But tweeters in a typical stereo loudspeaker are (often) points ! A sound coming from a point should sound like, well, like a point ! Thus the hifi term pin point localisation ! Hifi seldom has anything to do with realism.
- Elias
That's certainly true as a musical argument for large orchestral / choral works. As an argument in the context of accurate playback, it could be seen as an excuse for poor imaging. That's not a shot at the author (who has done much good for the industry), it's just an observation.
Side-step; related thoughts below is not commentary on the article.
I'm willing to admit that the Bose 901 sounds wonderful with the right room, placement, and source material. It's downright lousy for accurate imaging, and much of it's wonderfulness is the result of a manner of presentation that masks the harshness in recordings, and masks several of the distortions that all transducers have to address at some level. It is capable of a giant, dynamic, and non-fatiguing soundstage at realistic levels. This is very musically satisfying in many situations, and the presentation creates an impression of "realism" that is often interpreted as accuracy.
I'm a serious classical buff.
I've noticed, however, that many reviewers and audiophiles make concert hall recordings the sole criteria for judging a system's 'accuracy'. This is silly--Most of these recordings sound different, often quite better, than the real thing. Recordings on the CSO Resound label are a good example: Consistently world-class recordings in a hall that is widely recognized as mediocre at best. Not honest fodder for evaluating whether system playback is faithful to reality; only good for enjoyment of an engineered alternate reality.
Meanwhile, there are loads of recordings that indeed contain small-source sounds in specific locations - not an area. An intentionally intimate recording of a small group using a variety of hand-held ethnic percussion instruments is a good example. Chesky Records do this kind of stuff very well. Any recording that captures an intimate perspective can't be fully appreciated with playback in areas.
Accuracy and musicality aren't mutually exclusive, just harder for most setups to achieve. A good system plays everything well, and we are always free to introduce colorations to suit our tastes.
it is what it is
ps.
the statement of Eargle was not put forth in the specific context of "large orchestral / choral works" - "musical sounds" mean "any musical sounds" there
Last edited:
There
I made no effort to malign his context as it seems pretty clear. I provided a context in which I would heartily agree. His statement infers that sounds that are highly localizeable are not musical sounds. It's reasonable (not argumentative) to point out that there are many musical sounds that are well served by a system that renders imaging and scale accurately.
Localization cues are carefully captured on many recordings for the express purpose of increased realism.
A system that can image accurately will also render 'areas' when that is what the recording provides.
There
by "there" I mean the paper by Eargle
You clearly do not understand what I am talking about - I am not talking about somewhat undefined area of sound as opposed to clearly defined sound source - I am talking about precisely defined sound source that has appropriate realistic width and therefore it is bigger or smaller area and never a point
They are very well !
I just may have discovered the Grail of home audio reproduction ! 
An externally hosted image should be here but it was not working when we last tested it.
And how is your 3-way flooder progressing ?
- Elias
And how is your 3-way flooder progressing ?
- Elias
I have to wait, this is all I can say
OTOH if a thing like this is worth doing at all it's worth doing right
I expect the back-to-back thing to arrive in just about 2 weeks!
Elias,
I know that your argumentation above is highly ironic. So only for those who wonder about the facts: Any "area" information in a musical performance is recorded by the point-size stereo microphones as small differences in volume or timing. If we want to replicate those timing and/or volume differences in the playback room as good as possible, the stereo sources have to be as small as possible (compared to the wavelength involved).
If that is not the case, the smallest sound "area" achievable in reproduction is not controlled by the recording, but by the loudspeaker construction.
I expect the back-to-back thing to arrive in just about 2 weeks!
You have ordered the Stereolith ?
No I was not ironic
Because, the stereo mic cannot sample the size of the source as such. It only samples pressure and/or velocity in a single point (ideally), which do not contain information of the shape of the sound field and thus no information of the source size.
ASW is generated mainly in the reproduction end by first room reflections. The speaker size bears no role in it, other than generally a given size of a speaker naturally have a given type of directional behaviour which affect a specific first reflection pattern. But the size of a speaker do not have to be small, nor sound must come from a point.
- Elias
You have ordered the Stereolith ?
nah, why should I?
my own back-to-back, fourth incarnation
It is not sampled in the single microphone, but in the signal difference between the left and right stereo mics. If stereo can sample the distribution of individual sources on the stage, it can sample the size of the individual sources too.No I was not ironic
Because, the stereo mic cannot sample the size of the source as such.
individual sound source size perception is a complicated thing and relies on different set of data then perception of distribution of individual sources on the stage but yes - stereo pair of mics can sample the data BUT conventional stereo pair of speakers fails to reproduce it - that's the problem
Cool !
Don't want to depress, but I have just found there is an improved alternative over conventional back to back:
Elias Pekonen Home Page - Stereophonic Sound from a Single Speaker
- Elias
P.S. We're already pass the point where back-to-back technology can be called conventional
P.S. We're already pass the point where back-to-back technology can be called conventional
actually we have passed this point long time ago - have You forgotten Radugazon's back-to-back/OSD multi-way setups?
as to Your "3-channel" single box - have You considered/measured the effect of the back wave of one speaker through the other two?
No, because if only two spatial points are used to sample the propagating field, only a plane wave can be constructed because a line is the only curve that can be drawn through the two points without ambiguity. So stereo can sample the (horisontal) direction of the sound sources, but it cannot sample the size nor the shape of them.
- Elias
Elias,
Thanks for taking the effort to explain how stereo is only defined along a horizontal line. I thought that was evident from the start - isn't it part of the definition of stereo?
So, if "stereo can sample the (horisontal) direction of the sound sources", it can sample the horizontal size of them too. The latter is a stringent implication of the former.
Cheers, Rudolf
No
Try placing the source at the infinity, the (horisontal ) direction remains the same, but there is no information of the source size because it will be singularity no matter how big was your original source size.You need wavefield synthesis to capture the field from a source having non singular size and located at a finite distance. Stereo cannot do it.
- Elias
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.