nunayafb said:
Wrong.
I explained what would happen in my earlier post (#1931). I explained why that scenario would not give you any differential in soundstage height.
nunayafb said:
Our ear detects when a point source moves in a vertical way. We localize a moving point source because of a number of reasons (doppler, amplitude, FR, comb filtering, early and late reflections, diffusion etc,,,)
Speakers don't move.
nunayafb said:
That sounds plausible, but, no. We alter room acoustics around instruments all the time during recording sessions. If you put down a carpet around a drum kit that was previously recorded while on a shiny hard floor, it will not sound higher or lower. The ambience around the kit will be different for sure, but the kit will still stay in the same vertical "plane".
nunayafb said:
It's not difficult at all. I've been doing this for 20+ years, I'm explaining very simply why there is no "vertical" height imaging.
It would defy physics.
Has anyone flipped their speakers and made the vertical image go the other way yet? .....
Cheers
rdf said:
I'm afraid you are the one being incomplete, rdf.
I explained in easy, simple to understand terms what does, and does not happen. I'm not about to write a book here explaining doppler effect, comb filtering etc...
Please point me to any articles that explain to me how to mix in "height" in a recording, as I would be most interested, and most grateful.
What "modern" equipment has controls for "height" adjustment of sound? What controls are used? I have worked on dozens of mixers and I have never seen one that has a knob or fader that varies "height".
As far as reports from audiophiles, most are simply postulating, just as I do when I try and figure out magic tricks.
Cheers
Daygloworange said:
A listener in an anechoic chamber can't discern height for lack of all cues you've listed?
Somewhere, I have a test CD, which contains a tone which is positioned 30 feet in front of you, then moves upward in 8 foot steps. It uses the techniques described above - and is an auditory illusion, obviously.
My speakers are symmetrical full-rangers - flipping them would be pointless.
Another experiment - take your stereo speakers and position them centrally - one near the floor, the other 10 feet up. Keep your head level and listen.
I suspect you would get mono. No sensation of a vertical soundstage.
(btw I am not doing it with Sachikos). And if you tilt your head of course ...
If you wanted vertical faders - they would be implemented as an effects insert using a effects chain with a filter / reverb set up in a particular way. I presume that's how the test CD was done. If I find it I'll let you know.
In the same way you would find no "soundstage depth position" faders on your mixers - that dimension is implemented as reverb. You do not have to move your speakers physically away from you to appreciate this effect! (not trying to be cheeky, just drawing the analogy with the "inverted speakers" ???)
Stages are horizontal, orchestras ditto, heads ditto, people are comfortable listening to this. I can understand why the music industry has not bothered trying to give the illusion of height.
If I get the time (I've got Tracktion here) I'll try recording a piano moving vertically for you. 🙂
My speakers are symmetrical full-rangers - flipping them would be pointless.
Another experiment - take your stereo speakers and position them centrally - one near the floor, the other 10 feet up. Keep your head level and listen.
I suspect you would get mono. No sensation of a vertical soundstage.
(btw I am not doing it with Sachikos). And if you tilt your head of course ...
If you wanted vertical faders - they would be implemented as an effects insert using a effects chain with a filter / reverb set up in a particular way. I presume that's how the test CD was done. If I find it I'll let you know.
In the same way you would find no "soundstage depth position" faders on your mixers - that dimension is implemented as reverb. You do not have to move your speakers physically away from you to appreciate this effect! (not trying to be cheeky, just drawing the analogy with the "inverted speakers" ???)
Stages are horizontal, orchestras ditto, heads ditto, people are comfortable listening to this. I can understand why the music industry has not bothered trying to give the illusion of height.
If I get the time (I've got Tracktion here) I'll try recording a piano moving vertically for you. 🙂
jneutron said:
(Your) sarcasm aside, my example of a phone is not dissimilar to an outdoor noise source (to which our hearing is adapted), except it is indoors. Like a loudspeaker.
Here is another example that our localisation is not purely sonic. This time I will use loudspeakers, and indoors, to minimise attacks and be as relevant as possible. When listening to an audiovisual production eg home theatre or TV, voices seem to emanate from the location of the speaker's head, irrespective of whether they should do so based on loudspeaker placement and channel mixing. eg#1 centre speakers are often below the screen, but the mind overrides the ear and sources the voice where we see the speaker's head. eg#2 I have a homebuilt speaker on the right side of the TV my wife uses, and again the mind overrides the ear and has no problem locating the apparent source of voices on the speaker's head.
