"clarity" would have to be sufficiently defined.
From what I've seen, a move to a narrow directivity speaker in a small-room acoustic for professional purposes is specifically related to 2 objectives:
1. "Improved" lateral tracking in the mix of "sources"/images. (..and the improvement, depending on the small room and objects within, is highly debatable.)
2. Noise detection, particularly at higher spl's.
Note that both relate to the reproduction of sound sources (images), NOT sound reflections. Reflective sound is every bit as much a part of the performance/mix as direct sound. Listening to a recording that does not have reflective sound is generally unpleasant.
Anyway, #1 is understandable when you factor in typical mixing rooms and consoles. The rooms are often very small and have numerous *close* reflective surfaces (equipment racks, windows, etc.), not the least of which is often a large mixing console right in front of the loudspeakers.
As far as #2 is concerned, this may be more a product of the design as it relates to efficiency vs. distortion. High spl loudspeakers with low non-linear distortion will "allow" for a better "look" at distortion in a recording - which would even include distortions relating to reflections in isolation recording booths that are not intended to be part of the processed reflective environment for that recording. It just so happens that the lower distortion is often a product of horn-loading and high efficiency drivers (like compression drivers), operated at sp-levels that would otherwise compress and distort more traditional drivers.
From what I've seen, a move to a narrow directivity speaker in a small-room acoustic for professional purposes is specifically related to 2 objectives:
1. "Improved" lateral tracking in the mix of "sources"/images. (..and the improvement, depending on the small room and objects within, is highly debatable.)
2. Noise detection, particularly at higher spl's.
Note that both relate to the reproduction of sound sources (images), NOT sound reflections. Reflective sound is every bit as much a part of the performance/mix as direct sound. Listening to a recording that does not have reflective sound is generally unpleasant.
Anyway, #1 is understandable when you factor in typical mixing rooms and consoles. The rooms are often very small and have numerous *close* reflective surfaces (equipment racks, windows, etc.), not the least of which is often a large mixing console right in front of the loudspeakers.
As far as #2 is concerned, this may be more a product of the design as it relates to efficiency vs. distortion. High spl loudspeakers with low non-linear distortion will "allow" for a better "look" at distortion in a recording - which would even include distortions relating to reflections in isolation recording booths that are not intended to be part of the processed reflective environment for that recording. It just so happens that the lower distortion is often a product of horn-loading and high efficiency drivers (like compression drivers), operated at sp-levels that would otherwise compress and distort more traditional drivers.
Last edited:
yeah, this sounds to me like the most reasonable conclusion from this entire discussion - I think we all agree that there must be loss of clarity in a kitchen that is adjacent to a living room where a stereo is
indeed
geeezzzz WTF are we talking about?
I'm glad that you appreciate my point: that decreasing the direct to reflected ratio, either by changing speaker directivity, increasing room effect (decreasing mean absorption) or from increasing listening distance (yes, even into the kitchen) will reduce clarity.
David S.
'Resembling the production' is imo the only definition of accurate reproduction. So, if we standardize the production environment (for example a flat response and reflections smaller than -20 dB for the first 10 ms) this automatically sets the requirements for the reproduction environment. At the production side, this 10/-20 should be a suitable window to dial in the spaciousness of the production, perhaps with the help of Qsound or related tools to overcome the limitations of stereo. Then, a narrow directivity speaker at the reproduction side would suit this requirement for small and live rooms (typical european livingroom), more freedom of choice is there when the room is treated well.
Anything else is out of the producer's control. It probably will be enjoyable and satisfying, might even be better than the original, but the result is my-fi and not accurate reproduction.
So, back to the question of keyser:
I think the answer (and new question) should be: Depends if one wants to hear the recording as intended by the producer. If one does, this implies that the recording is produced in standardized production environment and reproduced on a system which complies to this standard. What is lacking in current stereo music production is a well defined standard (which fully describes the possibilities and limitations of stereo), so thats why we already need 60 pages to discuss this topic.
