Properties of speakers that creates a large and precise soundstage

[Duke]

Relying on Griesinger papers, sound is more accurate with stream separation, and less accurate without, due to auditory system, your internal processor! This doesn't mean anything absolute, it just means that if you find what you perceive is not accurate, first make sure you are having stream separation going on...

Citing Griesinger here as background information for those not familiar with:

"Envelopment is perceived when the ear and brain can detect TWO separate streams:
A foreground stream of direct sound.
And a background stream of reverberation.
Both streams must be present if sound is perceived as enveloping.

"Presence depends in the ability of the ear and brain to detect the direct sound as separate from the reflections."

(Note the implication that the perceptions of "envelopment" and "presence" are not mutually exclusive.)

In another thread @Bjorn posted an energy-time curve which imo visually illustrates "stream separation" in a home audio context. The spike at about 8 milliseconds on the horizontal axis is the "foreground stream of direct sound". The diffuse reflection energy beginning at about 18 milliseconds is the "background stream of reverberation". And the 10-milliseconds-ballpark gap between the two is "stream separation":

"

 

[jim1274]

Setting everything right as much as possible is only the recording that does the trick.
Strangely the best 3D image I perceive (not with test files as the ones to test those stuff) comes from the very old recordings such as this:

https://www.discogs.com/release/4155614-Rimsky-Korsakoff-Reiner-Chicago-Symphony-Scheherazade

Very well layered in all dimensions.
Very important is ample power so it will retain it's high dynamic range,it short of flattens out when clipped,even mildly.

Thanks for the suggestion—always glad to get some new soundstage test recordings.

I think this might be the same recording, just reissued with a different cover. Same Harth as soloist, so likely the right one—going on next.

 

[jim1274]

So, what are the most natural-sounding speakers in the world with the most 3D sensation, depth, and layering? For me, it's probably many models from Kef, as well as Focal's Shape series and most products from HEDD.

If there was a definitive answer to this, we could all take a break and get a pair. I’d love to clear my living room filled with speakers and cables everywhere trying to figure that out…

 

[theREALdotnet]

I think he mentioned that Madonna recording in regards to Qsound, didn't he? A track or two, can't remember if it was vogue or respect yourself, used Qsound to produce sounds that jumped out and surrounded the listener.

It’s the Immaculate Collection album. All tracks are in Q-Sound, but a few of them are effect show pieces, esp. Vogue and Express Yourself. In Rescue Me you can hear the rain and thunder from behind you.

I was just listening to a track by a 90's electronic band that used Qsound - very freaky. But I wouldn't use a Qsound track to judge soundstaging or imaging.

I think you can use it to judge the ability of a speaker/placement/room combo to produce good imaging. I have found a strong correlation between how well the Q-Sound effects work and how accurate a well-recorded a string quartet is presented. The Madonna tracks are among my tools for troubleshooting and tuning speaker placement, toe-in, wall absorption etc.

 

[theREALdotnet]

I'm sorry the post got very very long, but it's hard to write about the stuff shorty and that's because I think this subject makes great confusion all around the forums, seen almost daily, many many threads I have no time to write on, so hopefully occasional long post gets some attention

I think the point of this thread is that if you keep everything else the same – recording, listener, room, speaker placement – there are still differences in how well different speakers produce soundstage and imaging. Some will do better than others. The questions is, which characteristics of a speaker are responsible for that?

 

[MattHooper]

Tonal accuracy is very important, but how does one know what “accurate” means in terms of the soundstage characteristics noted above?

I think the various discussions on that on this forum have only shown how difficult that is to pin down.

For instance, in terms of flat frequency response and absolute lack of distortion, I'm not fussy in chasing some version of "total accuracy" because that's a bit of a rabbit hole and my position is that the essential, important sonic and musical information about a recording and performance can be heard through any number of systems. Similarly, with imaging, I think the essential information is easily heard on any number of systems, if you are just sitting in a sweet spot for stereo listening. I've cycled through lots of speakers and the essential positioning of instruments and voices in my favourite recordings have never changed. It's just, in the end, somewhat subtle differences in the character, like on one set up the guitarist will appear a couple feet to the side and in front of the drums, but on the next it will be the same, though seem just a bit more distinct and focused. But, that's subtle stuff, the panning and spatial choices in the mixing are obvious on all of them.

If you mean accuracy in terms of spacial localization precision, I will say that the Omnis may not match others.

I may differ, at least with my experience with the MBLs in my room.

