so on - that kind of thing becomes like a special effect to me, and I'm 99% sure that once I'd cycled through most of my favorite music with it for a couple of weeks, I'd turn it off and never turn it back on.

Also, just to avoid giving everyone the wrong impression, my current setup does not produce a "small" soundstage. It's not huge, and it's not enveloping in the sense that I would feel it surrounding me. but it's quite wide and has a good deal of illusory depth. To @MattHooper 's point (I forget if here or in another thread), I do periodically get the illusion of reality in various ways: a spookily realistic lower-midrange texture on a dry, center-panned vocal; the startling transient plug of a guitar string; that depth illusion I mentioned before when I listen to hard-panned sounds with my eyes closed; and in general with many recordings, the feeling that the group is performing right in front of me.

Good surround processing/upmixing (for music) is *very* different sounding than the Hafler matrix. The Haflex matrix has around 3dB channel seperation, is mono and has no delay to avoid the Haas precedence effect.

Good music processing will be 2,4 or even more channels of rear/surrounds/heights and the possibility of a center channel as well.

Can't argue against the extra complexity of course.

 

[krabapple]

A perfectly reasonable suggestion, and essentially a more complex - and much more properly engineered - version of the Hafler circuit I used for a little while when I was younger (the two rear speakers wired out of phase with the mains scheme, or whatever it is, that I described in my prior comment).

Upmixing has come a long way since then

For me personally, though, it's not worth the extra system complexity, the physical bulk of the rears (which would have to be on stands in my room), and so on - that kind of thing becomes like a special effect to me, and I'm 99% sure that once I'd cycled through most of my favorite music with it for a couple of weeks, I'd turn it off and never turn it back on.

The added complexity may consists of:
1) wiring 2 speakers behind or to the side of you, on stands or wall mount. (They needn't be the same as your fronts, though that helps)
2) telling your head unit that you have four speakers now --> one time, set and forget
3) pushing the surround button, or not

 

[MattHooper]

Piggybacking on this…

Specifically, I find that I prefer a wider-dispersion, more “enveloping” speaker… for a little while. The problem is that after a few hours or even just a couple of tracks, I get tired of it and feel a nagging desire for the precision, clarity, and impact

I get that.

For me, I go back-and-forth on this to a certain degree simply in how I arrange my speakers and reflections. Sometimes when I arrange them in a way to get that tighter focus but smaller sound stage, I really like it for quite a while, but… opposite from you… I start to miss the spaciousness of my usual set up.

As I mentioned before, it had always been something of a trade-off until I discovered using a curved diffuser in my room between behind the speakers. When I would sit very close to my speakers placed fairly wide, I could lose some of that density and focus, but once I employed the diffuser its like all the images snapped into place and became very solid and dense and palpable. So it became as close to the best of both worlds that I personally have heard. (An audio buddy of mine remarked on the same thing once I had that set up - he was shaking his head saying “ how do you do this? The sound stage is enormous sounds are coming from everywhere yet they are so focussed and “right there” sounding).

I don’t know how close this possibly comes to a Genelec set up. I seem to remember using Genelec speakers in some previous editing set ups, but I was never evaluating them the way I normally do for two channel. And I’ve only heard newer Genelec in a couple stores and not under the greatest conditions.

Finally, I noted earlier in this thread that I like to listen in the dark, often with my eyes closed. I’ve noticed that when I do that, and I point my finger at what I’m hearing as the L or R edge of the soundstage, when I open my eyes my finger is usually pointing to a location about a foot (give or take) outside the speaker. But when I listen to the same recording with me eyes open, the illusion usually collapses and the soundstage edge moves slightly inward to the outside edge of the speaker. It’s an interesting psychoacoustic/psychovisual phenomenon. It seems that with my eyes closed my brain interprets the reverb/ambience of a hard-panned sound as distance, but with my eyes open, it interprets it as what it really is - a hard-panned sound with ambience/reverb applied, coming out of that speaker’s drivers.

Even though my day job involves the interplay of audio/visual influences, in the real world I’m still constantly amazed by the type of things you’ve pointed out. Certainly the influence of visuals, even when it comes to loudspeakers and systems. That’s why when I think I might be hearing something from a loudspeaker system, I often close my eyes to listen to see whether it changes or holds up, and how much was influenced by what I was seeing. It’s also why I pay so much attention to the sense of presentation in my room (I like dim lighting with slowly evolving coloured lights on the projection screen behind my loudspeakers when listening).

But in terms of visual influence, just last night I was watching a musical performance on my iPhone, of two people, singing and playing acoustic guitar quietly. There was an amazing sense of sound quality in terms of my perception. When I was watching the hands playing the guitar strings combined with the sound coming through my iPhone speakers, there was a sort of effortless sense of “ wow that guitar sounds beautiful, and I’m hearing right through the picture to how the guitar actually sounds.”

But as soon as I close my eyes and my brain is no longer mapping the sound directly to the image of the guitar being played, that sound quality largely recedes. Now it sounds like a plinky little instrument being played through my iPhone speakers.

Brains-R-Weird.

 

[Arindal]

I hear you on that, but the Genelecs' symmetrical two-woofer setup seems to minimize the directivity differences between horizontal and vertical orientations,

A two-woofer setup with the two sound sources being vertically apart from each other, does quite the opposite - it increases lobing and creates a different directivity pattern vertical vs. horizontal.

If the x-over freq is low enough, that does not constitute a dramatic difference, though. D´Apollito, who has used the same principle for two midwoofers, put the minimum wavelength for crossing over at 3/2 of the distance woofer to woofer. Same applies here.

That doesn't change anything. Vertical and horizontal directivity is different once you pass over to the woofers.

Coaxial does change the ´horizontal vs. vertical loving´ issue for the transitional band between midrange and tweeter. Genelec coaxials are exemplarily here, at the cost of increasing directivity for some models.

It's quite possible that the ear is not sensitive to that directivity variation in that 200-600 Hz range, but I think it would be something interesting to test.

Common understanding is that the ear is pretty insensitive to the distinction direct vs. indirect sound below 600Hz (or a bit lower). Would say from own experiments that we do perceive the overall level of indirect sound in this band, though, so ideally it should not dominate. That's the reason why dipoles and cardioids sound so different particularly in the midrange. Increased lower midrange level contributes to bloated voices and dull reverb, but the reverb pattern itself does not really matter.