PDA

View Full Version : Do audiophiles get enough exposure to 'live sound'?



kraiker
21-01-2012, 11:44 PM
I recently applied Bass Boost to 3dB and noticed that the Bass capabilities of ...

{Moderator's comment: This thread has taken a different direction to the original post by kraiker. You can read the original thread about boosting bass here (www.harbeth.co.uk/usergroup/showthread.php?1514-P3ESR-with-Leben-CS300-300X-300XS-bass-for-free).}

A.S.
23-01-2012, 09:27 AM
The CS300X is 15wpc and not 12. You are scarring me now with the bad idea thing, can you please explain ?OK, let's step back and look at this logically.

Have a look at the picture (attached) of a full orchestra. Imagine for a moment that the conductor allowed us to walk onto the platform and set-up a pair of mini-monitors (like the P3ESR) on stands and connect them to an audio system and then to play back a recording of the performance that we'd just made - and (hopefully) impress the musicians with the fidelity of our audio system. What would be the outcome?

Logically, we can expect this:

... that the reproduced sound would be tiny compared to the live sound. So tiny that the musicians seated around our speakers would be straining (and chortling) at our efforts to reproduce their huge dynamic range. Our sound would seem very small indeed compared to the live sound. There are several reasons for this -


The amplifier has far too little power (12-15W) when we would need perhaps 1000W (guess)
The speakers are inefficient (as all speakers are)
There are only two speakers trying to reproduce the entire orchestra of 60+ sound sources
The orchestra produces sound across the whole stage (10m x 5m?) - the speakers produce sound from two tiny boxes

So how can the listener be fooled into believing that, sitting at home, his tiny speakers and amplifier in a small room can faithfully recreate the hall sound when self-evidently it's impossible to generate enough acoustic power in the listening room? Obviously there is some mental trickery going on here - willing self-delusion - in the same way that we look at a widescreen TV and allow ourselves to believe that we are actually in the jungle or up a mountain.

What brings the sonic illusion to an abrupt end? The reality check is that this illusion of an orchestra in all its glory in front of us only works at a low replay level within the capability of the speakers. Look at the surface area of all the instruments of the orchestra (it must be many m2) and then consider that the P3 woofer has a surface area of about 1/3 of a sheet of A4 paper and you'll appreciate that it must be working really hard, moving backwards and forwards pumping air, to even vaguely create the illusion of the big, full sound of the real instruments.

Adequate bass realism at a moderate loudness is allowed for in the design of the woofer. But what isn't is intentionally electrically boosting the contribution from the bass-heavy instruments (turning-up the bass control on the amp) so that the small drive unit is being asked to work excessively hard. That will run the risk of bottoming the voice coil deep in the magnet (destroys the woofer) and/or over cooking the voice coil and/or much increased low frequency distortion and muddy mid band.

Hope that helps.

STHLS5
24-01-2012, 02:27 PM
Thank you for the explanation.

Now, I understand why some users' speakers always suffer from burnt voice coil every few years.

In view of the danger of bass boost, what's your view on room correction softwares? Could there be hidden dangers of bass boost when we use such system? My understanding of electronic EQ is to use it to attenuate and never to boost any frequency. But a DSP may not work similarly.

ST

A.S.
24-01-2012, 03:30 PM
In view of the danger of bass boast, what's your view on room correction softwares? Could there be hidden dangers of bass boost when we use such system? My understanding of electronic EQ is to use it to attenuate and never to boost any frequency. But a DSP may not work similarly.STInteresting question.

The point that I hope is now firmly accepted is that it is completely and utterly impossible for two 110, 200 or 300mm bass units to generate the sound of an orchestra at home. The only reason that home hi-fi works as well as it does is because it deceives the listener much of the time with an illusion that is just convincing enough that we're carried along with the experience. It's the same as being in the cinema: for the first few minutes we notice how soft and grainy the picture is, how colored the sound, how uncomfortable the seat .... but once our mind is absorbed with the whole experience, we lose the ability to be critical. We accept unquestioningly the nasty reality of the soft-focus grainy picture with its tiny optical dynamic range as faithful to life and are carried along with it. Also, how many audiophiles actually experience live musical sound to have a valid means of comparion? And as we observed with one recent A/B musical comparison here, only 10% of contributors recognised the original sound. 90% preferred a compressed 'soft-grained' defocussed, analogue version of digital reality.

As far as I know, no attempt is made (or should be made) by a room correction system to boost weak notes in the bass. The fact that there are some or a series of musical notes which are a little or a lot less audible than adjacent notes when the loudspeaker is producing them as soundwaves of equal pressure indicates one thing: some sort of cancellation is occurring in the room. If you play the musical scale and certain notes are completely inaudible, this has something to do with the dimensions of the room, not the speakers. And the key point is this: if the notes are inadible because of the cancellation of soundwaves opposing each other, no matter how much sound power the speakers generate there will always be cancellation. So if bass notes are missing due to cancellation, EQ cannot ever fill them in, but it can tame those frequencies where there is an excess of bass by applying less power. That's my understanding.

Takis
24-01-2012, 07:28 PM
Interesting question.

... The only reason that home hi-fi works as well as it does is because it deceives the listener much of the time with an illusion that is just convincing enough that we're carried along with the experience. It's the same as being in the cinema: for the first few minutes we notice how soft and grainy the picture is, how colored the sound, how uncomfortable the seat .... but once our mind is absorbed with the whole experience, we lose the ability to be critical. We accept unquestioningly the nasty reality of the soft-focus grainy picture with its tiny optical dynamic range as faithful to life and are carried along with it....

Excellent!
This is the "break-in period" for every hi-fi component, that so many audiophiles believe...

Brain adjustment...

A.S.
25-01-2012, 08:09 AM
Excellent! This is the "break-in period" for every hi-fi component, that so many audiophiles believe...

Brain adjustment...100% correct. There is no real physical or electrical 'burn-in' (aside from a few hours of softening of the speakers suspension). What's called 'burn-in' in audiophile circles is entirely a mental issue of becoming acclimatised to a new experience. Of that I am absolutely and utterly certain.

Failure to accept this as fact demonstrates a basic lack of awareness of how the human mind works in all areas of observational experience, from engineering to love.

STHLS5
25-01-2012, 11:35 AM
.... but once our mind is absorbed with the whole experience, we lose the ability to be critical. ...

Yes, you said it well. Once we are absorbed with the music we are not only non-critical but also shouldn't be critical. It defeats the purpose of listening to music.*


*Also, how many audiophiles actually experience live musical sound to have a valid means of comparion?*

That's what I have been telling. We (some of us like myself) just do not know what a live unamplified sounds like in a proper venue. We do not know how a real high quality violin, piano or a woodwind is suppose to sound. If I am a musician and my sound reaches 90% of the listeners would I want change my music to be technically correct so that I meet the standard which is only recognizable by 10% of the listeners?

Some used to say the guitar sound in my room is like coming from nylon strings. i know they are referring to reverbs so I covered about 30 percent of the absorber in my room with aluminum sheets. The results was a "real" guitar sound even though it is colored with reverbs and I do not like the vocals due to the high reverbs but to others it was perfect.

Honestly, despite being pointed out to me about the *reduced reverbs in the clips I still cannot hear the difference. Is it because the music played in a venue that I have no prior reference. *Can a westerner tell if a veenai or a sarod or a gamelan's sound is correct or you could readily hear the trailing reverbs? But we live in a world where Louis Armstrong's voice is gift from god despite normal humans never sound like that.

Just my random thoughts .

ST

p.s. I have no idea why the asterisk appears in my posts. Must be something to do with my IPad.

A.S.
25-01-2012, 12:16 PM
I suggest that we move this topic to another separate thread as we are some way from the original theme of this one.

I've duplicated the last few posts so we can pick-up again with ease.

Discussion continues right here in this thread. You are in the right place to comment!

keithwwk
25-01-2012, 12:29 PM
I agree run-in = brain adjustment.

The more incorrect or distorted sound the more time need to "run-in".

My audiophile friends, non-Harbeth owners, always tell me how importance warm up or run-in and practise 24/7 switch on hifi in order to keep the sound in optimal quality.

For me, my Harbeth setup can sound natural and beautiful right after I switched them on and play music. The superb natural of RADIAL cone do not need any brain adjustment at all.

A.S.
25-01-2012, 02:02 PM
Yes, you said it well. Once we are absorbed with the music we are not only non-critical but also shouldn't be critical. It defeats the purpose of listening to music ... We (some of us like myself) just do not know what a live unamplified sounds like in a proper venue. We do not know how a real high quality violin, piano or a woodwind is suppose to sound.