There is one sense of precision in terms of image focus, that is how tightly defined an image is. So with two different speakers you can have the images in the same positions in the soundstage, but the specific images of a voice or sax or whatever may feel more tightly focused, even squeezed a bit smaller in being so, on one speaker, and a slightly larger or more diffuse in another. I'd say the MBLs never quite matched my Thiel speakers, which have a particular strength in image focus, but even so, I got pretty good focus so for instance a vocal could seem focused to the right size similar to a real voice etc.

Then there is precision in terms of the spatial relationships of the images - so a vocalist in the middle, stand up bass behind a couple feet and to her right, acoustic guitar a couple feet to the left but closer, keyboards floating more in front of the guitar, etc. I found on the MBLs the spatial relationships seemed actually more distinct, a bit like trying to judge distances with one eye, and then the MBLs were like opening both eyes for stereoscopic vision and the space between the instruments now seemed more exact and dimensional.

So....which is more "precise?" In some way I see the MBLs as more precise in those terms. Which is more "accurate?" Well, if you just assume the conventional speaker is the one more accurately re-creating the spatial relationships from the mix, that assumes the answer.

Was that near field listening in both instances such that differences in your room acoustics played no part in your assesment? Just curious. The room acoustics have significant impact on the final sound for any speaker.

Between the KEFs and my set up: similar listening distance, if anything a bit closer to the KEFs. I always do a bit of testing with listener distance, so when I listened on my friend's KEFs I also moved closer to nearfield to see what they could do. But his set up is not optimized for the KEFs, so the difference I was citing was not necessarily difference in the capabilities of the speakers, but in how I had taken the time to set mine up to achieve a certain goal, relative to if someone isn't thinking along the same lines.

 

[jim1274]

I think the point of this thread is that if you keep everything else the same – recording, listener, room, speaker placement – there are still differences in how well different speakers produce soundstage and imaging. Some will do better than others. The questions is, which characteristics of a speaker are responsible for that?

I’m taking that as optimal positioning for each speaker specific to its unique dispersion characteristics that will have its own best placement to maximize soundstage and imaging?

We are talking within ones specific room acoustics environment, so that is a constant, and level matched same recording used for comparisons is absolutely a must.

The last sentence is the million dollar question?

 

[MattHooper]

It’s the Immaculate Collection album. All tracks are in Q-Sound, but a few of them are effect show pieces, esp. Vogue and Express Yourself. In Rescue Me you can hear the rain and thunder from behind you.

Right. I have that collection which is where I used to hear it. Couldn't remember how many songs.

I think you can use it to judge the ability of a speaker/placement/room combo to produce good imaging. I have found a strong correlation between how well the Q-Sound effects work and how accurate a well-recorded a string quartet is presented. The Madonna tracks are among my tools for troubleshooting and tuning speaker placement, toe-in, wall absorption etc.

Yes, and I have used qsound in the same way. In fact, I just used a qsound track a few days ago (the one I mentioned) at a friend's to test out the speaker set up in the way you suggest. If a speaker set up doesn't do the Qsound effects right, there's something wrong.

But that's a bit of a different point than the one I'm making, which is that I think conventional stereo is overall what you'd want to use, and one wouldn't want to mistake the effects of a qsound track (like if you didn't know it had qsound) for a stereo effect. There are conventional tracks on my speakers that have quite a "wrap around" vibe in terms of imaging, which I love, but it would be less impressive if I then learned it was just a qsound effect.

 

[theREALdotnet]

But that's a bit of a different point than the one I'm making, which is that I think conventional stereo is overall what you'd want to use, and one wouldn't want to mistake the effects of a qsound track (like if you didn't know it had qsound) for a stereo effect. There are conventional tracks on my speakers that have quite a "wrap around" vibe in terms of imaging, which I love, but it would be less impressive if I then learned it was just a qsound effect.

I agree, Q-Sound (or other effect systems of that kind) are kind of gaudy and not what I’m looking for in classical or jazz recordings.

That said, I found that natural (single or minimally mic’ed) recordings of small acoustic ensembles never contain any surround effects, i.e. sound coming from the side or back of the listener. The sound stage always extends from just behind the speakers towards the front wall and beyond. If surround effects are present then I’m pretty sure technical trickery and manipulation during mastering was involved.

 

[jim1274]

Thanks for the suggestion—always glad to get some new soundstage test recordings.

I think this might be the same recording, just reissued with a different cover. Same Harth as soloist, so likely the right one—going on next.