So, I still maintain that your proposed test method (of identifying the sound of cables by changing the cable on one channel only and listening for localisation changes based on our stupendous powers of localisation) is based on an exaggerated sense of the ear's power.
If we can't hear cable changes in a straightforward blind listening test, it is beyond our powers of resolution while listening to music through speakers in a room.
P.S. I have been noticing the extent of L/R differences in the measured frequency responses of loudspeakers being tested by magazine reviewers. These are thousands of times greater than the difference caused by changing (functional) cables. If you want to do something useful for the sound of your hifi, guys, try enhancing the matching of your speakers based on measured frequency response. And place them symmetrically in a symmetrical room. And absorb first reflections. And equalise the bass. By the time you've done all that you will have long forgotten about cable crap. Until the next time you walk into a hifi shop and the salesman reminds you 🙂 (but we're diyaudio guys, we don't waste time in hifi stores, right? 🙂 )
I don't want to comment about cable sound. Anybody can verify it:
- take 10 cheap cables, but different types (coaxial, normal wire, high frequency cable, ..) and hear for yourself. Depends of driver you have, there may be or may not be any difference.
I just want to say something else: all audiophile cables are made from normal cables, add connectors, print company name, make 500+ price an that is it.
Sometimes R && C are added to change the sound.
We all have seen pictures from speaker factories, amplifier, cd factories. Some of us even visited this factories. (I was also inside a cable factory few times - normal electrical cables)
Anybody seen pictures from "audiophile cable factory"? Anybody was inside? 🙂)
- take 10 cheap cables, but different types (coaxial, normal wire, high frequency cable, ..) and hear for yourself. Depends of driver you have, there may be or may not be any difference.
I just want to say something else: all audiophile cables are made from normal cables, add connectors, print company name, make 500+ price an that is it.
Sometimes R && C are added to change the sound.
We all have seen pictures from speaker factories, amplifier, cd factories. Some of us even visited this factories. (I was also inside a cable factory few times - normal electrical cables)
Anybody seen pictures from "audiophile cable factory"? Anybody was inside? 🙂)
tpsorin said:
What exactly do you mean by "normal cables"?
Do you mean cable that's ready-made and bought off the shelf?
If that's what you mean, then you are incorrect.
se
Alan Hope said:
Incrementally moving toward you from 30 feet away is called proximity, I don't know what auditory "illusion" that is. Proximity is not an illusion. As you move the source closer, there is less diffusion of the sound source (ratio of direct vs reflected sounds).
Alan Hope said:
(You are correct) but how can you state that it's pointless without doing it empirically?
Full range single drivers would be the perfect speaker to prove the existence of vertical height.
Once again, here is the experiment to prove (or disprove) the existence of vertical height.
If the centerline of the single driver were 40" off the ground, and a feature instrument was heard to be 60" off the ground (20" above center axis), then flipping the speaker upside down would make the feature instrument go (20" below) axis (and be heard to be 20" off the ground).
You will not hear any change in vertical soundstage, nor L/R balance, soundstage, or depth either.
Alan Hope said:
Uhhh, ok...... Been using filters and reverbs for decades, still don't know how to make an image increase in height. Never seen it in a manual, nor had another recording engineer explain that one to me.
Please elaborate. Give me an example of which filters, which reverb effects do this, what algorithm parameters and how you adjust them to increase the vertical height placement of a sound. You can get as technical as you wish.
Also please list which equipment you know that does this.
Bad example. Soundstage "depth" is simply proximity. You alter proximity with miking to increase the ratio of direct vs indirect sound (room reflections), or through adding artificial reverb to the original sound source.
But you have "pan" controls that go from extreme left to extreme right.
You alter L/R (pan)positioning by altering the ratio of sound assigned to the L channel vs the R channel.
Where are you going to alter the (ratio of) sound from top to bottom?????
There is nowhere to "pan" a signal to in height in 2 channel You don't have an "Up" channel and a "Down" channel.
Ok. I'll be sure to check in. I'd love to hear this....
Cheers
Daygloworange said:
Whooooaaaa! DUDE! IT WORKS!
An externally hosted image should be here but it was not working when we last tested it.
😀
se
Daygloworange said:
Either I didn't make myself clear when i set up the post for option 1 and 2 or you didn't read carefully- I agree with you 100% that you cannot record height w/ two mics as described earlier.
Hopefully you know from reading my post that I agree with that, Yes we are in 100% agreement that speakers don't move. (I almost considered calling you captain obvious here but that would be rude), please don't assume I am a closed minded subjectivists who trusts his ears over reality.