Some used to say the guitar sound in my room is like coming from nylon strings. i know they are referring to reverbs so I covered about 30 percent of the absorber in my room with aluminum sheets. The results was a "real" guitar sound even though it is colored with reverbs and I do not like the vocals due to the high reverbs but to others it was perfect. ... Is it because the music played in a venue that I have no prior reference. Can a westerner tell if a veenai or a sarod or a gamelan's sound is correct or you could readily hear the trailing reverbs? ...Very thoughtful. Where to start commenting?

First, about the guitar sound: we the humble listener have no means of knowing how bright the original guitar sound was in real life, since there are many factors to consider about the guitar body, the strings, the positioning and selection of the microphone (very critical for recording strunged instruments), the room acoustic, the amount of equalisation that the producer/record company executive/marketing department/artist/artist's manager desired. Far too many unknowns.

We can say is that is is likely that for classical music featuring the acoustic, unamplified guitar with/without an orchestra that the guitar will indeed have nylon strings and should sound like a guitar with nylon strings i.e. a rather soft, warm, lacking-in-obvious-high frequencies* sound. So if your ill-informed friends were listening to classical guitar sound and expecting it to sound like bright, crisp pop guitar - more fool them. They sent you on a useless chase. If they were listening to a pop guitar recording, their oppinion was worthless because, for the reasons I listed above, neither you, nor they, nor I have the slightest knowledge of the tricks and techniques used to make the recording.

A case of the ignorant leading the willing?

I'll look out a guitar recording I made last year in a moderately dry acoustic. You'll be very surprised just how un-hifi like it is. Very very dull sound. But that's exactly how this lovely old guitar sounded in real life. With some EQ I can make it sound as bright and zingy as you want. In fact, although my original idea was to release the recording of guitar + singer as an 'audiophile recording' I didn't think that it had the "hi-fi" sound that audiophiles would recoginise as "hi-fi". So I abandoned the idea. Your friends would consider it worthless for sure!

STHLS5
25-01-2012, 03:33 PM
?.......
I'll look out a guitar recording I made last year in a moderately dry acoustic. You'll be very surprised just how un-hifi like it is. Very very dull sound. But that's exactly how this lovely old guitar sounded in real life. With some EQ I can make it sound as bright and zingy as you want. In fact, although my original idea was to release the recording of guitar + singer as an 'audiophile recording' I didn't think that it had the "hi-fi" sound that audiophiles would recoginise as "hi-fi". So I abandoned the idea. Your friends would consider it worthless for sure!

I remember the recording. I watched the live streaming and yes it sounded typically unamplified, i.e., flat. I always felt the same thing about some unamplified recording that I have.

The other thing I like to comment is the vocal recordings. Have you listened to your children singing several feet away and singing softly to your ears? I notice there is a difference in the sound texture. They are not the same. Similarly, a microphone placed few inches away from the mouth may pick a different sound characteristic of what we hear several feet away.

Thanks to our adaptability we are able to recognize the voice just like recognizing your loved ones voice over the telephone but as a designer this would be a nightmare to exactly determine the exact voice at the exact distance.

So many variables and unknowns and so many things to learn. I hope we could hear the difference in the A/B clips soon. Either I am not an acute listener or my brain just learned to filter out the excess reverbs.

ST

A.S.
25-01-2012, 04:30 PM
... Have you listened to your children singing several feet away and singing softly to your ears? I notice there is a difference in the sound texture. They are not the same. Similarly, a microphone placed few inches away from the mouth may pick a different sound characteristic of what we hear several feet away....It shouldn't be a surprise that especially in doors, the further the source is from you, the greater the difference in 'character' between the close-up sound and the more distant sound.

Let's take the human brain out of the equation and rig up a dumb microphone, that captures whatever presses on it's diaphragm and without a brain, just passes that along to the recording device as a voltage. Assume that we are using the perfect microphone to capture those voices, and that microphone of itself did not change character* according to how close it was to the source. So we put the mic on a stand** and ask the children to speak or sing at (say) 20cms, 50cms, 1m, 3m, 10m from the mic. What would we hear from the dumb microphone? (Actually I think I did this very thing when the children were young .... wonder where the DAT tape is).

* Some microphones change sonic balance dramatically depending upon how close the voice is. The warm, velvet tones we remember Bing Crosby for were much to do with the reports that he would only use one type of mic and that he even carried his own mic to recordings. "The ribbon’s natural sound can also be made to sound warm, big, and syrupy (Bing Crosby-like) when placed within two or three feet of the talent (generally, you can’t close-talk a ribbon without having a greatly exaggerated bass characteristic)" - more here (http://www.coutant.org/ribbons.html).

Microhones have their own character. The rich, chesty, full bodied, relaxed he-man 'all-American sound' has much to do with ribbon microphones used close-up (http://www.wired.com/rawfile/2011/01/gallery-ribbon-mics-part-1/). Example here (http://ia600508.us.archive.org/17/items/murrow_in_london_1942/murrow_in_london_64kb.mp3) from 1942. Why wouldn't you set that as your normal voice reference sound even though it was a (pleasing) coloration of the real sound?

** Funnily enough, at the X-Factor studio last week (another M40, M30 install) they told me how after a beauty parade of all the well known monitor brands how they came to select Harbeth. Their test was to place a well-respected Neumann U87 microphone down on the studio stage (where we see the presenter) and cable it back to the control room where they could listen to various voices. The only speakers that captured the naturalness of the live voice was the M40.1. I asked if the engineering staff would be willing to make a video of behind the scenes including repeating that test and they have gone off to seek permission and cost making a mini-documentary for us here.

A.S.
25-01-2012, 06:21 PM
Examples of Bing Crosby and how the selection of the microphone can give you a specific sound: the warmth we associate with his singing voice. And how many of use would be likely to meet and hear him in the street, unamplified. So our preception is inevitably that of the recorded voice.

Bing Crosby Audio Vault (http://bingcrosby.com/bing/bing-crosby-media/audio-vault)

Have a listen to the first clip 'Capital radio interview - UK'. It opens with the familiar warm recorded voice we know. Then follows a studio interview.

When he visted the UK in 1975 he gave a radio interview from which you can hear his real speaking voice at a typical studio voice to mic distance of about 30cms. The interview microphone at that time would almost certainly have been an AKG D202, a cardiod mic optimised to sound and measure flat when used close to the source.

The professional sound engineer will carefully select a microphone for the effect he wants. We, the listener, are not privy to that selection criteria so when judging a recording we are really unable to be sure about how true to life it really is.

A.S.
26-01-2012, 09:55 AM
Before your smart friends start to nit-pick your hifi system, you should be armed with some basic skills for putting them in their place. The most important skill to develop (in my opinion) is to be able to mentally deconstruct a recording. What does that mean? Since neither you nor they were actually present at any recording, and we can't really trust any publicity pictures supplied with the disc, we are going to have to listen forensically to the recording and decode what we hear. We are going to have to deduce, solely based on our knowledge of recorded sound, how big the recording environment was, where the performers and microphones were positioned in that space and even what type of microphones were used. We've touched on this so many times, even with recorded examples (here and here) but not reached the core. So let's develop some skill to do just that.

We don't need to know anything at all about music. We don't need to be able to read music (I can't) or play an instrument (I can't). We don't need to know anything about eastern or western musical scales, instruments or tonality. But we must know something about the art of recording, about microphones and recording halls and how they are used to artistic effect.

Microphones fall into two main categories: acoustic microphones set away from the instrument and contact microphones which are firmly glued or attached to the instrument. At the very least we must be able to recognise the sonic signature of these very different ways of capturing the sound we experience on the recording. The sound is totally different, and we can identify them by careful listening.

The contact or very close microphone: an intimate sound

Here is an excellent DIY video (http://www.youtube.com/watch?v=aOJuCYgmPPE&feature=list_related&playnext=1&list=SPDE23FAC8A681FA46) showing how a $1 crystal microphone (no moving parts) can be attached to a sound generator (guitar etc.), amplified, fed though some pleasing tone adjustment ... and bingo: a great rock guitar sound. You can even make the crystal microphone yourself with some simple chemicals! Remember: it has no moving parts. Cooking instructions here (http://www.youtube.com/watch?v=K3G2QM5a-9U&feature=relmfu).

Point to note is that a contact mic picks up nothing of the acoustic space around the instrument. The sound is therefore acoustically 'dry' with much higher intensity than a conventional mic set at a distance. Hence, using the contact microphone is the normal way of recording or playing pop/rock instruments where an intimate sound is required and crucially, performers can be completely isolated from other performers who have their own contact microphones*. This is the exact opposite situation with the recording of acoustic music (orchestra etc.) where performers are not contact-mic'd, and where the sound waves from instruments are allowed to float in the acoustic, absorbing the ambient character of the hall, before being collected by distant microphones. That's the general idea. More on that in another post.