View attachment 349505

Big thumbs up on this suggestion. As good or better on soundstage size of any recording of its type I’ve come across. I found the exact one you referenced on Tidal—didn’t show up in initial search but noticed when scanning all the CSO albums:

 

[theREALdotnet]

Strangely the best 3D image I perceive (not with test files as the ones to test those stuff) comes from the very old recordings such as this:

https://www.discogs.com/release/4155614-Rimsky-Korsakoff-Reiner-Chicago-Symphony-Scheherazade

I also find great imaging with anything mastered by John Eargle for Delos. Sadly, Delos never quite managed to attract the top tier of performers, although what they have is quite good.

 

[goat76]

I don’t think any signal reproduction changes will make a difference. IMHO, beyond a certain level of quality in the signal chain, I don’t think there would be any audible differences, the room acoustics and speakers/placement being the major defining factor. I’m not going to complicate this by introducing any other variables beyond speakers/positioning into the equation.

Don’t forget the listening position that is equally or even more important than the room acoustics and the speaker position. According to Jesco at Acoustic Insider, finding the best listening position in a room should always be the starting point if that is possible, which of course is way easier to do if the room is fully dedicated to music reproduction.

 

[tmuikku]

But….just thought of something…

When moving closer, the sound level increases from the reduced distance. Just did it several more times back and forth on listening distance,
focusing intently on clarity of one instrument in the mix. I’m not so sure this is change in clarity as much perception from music being louder, or maybe a combination of both? I stayed at my far listening position and adjusted volume up and down to mimic the SPL changes of moving closer and seemed to have a similar effect. Does this make any sense? I know it’s absolutely critical to match sound levels when doing an A/B—the louder sound sounds different and almost always better.

Hi,
Yeah direct sound gets louder close enough speakers, but not sure if the transition happens there, certainly it eventually would. It's not just clarity but localization, and also envelopment that seems to happen.

The sound of room is kind of level anywhere in the room, and swamps direct sound everywhere in the room except some perimeter around speakers, where direct sound is loud enough to swamp the room sound. This is the actual technical and measurable critical distance, and also detectable by ear. I think the transition in my place extends beyond this, listened to it at some point and should do again and also measure where the critical distance is about.

The trasition in perception might still be from crotical distance, if ear is sensitive at some particular bandwidth whose critical distance is further out in the room. As speaker directivity increases towards highs, usually absorption in room rises towards highs and also processing time for brain to handle direct sound before first reflections arrive increases towards highs, and this all is due to wavelength getting shorter. I would guess important bandwidth for perceptual transition is roughly the human voice, the midrange, especially higher part above fundamental, which contains most harmonics as it's the harmonics that are important according to Griesinger.

You are right that at any given moment multiple things are going on at the same time, and it is highly likely there is no some particular thing that makes a perception, but combination of basically everything. Level of direct sound would have some effect at least, be it from more power or by getting closer, except getting closer changes D/R. And not just D/R, but with stereo triangle all angles change, early reflections delay get longer and attenuated in relation to direct sound, and so on, a lot of things change.

Some key and insightful points here. At the end of the day, it’s what “sound” one prefers. The only conclusion I’ve tentatively come to is conventional forward firing box speakers lack the spatial image soundstage spaciousness qualities that the Omni and dipole provide. The addition of reflective sound from multidirectional radiation is additive (to me) if tuned properly in delay timing and level by optimal speaker positioning. The added reverberant sound creates an illusion of presence of live musicians. My testing will focus on what other aspects of good sound like clarity, timbre, and may be sacrificed in the process.

Yeah absolutely, if I remember correctly spaciousness, or feeling of space, was very high in list that affects preference score of speakers, I believe it was in Toole's or Olive's research. Feeling of space can be had with the clarity, and then it's called envelopment as Duke quoted few posts back, this happens with stream separation. You could also have spaciousness from your own room, without envelopment and without clarity, and this is when there is no stream separation.

You could also have stream separation with clarity, but with rather poor envelopment. Now that you mentioned the feeling of space I think this is one aspect that plagues domestic room acoustics, certainly my living room, and perhaps one important reason why many people would choose to arrange system so that it does not make stream separation, but instead provides the hazy spacious sound of local room.

Feeling of space that way happens without any effort basically, and requires no special knowledge on things, just put speakers anywhere they fit, and it's highly likely the sound is spacious but without stream separation. Conversely, if almost by accident the setup is so that there is stream separation, I think it's highly likely one would have no spacious feel to sound if envelopment needs work, and would play around with positioning until there is, right, which took very likely back to no stream separation and spacious sound it is, thanks. Be not mistaken, Griesinger says this is not envelopment though, and it has no clarity, no accurate localization, no engagement and so on, which only happen with stream separation. I think it's quite a task to arrange nice envelopment at home due to short reverberation time, and walls close by (modes, bass has importance with envelopment), but more over a lot of people do not even know or think about this, that our own auditory system caters the whole plate, and provides the easy solution for free. Gradient descent to a wrong valley.