Altering room acoustics during a session is entirely different than changing the height during a recording where the brain will(might) sense a tonal change. I think you missed the point of my statement, I am not saying carpet vs. no carpet will vary the perceived height difference and I am definitely not saying the source moves (captain obvious??) what I am saying is the difference in tonal balance between the ceiling and floor reflection will be enhanced. This would allow a reproduction room with the appropriate floor and ceiling treatments to ARTIFICIALLY create a sense of height which cannot be measured because it does not exist, it is an artifact of human perception.
It doesn't defy physics, you are ignoring the complex reflections both on the recording and the reproduction, and the nature of different speakers in different rooms with brains "interpreting" what they hear.
I have personally heard this a couple of times myself, let me explain the details:
first time, was with those little bose cubes (pardon my french) and I was listening to a test cd where the sound engineer used a clicker at a specified difference from two mics which he stated were some distance apart(I forget now) and he told you what height he was clicking at. Most of the elevations sounded identical until he said he was clicking 3 ft above the starting point which if I remember correctly was the height of the mics, at this point the clicking sounded like it was originating about 6-10" below the other clicks. So it was highly distorted and inverted, not a faithful reproduction by any means.
The second and third time were actually last week, I built some new dipole speakers, two 8" woofers and a Neo8 planar transducer, biamped, 24dB/octave electronic xover at 1200Hz. They are about five feet away from the wall and about 6 feet apart. They are in a corner only the "corner" has been cut off by a fireplace at an angle of about 45 deg with a tv above it in that cubby hole they put above gas fireplaces in apartments.
Now usually when I listen to music I can't define the height of the image, it isn't fixed but you don't really hear it moving either. But when I was watching a movie and some guy was speaking in a normal voice ie, no change in emotion the height sounded very clearly to be level with the top edge of the speaker ~12" below the bottom of the screen on a 32" 4:3 tv.
The third time was a different movie but no other changes, this guys voice was also clearly localizable but this time about 6" above the bottom of the screen.
What could cause this? Well it is an MMT dipole system so for a center image source containing info above and below 1200Hz there are 12 reproduction sources interacting with the room, 3 drivers per side front and rear radiation. As the spectral balance shifts downward the 8 lower sources dominate the room radiation and
this would sound lower and vice versa. The two 8's are about a 1/4" apart and the Neo8 is separated from the woofs by about 2"(frame edge to frame edge). Interestingly, if the woofs are treated as a single source(approximately) then the C-C distance is somewhat close to the perceived shift in height.... hmmm, might have a theory worth exploring here.
Sorry, no bookshelves to flip....
Obviously I have yet to perform a controlled test, since I am claiming some brain trickery here I don't need to. Rest assured I am going to
start thinking about how to test how c-c distances affect height perception and how one can alter a room to give a sense of height. Hey, even if it is fake and distorted imaging is cool.
Any suggestions...
nunayafb said:
It wasn't clear to me that you agreed with my earlier post, no.
Yes, speakers don't move, the point source is the same, never changes, therefore no "height" change is possible.
You tried to correlate to how we perceive height when listening to real world sounds acoustically , to how sounds occur from an audio system. The limiting factor is that in nature when the point source moves, it moves away in distance, somewhere in the hemisphere around our head, speakers don't.
I didn't miss anything (that's why they call me Captain Obvious).
What I get is you contradicting yourself.... on the one hand you clearly say altering room acoustics is entirely different than changing the height during a recording,....Then you say that we will perceive a difference in height because of tonal differences (being enhanced) by altering the acoustics of the room with (appropriate) treatments.
That contradiction there was pretty obvious....
BTW, what room treatment will ARTIFICIALLY create or exaggerate height? And how does that work?
Please elaborate. I am (again) most curious to the technical explanation on the mechanics, and (again) would be most grateful to learn this recording technique.
I am doing nothing of the sort (ignoring complex reflections, both on recording and reproduction).
In fact I have explained plenty on how sounds propagate (acoustically, and artificially).
I understand the theory very well, have a good grasp on the math, and have over 20 years of empirical knowledge on the subject of recording sound.
No one in this thread who claims that vertical height exists has given me any technical explanation how it exists in 2 channel audio, and how it is done. Why is that?
So let me ask you, why did all the other clicks stay at the same vertical height (despite them being recorded at other heights)?
Your speakers have the Neo8 (which I am very familiar with BTW), crossed at 1200Hz to a pair of 8" woofers.