Topic closed?
therefore in the end accurate reproduction thus understood cannot be regarded as a valid ideal for now
fortunately there is alternative ideal - of realistic reproduction
High Fidelity was from it's beginning first of all an idea of realistic sound reproduction, check out Hartley who was first to use the term in the 30s
fortunately lack of standards represents no impediment to pursuing the ideal of more realism in virtual presentation of music at home
but this is of course different discussion
Last edited:
This reminds me of something I was thinking about yesterday, but hadn't got around to posting yet.
I think most of us agree that lateral reflections, up to a point, can increase "envelopment" and apparent source width. Clearly this is something that many listeners like, even if it may be more Euphonic than accurate in nature.
It seems equally as clear that some degree of "clarity" is lost if the reflections are too strong and/or too early. For the sake of argument, lets define clarity in this context as "time domain accuracy", where excessive early reflections cause smearing or clutter in the time domain.
So we can't have it both ways - they're both opposite trade-offs of the same parameter. Or are they ?
What if it is not the delayed side-wall reflections that give the sense of envelopment per se, but the fact that they are a source of sound which mimics the sound from the speakers, but comes from a much wider angle than speakers are typically positioned.
In other words, our brain likes to hear sound sources coming from a relatively wide range of frontal angles to feel "enveloped" by the sound, and a standard 60 degree triangle whilst wide enough to provide image localization, is not wide enough to stimulate this sense of envelopment without the "help" of the side-wall reflections to artificially stretch the apparent source width.
Many of us can't even achieve 60 degrees due to our listening room limitations. I think my speakers are only at about 45-50 degrees separation at the most in my current room, which is far from ideal, and even when I've had larger rooms I've never tried more than 60 degrees for any long period of time, although I have tried it quite a bit wider for a few tests.
Now, what if we had a wide enough room, (helped by speakers being on the long wall) and positioned the speakers at a much wider angle - say 70 to 80 degrees apart, the speakers were fairly directional, and were toed in to minimize early side-wall reflection, (even just toeing them at the listener would be enough when that widely separated) would we feel the same sense of envelopment when the speakers themselves are as widely positioned as where the side-wall reflection spot would have typically been, without the help of any side-wall reflection ?
It's years since I've tried this, but from what I can remember, yes, there is a very strong sense of envelopment when doing this, but the phantom centre channel starts to become diffuse and fall apart beyond about 70 degrees.
What if we then add a centre speaker to this wide separation configuration, with a matrix decoder that properly isolates the phantom mono image to the centre channel speaker only, and keeps the left-right signal out of the centre channel. As far as I know Dolby Prologic (1/2) decoders can do this in the right mode.
Is this the answer to getting both envelopment / spacious feel, and definition at the same time ? Widely spaced directional speakers with a real centre channel, and a minimal amount of side-wall reflection ?
In a previous house I often used to listen from the next room when doing chores and find the same thing you do - the sound is very pleasant and well balanced. In fact "listen from down the hall-way" with no direct line of sight to either speaker is one of my rule of thumb tests for evaluating the off axis response of the speakers - what does it sound like after the sound has bounced around the room a few times and then exited the room...does it still sound balanced, or does it sound overly dull or bright etc.
There is definitely a loss of definition in the time domain though, and of course stereo imaging is absent...
Last edited:
Yes because it's pointless to focus on imaging only since artificial phantom images is quite easy to generate but it is very hard to make then sound realistic, like you would hear them in real life.
And considering David S failure in his stereo demonstration with unbiased listeners, he achieved some level of imaging because "sound was coming even between the speakers" but based on the given explanation the unbiased listeners did not appear to be convinced that what they hear was anything realistic, that they could have heard in real life situation, but the sound was coming from the speakers.
Now, the fact that sound was coming from the speakers must be due to the professionally biased speaker design (Are they designed by David S, who designs studio monitor speakers for profession?) i.e. they must have high directivity at the treble, and they must be omni at the bass.
They shall never sound realistic !
But Why David S is using such speakers? Because he, and Rudolf like to listen to the loudspeakers itself ! They don't care about realism.
- Elias
Err, you mean perhaps they like to listen to the content of the recording, rather than the contribution of their room ? How is adding stuff which is not in the recording more realistic ? You think a symphonic orchestra should fit into an acoustic space the size of a home living room with the sonic signature of a small room ?
glad You are glad and that You understand me so perfectly
I am sooo happy, I want everyone to be happy!