* I must stress this point fully. The entire pop/recording industry is underpinned by the rule 'one instrument, one microphone, one channel on the mixing console (http://www.oceanwayrecording.com/studios-nashville.php)'. That's because of the need to do re-records (re-takes) of individual performer's contributions. That implies that the recording engineer can isolate and get-at one performer's contribution (http://www.youtube.com/watch?v=IzuGdPvpiPA&feature=related) and replace or edit it without disturbing any of the other performer's acceptable contributions. That absolutely mandates that each performer is recorded in isolation and that means, their microphones must be very, very close to their sound source (instrument) or actually attached to it. So, the sound captured by these close-microphones will drown out any sound leaking from other performers. This type of dry sound is not what we hear in the concert hall.

TimVG
26-01-2012, 11:40 AM
The amount of variables which are part of how a recording sounds the way it does is too much to put in words. It's a good thing that a Harbeth is very true to the original source, a great reference speaker.

It's stunning how natural they sound even if you're sitting very closely to them, you don't hear the seperate drive units. With all other brands of speaker, when listening in the nearfield, I hear a tweeter and a (foggy sounding) mid/bass unit.

Macjager
26-01-2012, 12:38 PM
Interesting information about the contact microphone, yet many guitar players who record prefer that their guitar be hooked to an amp and then have the amp mic'd so that the more open sound goes to the mixing board.

Many artists like to have their guitar mic'd on stage, the same way so that the sound engineer can blend the sound with the rest of the music. It is only after getting involved in home recording (still not ready for prime time) and doing much research that I learned much about the various recording techniques.
Many recordings are done one insturment at a time, (the rest of the band does not even have to be there, thus the isolation, even with a mic'd amp, you can have the one:one:one ratio as noted). Also probably why many artists sound different while live vs "memorex"

STHLS5
26-01-2012, 12:51 PM
Thank you Alan. You have explained very well how you approached the Harbeth designs and how we should judge the sound. Btw, the "here and here" is without a link.

Some speakers are inherently coloured that it deprives listeners of appreciating the minuscule difference in the vocals. I have often heard remarks from my friends that some of the vocals in the recording that they are familiar with do not sound right to them.

One example was when I played a CD of Ella Fitzgerald the listener couldn't identify the singer. After he confirmed that he has the same CD, I switched few other tracks and in some he could readily recognizes Fitzgerald.

I have noticed the vocals sounded different from one track to another in my Harbeth but in the non Harbeth system it is somehow have a unique sound signature that it masks whatever difference that you may have despite one track could be a recording of her voice in her younger days. An acoustically clean room amplifies the difference more.

ST

A.S.
26-01-2012, 01:55 PM
... yet many guitar players who record prefer that their guitar be hooked to an amp and then have the amp mic'd so that the more open sound goes to the mixing board....Two points here ...

1. Whatever the artist may prefer about creating some acoustic space around their instrument, you can 100% bet that the record producer/sound engineer will do everything in his power to persuade the artist that is a very bad idea indeed, and a 'better' solution (that is, a better solution for the production team) would be to record very closely and then add some artificial reverb in the control room.

Can you see why the last thing the engineer wants is a nice, spacey air around the instrument? Because if the performer is playing 'to the room' either with a mic some way back from the instrument and/or playing through contact mic and PA amp speaker, the sound is splashing all around the studio and will be picked up by other musicians microphones. If he plays a wrong note, coughs or sneezes, becomes confused and loses his way or gets out of sync with the others, there may be insufficient acoustic isolation to be able to do any remedial work (i.e. edit, re-record) on his peformance alone because the original mistake will have already polluted the other performers recordings via their microphones. See how that could lead to a very, very expensive re-recording session with the entire band? And that's why multitrack recording was a godsend to accountants when it was developed in the 60s. They embraced it with even more gusto than recording artists. Because it saved money by minimising re-recording costs. Everyone else could go home except for the artist who needed to re-record his part and the sound engineer; he just played along with the original recording on headphones. Cheap and effective. But the down side is no air in the recording. So if you see recording artists wearing headphones during a recording, you can be fairly sure that it is a multitrack (isolated, airless) recording and they are just being fed whatever instrument they need to keep time with. They will not hear the entire band because the 'entire band' doesn't exist until it is synthesised out of the individual, airless, performances during the mix.

2. Relating to the above .... just because the instrument is playing over a PA speaker doesn't change the abservation about the missing air from close miking or contact miking. The mic will be jammed right up next to the speaker and intentionally not colecting the ambient acoustic for the reasona above in 1.

Listen to what you can do when you have all the performers/sound on different, isolated tracks here (http://www.youtube.com/watch?v=_vz_PU2DC8A&feature=related). Notice how on the second (clap) example the reverb aroud it is entirely synthetic. Also notice at the end the extreme convenience of being able to re-record only the vocals without disturbing any of the other sounds.

Here's something you don't hear these days: the band members all recording together in the same acoustic space at the same time and confident in their synergy (and musical ability) that they don't need to be isolated frome each other. Lovely open sound from the Beatles, 1965 (http://www.youtube.com/watch?v=6OWrlVT6vF8&feature=related). Also here (http://www.youtube.com/watch?v=TVX27TSPP7U&feature=related).

Interesting comparison with the previous links showing a modern mixing desk with hundreds of channels. Here in 1964 (http://www.youtube.com/watch?v=nFbArd5MxtE&feature=related), the Beatles and a view of the control room with just a few microphone channels. Implication: the four performers recorded in the same room, playing together in the same take. So, plenty of common acoustic is collected by the open microphones. Those days are long, long gone. Only the very best musicians who are completely in-tune with each other dare to record this way. Another wonderful example here (http://www.youtube.com/watch?v=JQWkNV2Nd3s&feature=related). And if you listen closely to this example of Paul and John singing (http://www.youtube.com/watch?v=kzgv9iIGSrE&feature=fvwrel) in harmony you can hear the guitar track which is fed to their headphones actually being picked-up by their vocal microphone(s). Imagine how small a sound that is and the true difficulties of recording many perfomers in the one space with isolation between them. (I wonder if they actually sang simultaneously because there is L-R pan and the picture, perhaps unrelated, shows a mono microphone. Did one recorded the vocal track first, the tape was rewound and the other sing along to it in harmony?).

>

Pluto
26-01-2012, 07:57 PM
...I wonder if they actually sang simultaneously because there is L-R pan and the picture, perhaps unrelated, shows a mono microphone. Did one recorded the vocal track first, the tape was rewound and the other sing along to it in harmony?

Both statements are possibly true because there are at least four parts here; it seems reasonable that they sang together onto two or three mono tracks to create the ensemble and thus create the stereophony.

Abbey Road (the album) was created using 8 track recorders although Abbey Road (the studio) was a comparatively late adopter of this technology.

A.S.
27-01-2012, 10:44 PM
I have a really perfect example of the deconstruction of a pop song: Queen's guitarist, Brian May, talks us through Bohemian Rhapsody (http://en.wikipedia.org/wiki/Bohemian_Rhapsody).

There are several very important points to note about this famous 70's recording.

1. The use of multi-track recording. Individual sounds are recorded to individual channels and then mixed-down in the desired relative loudness by the producer after all sounds have been recorded. That's called post-production, and will take far longer than the recording itself. This one track alone could have taken weeks to mix-down due to its complexity, but none of the performers need to be present for the mix-down.

2. It is absolutely essential that all the individual (microphone) channels are recorded dry - that is, without picking-up much of the studio acoustic or the sound of any of the other performers, just as we noted with the Beatles example. That's because at mix-down it is crucial that there is no unwanted reverberation or ambience because it cannot be sonically removed from the recording and may spoil the intended artistic effect the producer desires at mix-down. That in turn implies that the individual sound are close-miked to minimise bleed-through from other sounds. This close miking will give a high-intensity sound and little or no air about the sound.

3. Note that at about 25 mins in, Brian introduces the idea that for some of the drums, they wanted a more airy sound, and intentionally pulled the mikes back from them. But Queen made that artistic decision at the time of recording, because it would not have been possible (even with today's technology) to remove the ambience from those tracks and dry-out the sound. Had they regretted introducing ambience to the instrument at recording time they would have had to re-record the drums as a separate exercise days or weeks after the main session. The cost and inconvenience would be dramatic - and the drummer may have been away on tour. So the general rule for pop music recording is - Always record 'dry'. Avoid capturing 'air' around the performers. Record in proper, acoustically dead studios with mikes up close.

4) Just listen to how dry the various tracks are - devoid of any ambience. Completely the reverse of classical recording where the a sense of the hall and an ambience around performers is highly valued.

This one video video alone (http://www.youtube.com/watch?NR=1&feature=endscreen&v=Z85YsUAU6pA) encapsulates the art and reality of pop recording. Worth watching from start to finish.

It should be obvious that the creation of a pop/rock song cannot be spontaneous and involved hundreds of hours of careful adjustment. Because the end product is synthetic the question to ask yourself is this: As I will never be able to hear that exact performance in real life, is there a frame of reference to judge sound quality against? Should pop/rock music form any basis of my critical evaluation of high fidelity recording or playback equipment? The answer to both must surely be an unequivocal no.