The crossover point is quite low((which is great for imaging) from the Neo 8, The gap between it and the top woofer is a little bigger than I'd like....
In any event, large drivers tend to beam more as frequency increases, that coupled with the spacing (C to C) of the woofers, and the gap between the upper woofer and Neo8 could be placing the two different recordings (due to FR response (and formant) of the male voices being different) at a different physical height in the playback due to the different dispersion characteristics of the drivers. There could also be some lobing going on. I suspect that if your crossover point were higher, the effect might be exaggerated even more.
Try listening to those same recordings with a pair of single driver speakers, and I'm sure you would not get the same playback results.
An externally hosted image should be here but it was not working when we last tested it.
That's pretty bad **s Steve. Thanks for playin' ! :-D
Cheers
Daygloworange said:
You misread me: incrementally moving vertically - while staying the same distance away (ie proximity).
I'll try and find the CD. It also did an interesting (and similar) thing with a tom-tom beat. Pan R, pan L, OK heard from R then L speaker ... but then pan extreme R and extreme L, tom tom beat heard "outside" the L and R speakers!
tnargs said:
Trust me, if sarcasm were my intent, you would know it.
Your example is irellivant to the discussion at hand. Multiple omnidirectional sources in an uncontrolled acoustic environment with multiple reflection paths is a diversion.
Stick to the discussion.
tnargs said:
@daygloworange
Bounced the "vertical-positioning" concept off the guys on KVR. They suggest that this was done using binaural recording techniques/effects. Basically these are 3-d positioning techniques which incorporate the effects of the head-shadow and the shape of the ear pinna.
3-d effects can either be recorded with 2 microphones setup binaurally, or can be reproduced using plug-ins in a digital studio. Whether there are hardware effects for a traditional studio I don't know.
The effect only works using headphones where R/L ear sounds are fully segregated. There is a short wikipedia entry:
Wikipedia Binaural Recording
Cheers
Bounced the "vertical-positioning" concept off the guys on KVR. They suggest that this was done using binaural recording techniques/effects. Basically these are 3-d positioning techniques which incorporate the effects of the head-shadow and the shape of the ear pinna.
3-d effects can either be recorded with 2 microphones setup binaurally, or can be reproduced using plug-ins in a digital studio. Whether there are hardware effects for a traditional studio I don't know.
The effect only works using headphones where R/L ear sounds are fully segregated. There is a short wikipedia entry:
Wikipedia Binaural Recording
Cheers
I don't wish to cause pain but the distinction seems clear, you wrote "The purpose of the abx/dbt testing is to find small changes........"jneutron said:
I take that to mean 'finding out whether changing a cable changes the sound in the room'. I am saying the purpose (of the tests) is to find out if these "changes" are audible to humans. There may in fact be changes but we can't hear them (too soft) or there may be no changes but we could imagine that we hear some.
Well if the purpose of the tests is as you say then you would want to use instruments that you could trust ie they would not "adapt".
As you say
I agree absolutely with that. But to find out if some "absolute change" is audible to humans there is no alternative to using humans.
On your cable test I would use test instruments to find out if your changes to the cable caused any effects to the system before testing for audibility. Personally I think the changes that cables may make are not audible to humans. Time after time people who claim that they can hear these small changes are proved wrong. But of course there is always the possibility that there is someone who can and we just haven't found him or her.
cheers
Alan Hope said:
Yes I did, my mistake. Disregard my remarks in that instance.
Alan Hope said:
Yes, I've been exposed to binuaral recordings many years ago, and the stereo imaging can be astonishing. But it still won't give you true vertical height. Again, it can't.
Trust me, I wish it could.
I've also experimented with stereo mics with sound absorbing baffles in between them, and speaker setups with a wall that runs down the middle between the speakers to the listening position, in order to reduce channel crosstalk. They improve stereo imaging, it's a lot of effort, but it comes a lot closer to the effect of binaural recordings listened to through a headphone setup.
I have also done Q4 style quadraphonic recordings (many, many years ago. Again, a lot of effort, but the results are really impressive.
Alan Hope said:
In terms of acoustic sound capture with microphones without digital and/or phase manipulation tricks, you can't adjust height.
Software based effects for holographic imaging are something that I'm not interested in (at the moment). I think they still have a ways to go.
Cheers
@ daygloworange
OK. Er ... at risk of doing this one to death ...
Given that we have only 2 ears, do you think we can actually hear "height" at all? Why should a speaker near the ceiling sound higher up?
Assuming we keep our head level of course.