Last edited:
Interesting,
Although in a quick skim read it's not quite what I'm talking about, because (a) their left - right speaker separation was only 2.4 metres with a listening distance of 3 metres, which is fairly typical and less than 60 degrees, and (b) they didn't appear to be testing or rating "envelopment" of the sound, only image localization accuracy and "focus", with their interest being whether the centre channel improved stereo imaging for offset listeners. Their adjustable matrix may be ideal for what I'm suggesting though. (How they can claim to patent such a simple matrix arrangement when Dolby Prologic of 15-20 years ago can do something very similar I don't know...)
Last edited:
No, they declared themself that they want to listen to the speaker. Check back the posts in this thread, and find out this horrifying oddity yourself.
I fear there may be more of that kind of people around here too
Psychoacoustics !
I do think but not the same way like you
- Elias
Elias,
I'm honored to be mentioned in line with David S! You are too generous in doing so.
Most people who take the time to listen to my stereo setup are surprised, that they don't hear the music coming from my speakers. They had expected that the music would come from the speakers, because their "stereo" set at home works that way (bad setup, not sitting in the sweet spot etc.). Almost without exception those people are first fascinated by the strong "life like" center phantom image. It takes them some seconds (or minutes?) to realise that there are even more phantom images between the speakers. It needs some sub-standard recording to show them that an instrument/singer/effect can be "glued" to a loudspeaker too.
I'm quite sure that the lady who visited David S. reacted along those lines too. My dipole speakers have the advantage that the "stereo stage" starts 1 m "behind" the speakers - not at the same room depth as the speakers. This way the visual clue of the speakers doesn't misguide their ears as much as it does in conventional setups.
Rudolf
I do think but not the same way like you
- Elias
While talking about thoughts, how would a dipole line array help with speaker crosstalk? I've already asked that but you probably missed it.
If you want to lear more about optimum matrixes, check out Gerzon's work from the 70's.
You think it's too old ? No ! It is valid. Take into account that stereo hasn't improved since 1930's.
- Elias
Please provide a link to this (or at the very least a post #).
-wading back through this thread is headache inducing.
No need to risk a headache.
It is most likely that I have used phrases like "When I listen to my speakers, I find ..." more than once. It didn't occur to me how literally such terms could be taken by a member of this forum. Rudolf
I'm still trying to understand how sound coming from the speaker (that is, apparently coming from the speaker, rather than elsewhere) is a bad thing.
We place a sound in "reproduction space" by panning it between left and right channels and by varying the amount of reverberation attached to it. In that manner we can place a sound source in the distance (lots of reverberation relative to the direct level) or pull it up to the speaker itself (little or no reverberation). This can happen by the way raw sounds are mixed in the studio or it can be a natural product of microphone techniques and real spaces.
For example if we record spoken voice in an anechoic chamber and play it back over our speaker we would expect for the sound to be right there, at the front of our speaker. It may be your goal, but not mine, that a dry recorded sound element float around the room divorced from the speaker. I want the widest range of reproduction space from right at the speaker to well behind, from left to right over as much of an arc as possible, and with a good balance between diffusion and clarity.
As to the speakers they were Snell XA90, a symmetrical array with fairly normal lateral directivity and slightly higher vertical directivity due to the expanding array design. Read a review if you want at
Snell XA90ps loudspeaker | Stereophile.com
An anecdote about that design. We took a pair to a dealer in California and went to hook them up in a fairly large and reverberant showroom. I can't remember what system was playing initially but they didn't quite sound right, imaging vaguely. When we hooked up the Snells it was immediately obvious that the system was out of phase. It turned out that one of the balanced cables to the amp was reverse wired. So how can one system hide that it is reverse connected while another instantly reveals it? Which do you think would image better?
This is the tradeoff between a system with inherent diffusion and one with real imaging precision.
Hey Rudolf!
David S.
For *compelling* reproduction it should NEVER sound as if it's "coming from the loudspeaker" ..unless it's a purposefully processed digital sound artifact - i.e. an "effect".
Now an "image" superimposed on/over the loudspeaker is a possibility (for compelling reproduction) - but that implies more than simply a panning process, i.e. a vertical element exits there as well, as does a certain "diameter" to the image.
Clear as mud?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.