You can here Bohemian Rhapsody in its final mixed-down version (http://www.last.fm/music/Queen/_/Bohemian+Rhapsody) here (moderate quality). The CD is very nice indeed.

A.S.
28-01-2012, 09:14 AM
OK, before we move on to a forensic analysis of acoustic music (which we can use as a tool to reveal the latent abilities of our high fidelity systems), are we comfortable with the idea so far?

That is, if the air and atmosphere around musicians is the arbiter of great hifi that we must consider the approach to the recording itself before we chase realistic and achievable dreams of greater realism from any playback system in the studio or at home

STHLS5
28-01-2012, 09:41 AM
I hope I could speak for the non contributors but the two posts above require some understanding and I am still trying to understand your explanation about the nylon strings but just couldn't edit the samples that I have in mind to include in my earlier response. The last two posts are bit too advance for me but I may not be the average representative of your larger audience.

ST

HUG-1
28-01-2012, 10:31 AM
Appreciate your feedback. It is vital we convey the point Alan is making. Do others need help with this "close-miking-no-ambience" concept? If so please tell us. We have to find a way to explain this core element in hi-fi sound.

STHLS5
28-01-2012, 11:35 AM
..
We don't need to know anything at all about music. We don't need to be able to read music (I can't) or play an instrument (I can't). We don't need to know anything about eastern or western musical scales, instruments or tonality. But we must know something about the art of recording, about microphones and recording halls and how they are used to artistic effect.

Many threads followed after this statement to further explain your point about recordings and high-fidelity sound.

My point is you may be judging that many music lovers know what is good sound. We have seen many examples of demonstration based on Western Classical music which I know for a fact in this part of the world many do not have experienced it at all.

I included a short clip of snare drum being played in a loop. Which of the parts the readers think to be a correct representation of snare drums heard in real life?


http://www.youtube.com/watch?feature=player_detailpage&v=l0oaMT1IayE

ST

{Moderator's comment: But surely Alan is doing his very best to show you by example what he means. What more can he do? What is the problem?}

Pluto
28-01-2012, 12:06 PM
I included a short clip of snare drum being played in a loop. Which of the parts the readers think to be a correct representation of snare drums heard in real life?
Sorry, but I really don't get what a loop of a snare drum being whacked as hard as possible is supposed to demonstrate or prove.

A.S.
28-01-2012, 12:16 PM
Sorry, but I really don't get what a loop of a snare drum being whacked as hard as possible is supposed to demonstrate or prove.Nor do I. This seems like a distraction. But clearly ST thinks that this is most relevant so we must respect that. But personally, I need some more input as to what this clip is supposed to illuminate. It sounds like an electronic drum machine with a fake reverb to me. We could prove that by looking at the waveform in an audio editor.

We have been round this ambience/no ambience loop several times before in what I thought was minute detail. Clearly I am not breaking through. Would you please tell me what it is about my analysis that is indecipherable so I can see if there is a building-block missing. I'm more than happy to fill the gap.

It is impossible - totally and utterly impossible - to have the self confidence and credibility to comment on cables and the like unless the absolute fundamentals of deconstructing recorded sound is firmly in place. The only skill required, as I said earlier, is to make time to listen carefully with a truly open mind again and again and again until you sensitise your ears. That's how I learned about sound and everything I write about here. Simply through careful and curious listening.

STHLS5
28-01-2012, 12:47 PM
Thanks for the response. You have just shown that without a reference you are having difficulties in telling a dry and wet sound apart. The point that I wanted to make is if i were to add another short note of a strange instrument such as a didgeridoo (coming soon, if it is ok with Alan?) i wouldnt be surprised if we have users who cannot tell the dry and wet note apart unless I include the the trailing reverb.

But reverbs also change the orginal sound itself.

I have no problem of hearing the reverbs in the previous wet and dry demos. But when I and many failed to recognize the AB samples earlier I think we have to seriously relook our approach. I keep on asking why? And many people who spend their time in cables, speaker stands, spikes and cones should also do the same.

ST

EricW
28-01-2012, 01:54 PM
...This seems like a distraction. But clearly ST thinks that this is most relevant so we must respect that. ...

...Would you please tell me what it is about my analysis that is indecipherable so I can see if there is a building-block missing. I'm more than happy to fill the gap.



Actually, I thought your analysis was extremely clear and easy to grasp. It couldn't have been more straightforward.

I'm not sure ST is being entirely serious in his protestation. I notice that he keeps making the same point that many listeners in his part of the world may not be familiar with the sound of live Western classical music. Even if that's true (and I'm not sure the percentage is much lower than it is in Western countries, where it's also not exactly something everyone does), so what? You may use orchestral or classical pieces to illustrate, but the point of the exposition - I take it - is usually to demonstrate something about how a natural sound source (an unampified musical instrument, could be a sitar or an erhu as easily as a violin or a guitar) combines with an acoustic environment in some way. The surface attributes may be "Western" (though again, there are plenty of classical music listeners all over the word), but the scientific essence of the idea is universal.

I suspect that ST is understanding you perfectly well, but is trying to make some other kind of point.

However, I would like to pose a different kind of question. The Bohemian Rhapsody video was brilliant, and you convincingly use it to make the point that one can't judge the ultimate quality of a loudspeaker by using that kind of heavily processed, "unnatural" (though perfectly valid artistically) sound. And yet .... won't even a multi-tracked recording like Bohemian Rhapsody still sound better through a pair of Harbeths than just any old speaker? If so, why is that the case?

{Moderator's comment: and we know that western classical music is taught in China (and many other eastern places) and played to a really high standard there. So there really is no east/west divide on this issue.}

STHLS5
28-01-2012, 02:29 PM
Cont/

Let's look at a bicycle's bell. We are very familiar with the sound. How the sound changes in the garage, in the rain, on the street and when you ride in the field.

You have a general idea of a bicycles bell sound and you may have a place where the bell sounds the best. That will be your reference.

Now, let's say a conductor decides to use the bell as part of orchestra performance just like one conductor who used an animal bone, how would one describe the bell sound? That will be a sound that you or I never heard in our childhood. Can we say that should be accurate and natural just because it was played and recorded in the best hall?

Wasn't it in the 50s or 60s when Ravi Shankar the sitarist played in the US for the first time, the sound of sitar was described as screeching sound of a cat. Well, things changed after that but what matters here is the first impression. I have no problem with any kind of music but the very important point I would like to make is we shouldn't learn to recognize the correct sound.

We as Harbeth users instinctively know the natural vocal sound but our (ok, my problem since no one is going to admit) we do not readily know the natural sound of musical instruments. There maybe no such thing. It is is not how natural or correct the instruments sound but how good the music is.

ST

A.S.
28-01-2012, 02:48 PM
Re: Snare drum

I'd like to look at this more closely. First, what we were presented with is an audio-in-video clip of unknown quality and origin. Retrospectively I think that this was a more sophisticated example than it appeared to be on first play through. I assumed that the alarming glitch at about 5.8 seconds in the sound where both the tempo and pitch change was some video editing malfunction and as a result, I mentally discarded the jump and switched off at the edit point. Now I've actually extracted the audio from the supplied clip - below - it's obvious that this clip is assembled from two 'takes' either intentionally or very poorly joined at about 5.8 seconds. Examination of the waveform shows how obvious the transition is: see attached.

Am I reading too much into this clip? Was the intention to convey something different in the acoustic of the front and back edits? But what can be meaningfully deduced from what are, judging by tempo/pitch alone, two different recordings possibly made months apart in different studios? I'm really baffled.

>

A.S.
28-01-2012, 03:00 PM
... I have no problem with any kind of music but the very important point I would like to make is we shouldn't learn to recognize the correct sound.

We as Harbeth users instinctively know the natural vocal sound but our we do not readily know the natural sound of musical instruments. There maybe no such thing. It is is not how natural or correct the instruments sound but how good the music is...I profoundly disagree with you from the bottom of my soul. To indulge in the high-fidelity game and then (collectively) to pontificate about audio minutia it is not good enough to say that there are no natural or correct sounds and to solely indulge in the romance or emotion of the tune and experience. That's the fast track to audio hell, via ever more compression, more EQ and coloration and smaller, nastier audio equipment. You cannot mean what you have written because if you truly believe that, you have wasted your money buying quality audio equipment. You could have reached the same level of satisfaction with a mini supermarket hifi system - surely?

Just because you (or I) are not familiar with the live sound of some obscure instrument, there is no excuse whatsoever for not turning off the hifi, putting on a clean shirt and actually going out and finding someone playing a conventional instrument in public. Combine it with a drink in the lounge bar of many better hotels where you will be entertained by a pianist or violinist for free every night of the week.

So what do you really mean?