See, finding the transition, knowing if you have stream separation or not, is crucial. When you get over the transition closer to speakers you are now able to listen envelopment and do something about it, you know stream separation happens and you should have clarity and you should have envelopment. You could nowfurther adjust positioning, toe-in, speaker directivity, room acoustics and know exactly how to listen the envelopment, find the transition, go over and listen as long as needed to understand how your room sounds and what you need to do about it. Move back and forth at the transition to AB how your envelopment compares to spaciousness! Finding and utilizing the transition is key to be able to tune the system to your liking, but it takes some effort.

Since your wishlist of sound includes clarity and feeling of space, you must have MLP closer than the transition to get clarity, and if the feeling of space is not good enough now you can now listen to it and improve on it, always referring your perception to the transition. Your logic rooted to your perception, key to be able to pull it through I think.

In the vein of topic, listening the envelopment for some time tweaking with positioning and thinking about it all one gets an idea what kind off directivity could help, what kind of room acoustics adjustments could help, and so on. And as mentioned many times earlier, power is eventually just the knowledge, ability to listen at any position that feels good to you, go over transition at will if you feel like it. If the system is optimized you don't have only one good sound available, but two!

 

[tmuikku]

Another point made more than once along the way is the powerful effect of adaptation. During this A/B testing phase, I’ve often done extended listening to both whatever A or B is playing, forgetting which is playing after distractions or interruptions. While the difference is significant and obvious when doing instantaneous A/B back and forth, when I guess which is playing after an interruption and forgetting which is which, I’m often wrong, especially when it was Omni/dipole in the A/B. It’s easier to guess between the BMR box and the AMT dipoles due to the soundstage reverberant aspects, but not batting a thousand on that for sure.

Yeah, in the end its all kind of subtle, and many people are just unaware of such details even exist in a way, and this this is the audiophilia thing except we are not buying gear here but learning to listen. I think our systems are only as good as our listening skill, which is always optimal isn't it, as good as it has ever been Removing confusion from the stereo/spatial auditory experience is huge boost to listening skill, connects logic to perception, builds trust, which enables effective use of perception. But it's not all of course, seek for Harman How to listen software for example.

 

[goat76]

After doing more evaluation of forward firing speakers (BMR in this instance), your points are becoming clearer. When getting the geometry of listener and speaker location just right, the soundstage becomes more 3-D. I clearly did not yet evaluate forward firing speakers enough in terms of soundstage. I’m still not convinced that properly tuned rear (diople) or 360 degree (Omni) added reflective energy is not additive, but the gap has narrowed now that I’m getting more optimal placement of the forward firing speakers. Hope I don’t have to eat too many words before this is over..

All that depends on if you mostly want to hear the recorded space (or reverb), or if you mostly want to hear the acoustics of your own listening room.

The added reverberation you get with your dipole and omni speakers can likely be to your liking, but it will also introduce more “sameness” to everything you are listening to.
With the forward firing speakers and you in a position and an acoustically treated room with a high ratio of direct sound vs reflected sound, you will get a clearer view into the recorded space and you will hear more variations from recording to recording. This is most likely the sound the mixing engineer heard while making all the sound decisions for the mix.

I don't see anything wrong with either approach, whatever rocks your boat the most is the thing you should go for.

 

[Sokel]

Big thumbs up on this suggestion. As good or better on soundstage size of any recording of its type I’ve come across. I found the exact one you referenced on Tidal—didn’t show up in initial search but noticed when scanning all the CSO albums:

View attachment 349509

Apart from the galaxy of genius performers and the conductor of course,people involved with this series of recordings (Reiner,RCA) really did a amazing work with the means of the era.
Lew Layton is my hero.

 

[tmuikku]

After doing more evaluation of forward firing speakers (BMR in this instance), your points are becoming clearer. When getting the geometry of listener and speaker location just right, the soundstage becomes more 3-D. I clearly did not yet evaluate forward firing speakers enough in terms of soundstage. I’m still not convinced that properly tuned rear (diople) or 360 degree (Omni) added reflective energy is not additive, but the gap has narrowed now that I’m getting more optimal placement of the forward firing speakers. Hope I don’t have to eat too many words before this is over..

