Good Morrning all!

I just thought I would post this link to the Harbeth "designers notebook" regarding the correct sound stage images and how incorrect images can be mistaken for correct ones.

It also has some interesting comments on blind testing.

http://www.harbeth.co.uk/designersnotebook/chapter7/chapter7-1/index.php

Enjoy.

Cheers

DIYnut

Thanks for the site- good find!!

that was a good read.

dont agree with some of the stuff in the first little bit. stuff like "stage width extending beyond the speakers" is a bit of a misnomer.

i'd also like to know what he means by 'avg reverb time'. i'd like to know what he is trying to say there.

and lol @ "Harbeth speakers are balanced to provide this natural perspective no matter how far the listener sits from them"

Useful to have a graphic correlate to what we hear, but a few things I dont agree with or understand.

The term recessed I usually take to mean recessed in the midrange, and since vocals reside here this typically would mean that the singer, which is often dead centre, would appear further back in the soundstage. It seems they are talking about FR as they mention the BBC dip. But then their illustration suggests speakers which portray the centre images further back. What has this to do with FR? What of the vocalist standing on the left side of the stage?

I dont understand how one speaker can have better depth that another apart from its ability to reproduce low level ambient clues. What is it about a driver that could warp spacial perspectives as they suggest.

I've also found that spacial perspective seems to be more dependent on room and speaker positioning that speaker. All speaker, especially all dipole speakers I have used (provided they have plenty of room behind them) seem to image quite far behind the speakers, but I havnt heard a speaker yet that doesnt put a hard panned image closer to to the speaker in depth i.e if a vocal is coming out of only the left speaker it always sounds to me, regardless of speaker, that it is emanating from the plane of the speaker when all the other instruments are about 6 foot back, even if they are just inside the speaker, so not the concave soundstage they show but more a rectangle. Maybe Harbeths have special powers? Maybe its just a psychological thing from looking at the speakers, I should try listening to this with my eyes closed perhaps.

Would like to hear others thoughts re soundstage shape. It's something I take quite a bit of interest in when setting up speakers and experimenting with toe-in.

cheers

Bevan

interesting.

some of the other notes are good too

Good Morrning all!

I just thought I would post this link to the Harbeth "designers notebook" regarding the correct sound stage images and how incorrect images can be mistaken for correct ones.

It also has some interesting comments on blind testing.:eek:

http://www.harbeth.co.uk/designersnotebook/chapter7/chapter7-1/index.php

Enjoy.

Cheers

DIYnut

Oh yes, I always enjoy a good read on blind testing early in the morning. Preferably with a subtle wiff of napalm in the background to set the mood.....

Interesting article.

Oh yes, I always enjoy a good read on blind testing early in the morning. Preferably with a subtle wiff of napalm in the background to set the mood.....

Interesting article.

Allo , Mustard's back in town with a sprinkle of napalm to boot .

Hmm if we napalm your room can we put it back together the long way .

Useful to have a graphic correlate to what we hear, but a few things I dont agree with or understand.

The term recessed I usually take to mean recessed in the midrange, and since vocals reside here this typically would mean that the singer, which is often dead centre, would appear further back in the soundstage. It seems they are talking about FR as they mention the BBC dip. But then their illustration suggests speakers which portray the centre images further back. What has this to do with FR? What of the vocalist standing on the left side of the stage?

I dont understand how one speaker can have better depth that another apart from its ability to reproduce low level ambient clues. What is it about a driver that could warp spacial perspectives as they suggest.

I've also found that spacial perspective seems to be more dependent on room and speaker positioning that speaker. All speaker, especially all dipole speakers I have used (provided they have plenty of room behind them) seem to image quite far behind the speakers, but I havnt heard a speaker yet that doesnt put a hard panned image closer to to the speaker in depth i.e if a vocal is coming out of only the left speaker it always sounds to me, regardless of speaker, that it is emanating from the plane of the speaker when all the other instruments are about 6 foot back, even if they are just inside the speaker, so not the concave soundstage they show but more a rectangle. Maybe Harbeths have special powers? Maybe its just a psychological thing from looking at the speakers, I should try listening to this with my eyes closed perhaps.

Would like to hear others thoughts re soundstage shape. It's something I take quite a bit of interest in when setting up speakers and experimenting with toe-in.

cheers

Bevan

soundstage is almost entirely geometric. the FR can slightly alter a soundstage. boost 8-10kHz, u get more stage height etc etc, but it is typically only minor.

u have to go back to how humans determine where sound comes from. the article sort of hints at it.

there are three ways in which a person can determine the direction of sound

the Interaural Time Difference, Interaural Intensity Difference and the pinnae effect (ok, so there is four, the fourth being the precedence effect)

the Interaural Time Difference is the difference in time it takes for one wave to reach the closest ear and the extra time it takes to reach the furtherest ear from the sound source.

the Interaural Intensity Difference is the difference in intensity that occurs because the wave has to travel farther to the furtherst ear and therefore loses energy.

now, the brain couldn't possible understand whats going on if both interaural effects occur at the same time. in fact, we use the interaural time difference to locate sounds less than 1000-1500hz and more than 270hz and we use the interaural intensity difference to locate sounds more than 1000-1500hz.

so how does the brain do this?

quite simply actually, the ear relies on phase. when the delay between the ears is exactly a half period of the wave, then the information is ambiguous. if delay is less than a half period then we can detect the phase difference. if it is more than a half period we cant. why is this the case? because if the delay is more than a half period then there is more than one cycle occuring, therefore the information confuses the brain and it 'discards' the information.

the above paragraph makes sense, because in higher frequencies, there could be several cycles of the wave (think of a wave as a sine wave) and there is no way the ear could possible tell how many cycles has occured (unless it is retrospective, which the brain simply cannot handle).

so lets do a little test. put ur hand over one ear and close ur eyes. get someone to make a high pitched sound. turn around until u think u r facing where the sound comes from. open ur eyes and be amazed. how could u possible be facing in the right direction if there is no interaural intensity difference?

because of the pinnae. now the pinnae is the misunderstood little brother of the interaural differences. Blumlein (the guy that invented stereo) actually casts doubts as to how well stereo would work because stereo can only account for 2 of the 3 (the 4th is only important to the interaural differences and i wont go into it to answer ur question) methods we use to locate sound.

the pinnae acts as a little mechanical filter in the exact same way as ur room works only it is exceptionally complex. before the sound can reach ur ear canal, it must pass the pinnae. the pinnae causes the sound to reflect and superimpose itself so that peaks and troughs occur. now the only way the brain can use these reflections is through learning. as a child ur brain learns the direction from which sound comes from by actually looking at where the sound comes from. the brain then memorises the exact change in the FR caused by the pinnae and then the next time that FR occurs, then the brain knows where that sound has come from. the pinnae effects start to kick in around 1500hz or higher.

the difficulty for reproduction of music is that the pinnae and the Head Related Transfer Function (google it if u want to know. its not important for this discussion) are like fingerprints. everyone has their own individual set.

SO. how does a speaker influence the sound stage? well its FR can impact on the interaural differences and pinnae differences. BUT, how significant is its impact? well, not all that much really. the pinnae will cause anything up to a 30dB change in the FR. how much of a difference does a 3dB wobble in the FR of a driver make? a little bit but not that much.

all we have to remember is how does something alter how the 3 (or 4) methods of sound source location that the human brain uses.