EricW
28-01-2012, 05:33 PM
Also, even the accurate reproduction of voice is a big achievement, as I've discovered.

I have two pairs of speakers, one Harbeth, one not (purchased pre-enlightenment).

When I bought my P3ESRs, I naturally put them in my study which is also my listening room, for serious music listening, and moved the other speakers (well-regarded, well-reviewed, not cheap) out into the leaving room next to the TV, for AV use and more casual music listening (background music when having people over for dinner, for example).

I've since switched and put the Harbeths in the living room. Here's why. Although the other speakers are enjoyable for music listening, I found that I was struggling to hear the dialogue clearly in movies or TV shows: there was always a strange, phasey coloration right in the vocal band that seemed to affect intelligibility. Turning up the volume (my first response) didn't really help. At first, I put it down to poorly-recorded sound, compression on the DVD, whatever.

But then I thought (prior to playing a film that had dialogue I'd actually want to hear) hang on, let's try the Harbeths in the living room. Bingo! Suddenly voices are perfectly clear, sound exactly like voices, and are clear and intelligible even at low volume levels. I can relax and enjoy the film or show, without strain. Very nice.

So don't discount the important of even just making the voice sound right. It makes a profound difference.

STHLS5
28-01-2012, 05:54 PM
Rather straight-forward samples retrieved from recording made using a Crown SASS-P stereo PZM microphone (http://www.sweetwater.com/feature/technotes/issue28-reverb/). Snare drum dry here (http://www.sweetwater.com/feature/technotes/issue28-reverb/mp3s/SNARE_dry.wav)and wet of Trinity Church here (http://www.sweetwater.com/feature/technotes/issue28-reverb/mp3s/SNARE_IR3.wav). It was edited to snip away the trailing reverbs. The point I was making, that without the trailing reverbs one cannot be certain of which sound supposedly be the true representation of the recorded sound of real instruments more so if you do not know the real sound.

So how are we to know the real good natural sound of instruments? I emphasize instruments NOT vocals. Recognizing the natural voice in recordings comes naturally. I need not have any orchestra hall recording to know that unlike instruments.

Long time ago Alan made a remark about an edit in the Take Five recording where the surroundings changed. He expressed his surprise than no one actually picked it up. Shouldn't we be asking why? Were we too drowned in the music and stop being critical? Or should we as ordinary music lovers noticed that? What level of alertness is required in us while we are supposedly be enjoying the music?

I think I have caused enough distraction for today. Hopefully, others would step forward and critically *look at these issues. Thanks Mod for tolerating me this time.

ST

A.S.
28-01-2012, 06:21 PM
... The point I was making, that without the trailing reverbs one cannot be certain of which sound supposedly be the true representation of the recorded sound of real instruments more so if you do not know the real sound....OK I went out for a walk with the family to the local giant scone tea room (refreshes the parts other scones cannot) and mulled over where we are at and where the sticking points may be. I think we are slightly at cross purposes, and the gap between us has actually been long standing despite various audio clips and explanations I've given over the years.

What it boils down to is this:

We all know that a supermarket audio system and a quality audio system (let's call it a high-end system - just to differentiate it) fundamentally do the same job but at perhaps a 1:100 price range. They both generate sound waves from a recording, modulate our ear drums and provide a sensation in our brains. Even the most hard bitten audiophile can sometimes have an enjoyable musical experience with a portable radio (I can), or in his car. If we are totally honest we know that our partners can get at least - perhaps more - pleasure from music over these low-fi sources than we can, despite limited bass, colored middle and distorted top.

We also know that if we stumble across second hand audio equipment from twenty, thirty even fifty years ago that we can be shocked how enjoyable a sound they make for a very small investment. We also know that if we dust-down some BBC monitor speakers from the 1960s or 70s that they can (under optimum conditions) give us 80% of the performance of today's BBC-derived monitors, such as ours. So why are we even in business? Why didn't the development clock stop in 1975? Why did we even lift one little finger to start on the development of the RADIAL cone project? Why indeed.

The answer has to do with addressing that 20% performance shortfall, and the first step has to be that someone with time, energy and resources to invest recognises some sonic limitation in the existing loudspeaker technology, a temporary performance ceiling which could - and should - be investigated. How do they arrive at this conclusion when there is a plethora of conventional loudspeakers available, all of which have found a market position and hence satisfy particular consumers. Unless they are motivated by, and have the pockets to fund blue-sky research regardless of commercial success (we don't) they must be able to actually hear the limitation of the conventional technology and be able to translate that into hard, factual, comparative technical data that can drive a science based R&D project (we did).

As I said before, I cannot read music, and I do not play an instrument. Nor do I regularly attend live concerts. So what gives me the right to judge this or that sound? Do I have some unique sonic acuity? No, I absolutely do not have special hearing. I know this because when I talk with recording engineers and others who listen to the micro tones in music, we share a common experience. I also know that when talking to senior audio people (in Japan for example) they are completely and utterly in step with what I hear; astonishingly intuitive and insightful at a deep, humbling philosophical level. When I hear live music it just reinforces and refreshes my audio memory of how instruments sound.

I said that the performance of 70s BBC speakers speakers reached a high performance plateau which is remarkably good. I know this because I have many of these speakers to hand in the R&D centre and they give you a sharp reality check with how little real engineering progress there has been in thirty + years. But what can't they do that the best contemporary BBC-style monitor can do? Is it to do with bass quality or extension? A little. Or power handling? Also a little. Or the top end? Perhaps. Distortion? Also a little. What then? The real performance that 35 years of development evidences is not immediately obvious until your audio palate is sufficiently sensitive: it's to do with a reduction in coloration between the notes ..... not on the note, between the notes. And that means revealing more detail in the 'reverberation tail' after the note leaves the instrument and hits the microphone and the reverberant echo that follows the note, bouncing off the walls. And that exceptionally small sonic detail improvement is what a generation of loudspeaker development can offer in the best examples of the BBC monitor today. In short, the macro development of BBC monitor speakers was perfected 30+ years ago; since then it's been micro development.

It is, in my opinion, crucial for you to have the skills to identify and justify to yourself whether you should be investing your hard-earned cash in Harbeth speakers (as opposed to buying any other perhaps cheaper conventional speakers) and that is why it's so important to be able to identify the sonic signature of the cone. And the first step is to be able to identify in your mind, solely by listening, what a dry and wet acoustic space sound like around performer because it is in that gossamer-like region that all the differences exist.

Note: you do not have to know how the instrument sounds in real life as you suggest, although it helps fix the tonal colour of the instrument in your mind. What you have to be able to do is listen to, at, in and through the ambient halo that follow the notes. That is where 30+ years of technological progress is to be found. Once identified you will recognise its shape and nature forever.

Make sense? I just do not know what more I can bring to the table on this one.

Continued on post #44

Macjager
28-01-2012, 07:37 PM
Thanks Alan
Brings a lot of clarity to the discussion, and also has sent me off looking for more information on recording and how best to record my guitar based songs.

Cheers

George

Pluto
28-01-2012, 08:18 PM
Rather straight-forward samples retrieved from recording made using a Crown SASS-P stereo PZM microphone. Snare drum dry here and wet of Trinity Church here. It was edited to snip away the trailing reverbs.
If your intention was to demonstrate the reverberant characteristics of two different venues, I'm not at all sure that this close a recording of the direct sound has any chance of making that point. How far from the drum was the mic positioned?

Likewise, a snare drum seems an odd choice because the rattle of the snare could easily "drown" residual reverberation. A sharp sounding percussion instrument such as wood block or temple block might have been a better choice, accompanied by the question, "which of these sequences was recorded in a church, and which in a dry room?"

STHLS5
29-01-2012, 03:06 AM
..Likewise, a snare drum seems an odd choice because the rattle of the snare could easily "drown" residual reverberation. A sharp sounding percussion instrument such as wood block or temple block might have been a better choice, accompanied by the question, "which of these sequences was recorded in a church, and which in a dry room?"

The first half was dry and the second wet. Each impulse was 0.423s and played perfectly well till it went to youtube (or maybe, the window movie maker?). The increase of tempo was unintentional and was due to bug.

Thanks for the attention.

ST

STHLS5
29-01-2012, 08:33 AM
.... it's to do with a reduction in coloration between the notes ..... not on the note, between the notes[/COLOR]. And that means revealing more detail in the 'reverberation tail' after the note leaves the instrument and hits the microphone and the reverberant echo that follows the note, bouncing off the walls....

"Not on the note, between the notes.". This is what I have been waiting to hear. Finally, a definitive point where to listen for colouration. If you were to ask the same question like what is natural sound or open sound, I am sure of the response from the majority of audiophiles will be different because colouration carries a different meaning to "us."

I know the word "dynamics" means a different thing to many. Many wrongly think it alludes to the speed of a system or speakers.