If you found 3D kind of a feeling of space, you have very high chance the transition is right behind you! I urge you to find the transition, where 3D feeling collapses back to 2D.

Try if you can collapse the feeling of 3D space into more 2D, that is in front of you, just by getting bit further away, should happen quite fast, maybe within few feet. I don't know what your current positioning is, but assuming you pulled the speakers few feet closer to you the transition is within the few feet behind you I think and you could go standing behind your chair and try to spot it. If you like to listen seated, try moving the speakers bit further away from you, like half of what you moved them in to get the 3D. Idea is to get the transition about at your listening chair so that you can lean forward to zoom in to get the 3D sound, lean back to get 2D. It's lots of fun, could be like sticking your head inside the sound in a way

 

[tmuikku]

I think the point of this thread is that if you keep everything else the same – recording, listener, room, speaker placement – there are still differences in how well different speakers produce soundstage and imaging. Some will do better than others. The questions is, which characteristics of a speaker are responsible for that?

Hi, yes, see post https://www.audiosciencereview.com/...rge-and-precise-soundstage.48542/post-1881999, maintain original harmonics and dynamics for example, in general just fine speaker without obvious audible issues.

In short, my current understanding of all this is that almost any speaker that has no obvious problems with any reasonably smooth directivity is fine, but specifics depend how you arrange the system in your room. You must get the stream separation happening in brain for attention, and after that you could optimize more but how that is done depends on your room and practical positioning opportunities. So, the suitable speaker very much depends on room, but if you can arrange the room at will there is high chance almost any nice speaker works just fine, you just must use your ear to position everything accordingly.

If placement needs to be static, then you might have some requirement for directivity due to room acoustics in order to have stream separation extend all the way to listening position, the further you are and the poorer the room acoustics everything lumped, the better your speaker must be, and narrower the directivity must be.

Or, if you don't mind about stream separation and like relaxed sound better, then any speakers that have nice balance and left/right well matched should produce the spacious image. In general, just problem free speaker and then rest depends how you position it all. Positioning locked, you could try narrow coverage speakers if you wish stream separation to happen.

 

[theREALdotnet]

In short, my current understanding of all this is that almost any speaker that has no obvious problems with any reasonably smooth directivity is fine, but specifics depend how you arrange the system in your room.

From my limited experience (of almost 40 years listening to many speakers, yet still a tiny fraction of what’s out there) this may not be the case. I have found it difficult to find speakers that can cast a “holographic soundstage”, to borrow a term from audio sales speak. Most I’ve heard cannot do this to a satisfactory degree. Now, it is perfectly possible that the speakers which failed this assessment would have come through in circumstances vastly different to my domestic listening spaces, or at impractical (for me) listening distances, so your assertion may still be true.

 

[tmuikku]

Yeah practical positioning is problem in sense that instead of changing positioning to optimize sound, which is free, one would need to start looking for gear and acoustic treatment that happens to support the practical positioning with given acoustics, so that sound at listening position makes your auditory system to a sate you wish to perceive

I think prerequisite for 3D sound is the stream separation and then optimizing for fine envelopment, otherwise the sound kind of localizes to your room to a frontal 2D sound, especially if frontal early reflections come short and loud. And virtually no marketing leaflet says anything about it, even the forums have been concentrating mostly on the gear instead of auditory system as such. Also on early reflections, but how they actually sound like has been in my view just a sideline with all kinds of detail to it but how to actually listen to the effect has been hard to understand, at least for me.

So, us hifi enthusiasts and consumers have been in confusion with all this, power has been on the sales departmemt side. But now, knowing Griesinger work and trying to attach my own perception to my situation allows me to take control, remove confusion on what I hear and what directivity and all that mean perceptually to me, in my context, and gravitate toward sound I want, take advantage of my own perception with my own setup.

Perhaps professional acousticians and others on the industry might have had an idea for long time, after all studios have had nearfield monitors half a century at least but for example Griesinger papers and what he bases on are relatively recent. I see his studies work like a map to our perception, allowing connecting things to perception. Finding the perceptual transition that happens with stream separation you'd know otherside could provide 3D like perception, while the other could not, at least not in similar way. And these are tough concepts to have an idea and communicate over unless everyone has similar perceptual experience and ability to communicate it over. Learning to listen with the transition and looking Griesinger papers what one should perceive on each side of the transition (limit of localization distance) is common ground, as it is not speakers or room but our auditory system we all share. The transition allows AB compare your own system, with your own perception, if either side gives you the sound better what you are looking for