Perhaps we just contextualise sound height: ie a distant-sounding helicopter, or thunder, sounds instinctively "high" whereas vocals sound on a level, and perhaps tap-dancing (can't believe I am writing this) would sound lower down.
I can feel a double-blind test coming on 😀 😀 😀
OK. Er ... at risk of doing this one to death ...
Given that we have only 2 ears, do you think we can actually hear "height" at all? Why should a speaker near the ceiling sound higher up?
Assuming we keep our head level of course.
Perhaps we just contextualise sound height: ie a distant-sounding helicopter, or thunder, sounds instinctively "high" whereas vocals sound on a level, and perhaps tap-dancing (can't believe I am writing this) would sound lower down.
I can feel a double-blind test coming on 😀 😀 😀
Good point, if you look at an ear you will notice it is not symmetric about any axis, so when sound comes in from different directions it will sound different. Our brains learn to localize sound, obviously...
daygloworange....
It is clear to me now that you are just trying to "be right" without actually listening to anyone, you are repeatedly misinterpreting what is being said and stating 20 years of experience with recording and that is all that matters. You are disregarding how the brain localizes vertically while repeating the obvious things like speakers don't move and microphones don't record height. Your lack of effort to listen and learn and more importantly think outside your field of expertise is making this discussion pointless.
When you change height of the source the microphones WILL measure the difference in tonal balance for the reasons stated repeatedly, and since the speakers reproduce what the mic measures an open minded person would think about how that "difference" is heard by the two human ears which are adept at localizing sounds.
You clearly are not understanding this concept,
...recording technique, come on man I have already addressed that.
If you were actually thinking about my statements instead of just looking to debunk them you might see how I explained that.
Keep thinking like this and you might approach an understanding of what I am talking about.
And if you were objectively reading my posts you would have noticed that the first time I heard this was with small full range drivers.
With all due respect you are acting like the "objectivist" that the subjectivists get irritated with, you keep talking about how it is physically impossible to record height while ignoring how ears localize sound and the likelihood that you don't understand everything.
When we talk about cables objectivists say that all cables that measure the same wrt audio band transmission will all sound the same, subjectivists argued that they routinely hear differences in cables that measure the same. The claim that there exists some non measurable parameter that can only be heard in a familiar setting and that no testing can reproduce it. That is not the case here, the sound is clearly measurable, now we need to test how the brain interprets it.
Alan Hope said:
I think we can. We can certainly localize in 360 deg, around us. I'm not entirely sure about height though...We can localise some sounds better than others. And we can localise better or worse depending on conditions. As we age, our hearing becomes less acute, and I believe that it begins to deteriorate in late adolescence.
Understanding how sounds radiate from a point source is important. For the most part, they eminate in a hemispherical (omni-direction) from a point source. There are too many illustrations that show sounds dispersing as "arrows", as if sounds beam like lasers, which is wrong. It is easier to understand in a 3D model.
Microphones have what are called "polar patterns", which describes how they capture sound in the hemisphere around their diaphragm. It is an important aspect to know which mics to select for a specific application, and how to use ( and intentionally ) misuse them.
Our ears are 2 independant transducers that feed 2 independant channels to our brains. Sounds are processed independantly at out ears. There are differences in amplitude, FR, and time arrival of the initiation of a sound at each ear. There are also early and late reflections, echoes etc.. that we use to localise sounds in proximity and location.
The "attack" or envelope of a sound is very important as well, sharp percussive sounds are easier to locate, as well as sounds that are in the mid range (where our ear is most sensitive).
Low frequencies are much more difficult to localize. Apparently, we feel low frequencies through our body, much more than we actually "hear" them. IIRC, frequencies below 200 Hz are more felt than heard...
It's a fascinating topic.
I think the most important thing to visualize on the topic of imaging in 2 channel audio, is that (for the most part) speakers are fixed height point sources. To understand what I have been trying to explain, it's best to envision all speakers as full- range single driver speakers (without any of the physical limitations of course).
You cannot get "height" from L/R stereo no more than you can get stereo from a single (single driver) speaker. You can't have sounds come from in back of you either, with only 2 channels. Q Sound type effects excluded...
If you had 4 channels, as in L u/R u and L d/ R d (L up/R up, L down/ R down) with 4 speakers in front of you, then yes, you would have "height" as well.
That would be cool. It would be an awesome added dimension. You could exaggerate separation even more than real life, and get quite interesting mixing possiblities.
Hmmmmm.....
Cheers
- Status
- Not open for further replies.