Make sense? I just do not know what more I can bring to the table on this one.

You are doing it perfectly. The thread in Sound of musical instruments was an enlightened (http://www.harbeth.co.uk/usergroup/showthread.php?1170-The-sounds-of-musical-instruments/page4). I have searched the net over an over to find just one example of explanation of sound as deep and illustriousness like yours but found none so far.

I was just pointing out what seems to be plain and obvious to you may not have been fully understood by "us" which could be due to not getting the good grasp of the terminologies used or simply having a preconceived idea of what's smooth and natural sound is.

The video example of Queen - The Making of Bohemian Rhapsody shows an excellent acoustics in the room but in the Lady Gaga's producer reveals his secrets video we can hear echos in the room. Listen at 1.17 to 1.23. If RedOne is going to use the acoustics of the room to master Lady Gaga's tracks then I cannot expect the Cd to sound the same in my echoless and slightly dead room. I was saying in this context when I said "There maybe no such thing. It is is not how natural or correct the instruments sound but how good the music is". But musically, it may be a big seller.

ST

STHLS5
29-01-2012, 10:05 AM
.........., accompanied by the question, "which of these sequences was recorded in a church, and which in a dry room?"

That wasn't my intention. I wanted to know which of the parts represented the real sound to you. Better still which one is liked better.

ST

A.S.
29-01-2012, 10:26 AM
"Not on the note, between the notes.". This is what I have been waiting to hear. Finally, a definitive point where to listen for colouration. If you were to ask the same question like what is natural sound or open sound, I am sure of the response from the majority of audiophiles will be different because colouration carries a different meaning to "us." ...You have introduce the word 'coloration' to describe some changed quality between the notes. I wouldn't have jumped to that conclusion quite yet. That's a really big step. I hesitate to say coloration because, as a general approximation, coloration is often used to describe the addition of some (unwanted) tonality to the underlying sound. If you cup your hands around your mouth and talk to a wall, you will hear an additive peakiness to your voice which many would describe as coloration. But it's much more difficult to imagine mechanically removing a sound from your voice (other than electronically by processing your voice in a computer), yet technically, that would still be a coloration of your voice, but a negative one. Let's wind back a bit.

It occurred to me in the middle of a sleepless night going over and over how to move this forward, I have these point to make. First, perhaps more accurately I should say not literally 'between the notes' but between the pauses - be they phrases, trills or whatever. But the concept is still valid: it's something to do with that brief moment between the pianist lifting his finger off the key, the energy input into the instrument ceasing, and then hitting the (same) note again. OK?


Let's forget about the sonic quality of the note ...
because even a cheap supermarket hifi speaker can reproduce the note (adequately well that the tune is recognisable)
... and let's assume that there isn't gross coloration issues ...
such as obvious ringing or peakiness (can you imagine those in your head? - like the cupping of your mouth, above) ...
that would give undue emphasis to certain notes in the music ....
.... let's even forget entirely about quality speakers and concentrate our attention on completely standard, low-fi speakers that would be included in an all-in-one mini hifi system ....
We've just said that they can reproduce all the essential notes in music so that we can perfectly follow the tune ....
and ignoring the fact that their little cones and cheap boxes cannot produce deep bass or play loud ...
or the very highest harmonics on the cheap tweeters ....
assuming all of these prerequisites of adequate performance ....
in what area of performance are they going to be outclassed by a great speaker such as one with a RADIAL cone?

Now, you mentioned 'coloration', but in my list above I've said that we should assume that even a cheap low-fi system these days (with wooden box speakers) is not grossly colored. So why pay more? What performance edge is spending 10-100 times more bringing?

As we said, the issue is not the reproduction of the note. The low-end speakers can replay the note as well as the fancy speakers. 440Hz is 440Hz. C sharp is still C sharp. No ambiguity about what note is being played.

That leaves two possibilities as I see it: The fancy speaker must be revealing something in the sound either before and/or after the note has been generated and ceased. And what is that 'factor x'? Is it an additive effect or a negative one. Is the fancy speaker giving us more or less than the adequately good cheap one in this inter-note gap?

That is a nutshell is the entire situation.

A.S.
29-01-2012, 10:53 AM
...The video example of Queen - The Making of Bohemian Rhapsody shows an excellent acoustics in the room but in the Lady Gaga's producer reveals his secrets video we can hear echos in the room. Listen at 1.17 to 1.23. If RedOne is going to use the acoustics of the room to master Lady Gaga's tracks then I cannot expect the Cd to sound the same in my echoless and slightly dead room...STYes, there is a considerable difference between the sound we are fed on those two videos. But we do not have the slightest clue about the monitoring. It is possible that for the Queen video we are being fed a direct sound from the mixer (I suspect not); it's more likely that the position of the microphone(s) is much closer to the monitor speaker in the Queen piece. You can tell that it's an expensive, authorised, planned shoot and a proper sound crew would have been used who knew how to get the best sound out of the room.

But what you hear on the Lady Gaga video is normal, what I would expect in even a fairly well damped studio. You'd be amazed how much sound bounces around, eventually reaching the mic. Do an experiment yourself. Do you have an MP3 player/phone with a record facility? Put it at the back of the room and play some pop music. On replay you'll be surprised how reverberant your room is. The human brain is very good at filtering echoes - the dumb microphone has no discrimination (except from angle of sound).

A.S.
29-01-2012, 11:00 AM
The first half was dry and the second wet....I suspect that by packaging your clip in a video wrapper you have confused the listener. Also, you didn't allow the instrument to decay fully before you looped the next beat as you can see from the spectrogram in post #30.

Since you were trying to made a point about dry and wet (about reverberation) you should have given us more tail on each example.

Pluto
29-01-2012, 12:18 PM
I wanted to know which of the parts represented the real sound to you...
The sound of a snare drum being beaten into submission doesn't seem a particularly useful test, in my view. As I explained, the recording did not show much distinction between reverberation and snare rattle.

A.S.
29-01-2012, 12:54 PM
Continued from Post #40

OK, I'd like to give you some randomly selected samples from YouTube of the piano being played. I'm not sure if you've been lucky enough to get up close and personal with a Steinway D with the lid open, but if you have a chance take it. It's like being in the company of a magnificent race horse: a huge latent potential just waiting to be released.

The sound quality of these videos is unimportant. The musical selection is also unimportant, as is the venue. I just want you to take a look at the performer's use of the foot pedals. There are three on a concert grand - left, middle and right. Which one or ones is most often used?

Down is 'on' for the effect to be actuated inside the piano. Note, the sound is out of sync in many examples - I'm not asking you to listen, only to observe.

Video 1 (http://www.youtube.com/watch?v=VzshqKp3Giw&feature=related)
Video 2 (http://www.youtube.com/watch?v=ucIdjPG80wo&feature=related)
Video 3 (http://www.youtube.com/watch?v=FwO0NAIcnaw&feature=related)
Video 4 (http://www.youtube.com/watch?v=eqqq-FBCjT8&feature=related) (very interesting indeed - the child is fully aware of the change in sound from expected)
Video 5 (http://www.youtube.com/watch?v=sa467dheCuU&feature=related)
Video 6 (http://www.youtube.com/watch?feature=endscreen&NR=1&v=ms0Ax4EFB-Q)
Video 7 (http://www.youtube.com/watch?v=hfs8LZ9zuz0&feature=related)
Video 9 (http://www.youtube.com/watch?v=hfs8LZ9zuz0&feature=related)
Video 10 (http://www.youtube.com/watch?v=3kCqTHkqBdM&feature=related)
Video 11 (http://www.youtube.com/watch?v=TQtwqkM4V4c&feature=related) If you listen closely you perhaps can hear the effect of the pedal but that's not important at all.

In my next post I'll show you what's going on inside the piano. But for now this is enough. There is a reason for this.

STHLS5
29-01-2012, 01:14 PM
The sound of a snare drum being beaten into submission doesn't seem a particularly useful test, in my view. As I explained, the recording did not show much distinction between reverberation and snare rattle.

I wasn't expecting anyone to hear any difference. That the reason I snipped away the reverbs so you only hear the original sound.Earlier, I mentioned about the nylon guitar sound and increasing my room reverbs with aluminium sheets which produced a sound that was well received and described as closer to true to life by the non initiates.

I was pointing out that for some, the accurate sound was the sound with the trailing reverbs which wasn't part of the recordings. Now, let's think for a moment, if for whatever reasons they think the room reverbs portrays the accurate sound to them then could they always be fooled with some speakers where the drivers or cabinets continuously resonate adding colour and corrupting the original sound to be perceived as realistic to untrained listeners?

ST

A.S.
29-01-2012, 03:56 PM
Ok have you looked at the piano videos? What did you notice?

EricW
29-01-2012, 05:21 PM
The most-used pedal appears to be the sustain pedal (the one on the right).

Generally, it is modulated rather than held down continuously.

A.S.
29-01-2012, 05:40 PM
The most-used pedal appears to be the sustain pedal (the one on the right).Yes, that's what I observe to. In fact, the left and middle peddles only very rarely seem to be used. So there is something rather special about the action of the right peddle, whatever it is. And as we'll see, the decision whether or not to press a peddle is defined in the musical score. It's not (or shouldn't be) a decision made by the performer according to how he feels on the day. The composer must have had in mid both an awareness of what the peddle does to the sound and when he wanted that effect to be heard by the audience.

Just to recap a little. We're trying to examine a certain je ne sais quoi that may or not be audible between notes or phrases in reproduced music at home. ST called this coloration. That's one possible term. But I said that coloration implied to me the addition of some element to the sound and I'm not completely happy with the word coloration in this connection. It's not exactly the experience I have. Let's see if we can nail down a really appropriate word. Back to the piano to help us. The piano is really useful to demonstrate acoustic because it's a like a little room on its own, and has the characteristics we associate with rooms: standing waves, echoes, reverberation, decay and tonality so we should be able to recognise those. These next few steps are extremely important relating to the Harbeth musical experience and if you appreciate the importance of the upcoming points then we can close down the HUG as job complete, I can retire, and sales of RADIAL cones Harbeth speakers will rule the audio world!

First step: what exactly does that magical right-foot pedal do? Have a look at this video 12 (http://www.ehow.com/video_2372229_piano-sustain-pedal-tips.html). As EricW said, the right peddle is the sustain peddle. Sustain means last for longer - agree? But the key has already been hit by the performer so how can the note last longer? Surely the energy input is complete.

Step 2: We need to look inside to get an idea of what happens when the keyboard is pressed. Video 13 (http://www.ehow.com/video_2372224_the-parts-piano.html). This is an upright piano, ideal for use at home. We'll also see inside a concert piano which is exactly the same concept but horizontal.

Step 3: Exactly what mechanical-acoustical process goes on inside the piano? The next video shows felt-covered hammers under the strings (connected one-to-one to the keys) in action, and notice how there needs to be a certain minimum force applied by the pianist to accelerate the hammer from rest and to strike the string.

Also - and this is really critical - note the pyramid shaped, felt material tooth-like wedges that slip in between the strings and make intimate contact with them. The rest position for these dampers is down, in contact with every string. They are only very briefly automatically raised and dropped after the note is struck, but all can be simultaneously lifted away from the strings for as long as the pianist applies pressure to the sustain peddle. Video 14 (http://www.youtube.com/watch?v=q3qgZemce9c&feature=related).

In close-up we can see the dampers, and the sustain pedal in operation here, video 15 (http://www.youtube.com/watch?v=XArz6uCsrA0&feature=related). The action of the light damping of the felt on the strings greatly alters the ambience of the instrument's voice. Damping does not add something, (as we often associate with coloration): it removes or suppresses something in the sound. Just as the word implies: damping alters the air, the ambience, freshness, a sense of spaciousness and realism, of 'being there'. If you remember nothing else, remember this point: damping controls resonance. Too much damping kills musicality. The optimum amount of damping is the difference between a xylophone and a real piano.

Step 4: Now an interesting observation. We looked at a random selection of videos in post #44 of piano's being played, and whilst this was not a scientific sample of all available piano videos, it's self evident that the pianists right foot is constantly called upon by the composer to operate the sustain peddle. Several pieces demonstrated that the normal playing mode was playing with the peddle down - that is, applying sustain, just occasionally dropping the dampers on the strings to damp the tone. So the application of sustain, and the consequently brighter tone, is something that is is much desired by composers for sonic effect. The pianist has clear instructions when to apply the pedals, here (http://www.youtube.com/watch?v=0Xj4aFGukHM&feature=related). It's what we need to capture in our hifi systems at home.

Step 5: As I mentioned in step 3, the normal arrangement if you looked inside a piano at rest is dampers in contact with the strings. We know that for sonic reasons, much of the music we hear is seemingly played with the dampers away from the strings, sustain peddle pressed. Example here, video 16 (http://www.youtube.com/watch?v=_pBK8cQ-tP0&feature=related).

Remember: the operation of the sustain peddle is usually instructed by the composer. The composer and performer want the piano to sing-on with a clean, bright tone. It's a vital part of the musical performance: the open voice of the instrument - the sense of air that this creates around the instrument. Even the child performer in video 4 (http://www.youtube.com/watch?v=eqqq-FBCjT8&feature=related) could hear that the piano's voice was too dull, too dampened because she couldn't reach the sustain booster-peddle to raise the dampers from the strings. She didn't want to continue playing: the piano's voice was not as the composer intended.

Step 6: So, what is the undamped, open voice of a piano? What sonic microtonality is it capable of yielding? It's really astonishing. Count how many seconds the reverberation tail extends in this gloriously balanced instrument under repair here, video 17 (http://www.youtube.com/watch?v=W17xD2p8u0g&feature=related). The sound you hear is not the room echo. You're hearing only the sound racing around in the iron frame, soundboard, strings and body of the piano. Note that by implication, the fact that we can hear it decaying so slowly - it would take more than a minute to fully decay, means that the damping in the instrument's structure is extremely small. Interesting isn't it: negligible damping in the structure, and significant felt damping on the strings. The entire sound of musical instruments is about the optimisation of damping: they can all play the same notes.

------------------------------------------------------------------------------------------------------

If we as hifi enthusiasts want to capture what just this one fabulous instrument is capable of we have to pay serious attention to the way our audio system - and especially our loudspeakers - reproduce microtonality. In other words, we have to avoid smothering the vital transient information between notes as one note decays into the next. As we have seen, composers and Steinway & Sons strive for a brightness and openness of tone, invoking regular even constant use of the sustain peddle. But it's the inappropriate and excessive application of damping in conventional loudspeakers cones (the working part of the speaker) that's robbing the audio of it's real-life freshness. They just sound too muffled - like the dampers are down when we should be hearing sustain. They play the notes perfectly; they can't release the information between the notes properly. So the replay experience is not as open and fresh as it is in the concert hall, live. It's too dark, just like the little Japanese girl experienced.

Harbeth's own RADIAL cone material is, as far as we know, the only purpose-designed acoustic loudspeaker cone plastic available in the world. We engineered RADIAL specifically not to over-damp the sound of reproduced musical instruments because that's what I heard conventional polypropylene speaker cones doing. When you hear the low-level transient resolution of the Harbeth RADIAL cone, there is no way back to the foggy sound of conventional, loudspeakers!

Here is a recording I made of Steinway D at the Fairfield Hall, Croydon. You can imagine that the sustain pedal is in much use. This is exactly the bright, clean atmosphere that the piano should have - for classical performance. It is extremely challenging for a loudspeaker to reproduce all the fine grained detail in the decay whilst at the same time being a resonant system itself - as all cones are to greater or lesser extent. A loudspeaker is a controlled resonator masquerading as a transparent window onto another resonator (a musical instrument).

/library/mp3files/ozgur_mixdown-x.mp3

And now we know why conventional closely-miked pop music is quite useless as a tool to reveal latent loudspeaker quality. There is intentionally no air around the instruments and performers and hence the recording presents no real difficulty for even a low-fi loudspeaker. Pop recording is the equivalent of recording the classical piano with the foot permanently off the sustain pedal and the string dampers fully down. To really be able to test a loudspeaker's forensic ability to resolve low-level detail, you have to ask it to reproduce the acoustic space of a hall - and you have to listen carefully between the notes.

Finally, a video demonstrating the effect of damping in the piano, as it is played. A Harbeth RADIAL coned loudspeaker is designed to catch and replay that rich liquid freshness of the sustain pedal in action (when it is being operated) - just as you would here in real life. It's a really, really difficult technical challenge for a loudspeaker. Video 18 (http://www.youtube.com/watch?v=FFehlpjuCIY&feature=related) (watch and listen - comments are in top left corner). In the pause 1'12 to 1'18 or 3'19 to 3'28 the piano tone is still decaying: conventional loudspeakers have difficulty resolving the microtones in that decay period. Can you imagine why?

I developed the Harbeth RADIAL loudspeaker cone because, reacting as the little Japanese pianist in video 4 did, I couldn't hear the freshness of tonality of the concert hall experience on polypropylene-coned speakers.

I hope this is useful for when you next go out auditioning hifi speakers - or better still, to a concert hall.

STHLS5
30-01-2012, 01:38 PM
1) It is extremely challenging for a loudspeaker to reproduce all the fine grained detail in the decay whilst at the same time being a resonant system itself - as all cones are to greater or lesser extent.

2) To really be able to test a loudspeaker's forensic ability to resolve low-level detail, you have to ask it to reproduce the acoustic space of a hall - and you have to listen carefully between the notes.

When you say "to reproduce all the fine grained detail in the decay whilst at the same time being a resonant system itself" am I correct to conceptualize it as to mean while the preceding note decays, we should hear the subsequent note clear and vibrant and yet the decay is still heard without any changes as if the new note wasn't there?

Or like putting a drop of red ink and a drop of green ink to a glass of water and they both remain separate as red and green and not brown? Can't think of a better way to describe your statement.

I still need to think about the second statement.

ST

Pluto
30-01-2012, 03:15 PM
Video 18 is excellent. It also provides a good insight into how the left pedal operates mechanically; you can see the entire keyboard move when the pedal is pushed or released!

It might also be worth noting that the middle pedal is relatively unusual. The left pedal adjusts the impact of the hammers on the strings so that only two of the three strings (that contribute to the higher notes) are hit. The middle pedal, where it exists, keeps the damper raised on notes that are played while the pedal is pressed in contrast with the right pedal which lifts all the dampers. The middle pedal is designed to negate the effect that can occur when the right pedal is used, which is that resonances are easily set up within the instrument e.g. if a middle C is struck with the dampers lifted, all the other Cs will sound to a greater or lesser extent.

A.S.
30-01-2012, 06:50 PM
When you say "to reproduce all the fine grained detail in the decay whilst at the same time being a resonant system itself" am I correct to conceptualize it as to mean while the preceding note decays, we should hear the subsequent note clear and vibrant and yet the decay is still heard without any changes as if the new note wasn't there?

Or like putting a drop of red ink and a drop of green ink to a glass of water and they both remain separate as red and green and not brown? ...STBoth of your observations are absolutely 100% correct.

Technically, there is a word that describes how sound can co-exist, just like the red and green ink that you mention, whilst remaining completely pure and unmixed. If you play a piano at the same time as an electric bass, at the same time as you sing, at the same time as a police siren passes by, these sound waves do not mix together into a muddy brown as children find when they mix the individual bright vibrant colours of their paint set together. In fact, for me, by far the most engaging aspect of attending a live concert - especially a choir - is how much separation there is between the voices. Look and listen closely and you can identify individual performer's contributions - even little mistakes. Their sound waves are not at all mixed in the air of the hall ... and that coexistance without mixing is called superposition.

Can you see the irony of home hifi reproduction compared to the superposition of live sounds in the hall or studio? We collect all those complex, independent red and green colour tones and pass them along to a loudspeaker, itself a complex mechanical resonator with it's own tonality. We take it for granted that having merged the colours into a cone driving force that the individual red and green sonic colours will magically be regenerated in our room, completely independent of each other exactly as the live experience right down to the smallest micro tone. That's a technical impossibility. The loudspeaker has its own internal resonances - some creating obvious sonic additive effects which we call colorations, and many more very subtle ones related to the material 'DNA' of the moving parts, specifically the relatively massive bass/midrange cone.

When you experience live sound it is as light as a feather, weightless, effortless, dimensionless: feed that to a loudspeaker cone and you have then to accelerate, decelerate and keep it flapping about thousands of times a second to create a vague impression of the original sound waves. The loudspeaker is tasked with generating the individual, isolated, unconnected sounds from its one cone but as you suggest, we want only the red and green primary colours out, but to one extent or another, we get muddy brown. It must be - and it is a fact - that beyond all other elements in the reproduction chain, the expectations we place on the speaker cone (not cabinet, tweeter, biwire links, cables, crossovers) are technically demanding. Of all the parts, the most critical for high fidelity sound is the bass/midrange cone, as that is the transformer from electrical signal to the acoustic one. It is the arbiter of reproduction quality.

To run with your analogy: the fundamental issue is that if you want to keep the red and green water-soluble colours separated, you cannot make the speaker bass/midrange cone from ordinary tap water. That would guarantee that they would mix. What cripples polypropylene as a cone material is that, at a molecular or 'DNA' level, it is substantially homogeneous just like the water in your glass. It does not have any inherent chemical mechanism to inhibit the colour mixing. A analogy to using polypropylene as an acoustic transformer (a speaker cone) is to dissolve a kg of sugar in your water glass then add the ink colours. They'd still mix of course, but you'd have made a stiff syrupy brown goo. The resulting sound will be sweet, charming, easy on the ear, inoffensive but acoustically dead. And the goo will resist stirring - an analogy for acceleration - which in turn brings us back to the microtonality problems with conventional loudspeakers.

So, the fresh Harbeth sound is a manifestation of our research - perhaps even mastery - of cone material science. If you can see your face reflected from a midrange speaker cone, there is a chance that it could make great music. If the surface is pitted and dull, so will be the sound: the equivalent of a kg of sugar in the water. In my long experience of listening to loudspeakers I'd say that is a universal truth. It is no marketing gimmick that you can see your reflection in the mirror-like finish of the Harbeth bass/midrange cones from the P3ESR up to the M40.1. That high-gloss, polished, mirror-like finish reflects a fundamentally different cone technology to that of conventional speakers. RADIAL is a molecular modification to the cone which reduces colour mixing, and that means more resolution, more detail between the notes.

Ask yourself how much thermoplastic and rubber you find in musical instruments. The answer is none because these materials are far too acoustically inert; they would kill the 'air' around the notes. Can you imagine the sound of a concert bell moulded in polypropylene?

Macjager
30-01-2012, 07:17 PM
Fascinating! And now I have gone off in search of my Van Cliburn Tchaikovsky concerto No 1, 1958 recording, to hear the piano with a whole new appreciation!

George

EricW
31-01-2012, 01:55 AM
...the fresh Harbeth sound is a manifestation of our research - perhaps even mastery - of cone material science. If you can see your face reflected from a midrange speaker cone, there is a chance that it could make great music. If the surface is pitted and dull, so will be the sound: the equivalent of a kg of sugar in the water. In my long experience of listening to loudspeakers I'd say that is a universal truth. It is no marketing gimmick that you can see your reflection in the mirror-like finish of the Harbeth bass/midrange cones from the P3ESR up to the M40.1. That high-gloss, polished, mirror-like finish reflects a fundamentally different cone technology to that of conventional speakers. RADIAL is a molecular modification to the cone which reduces colour mixing, and that means more resolution, more detail between the notes.



Those who have not been following the HUG for a period of time might read the above and say, well. don't metal-coned speakers fit the above description? Many manufacturers are using metal cones these days, presumably because they appear offer an answer to the "smearing" problems caused by conventional materials.

As I've understood past explanations, it's not only that RADIAL "reduces colour mixing" by its greater rigidity, it's that it also does so without introducing the audible resonance problems that e.g. a metal-coned can (or is likely to?) have. Since you're making a case for the critical importance of the cone material, I thought it might be good to be reminded that control of cone resonance is a key factor as well, and the combination of high rigidity and low resonance (if I have that right) another unique RADIAL advantage.

A.S.
31-01-2012, 09:26 AM
Those who have not been following the HUG for a period of time might read the above and say, well, don't metal-coned speakers fit the above description? Many manufacturers are using metal cones these days, presumably because they appear offer an answer to the "smearing" problems caused by conventional materials.

As I've understood past explanations, it's not only that RADIAL "reduces colour mixing" by its greater rigidity, it's that it also does so without introducing the audible resonance problems that e.g. a metal-coned can (or is likely to?) have. Since you're making a case for the critical importance of the cone material, I thought it might be good to be reminded that control of cone resonance is a key factor as well, and the combination of high rigidity and low resonance (if I have that right) another unique RADIAL advantage.Great - I was waiting for that perfectly logical observation. Indeed, on the face of it, if conventional semi-rigid plastic cone material acts like an acoustic sponge, then surely the opposite solution, a cone made of rigid metal would make an ideal loudspeaker cone. A perfectly reasonable assumption. If we continue with the great analogy of red and green inks in the glass of water - if we add the sugar we have not only mixed the coulours into a muddy brown, but made the whole mix viscous and dead. Alternatively, at the other extreme, find a fresh dry glass, drop in the red and green inks and stir with a teaspoon. In both cases the result is a an undesirable brown colour mix. In the sugar/water mix as the spoon touches the side of the glass there is a deadened, soft tinkling sound. In the second, the dry glass, there is a bright sound as the spoon touches the sides. Both result in a muddy brown ink. But the sound quality is very different and it's that which this entire thread has been focusing on. The dampers up or down.

I hope that if you have followed this entire thread and in particular, understood the operation of the piano pedals, we can neatly move towards an answer to the selection of the cone material, and as to why a metal cone may not be any more ideal than polypropylene. It's another 'take' but not perhaps the optimum solution to faithful reproduction.

1. We set aside a year or so, we trawl the world looking at all known materials materials that could theoretically be formed into speaker cones* and we devise a way to technically evaluate and grade their sonic anti-smearing potential using very sensitive and novel measuring equipment (we don't trust our ears at this stage).

2. Material by material, over the months we start to plot all those graded results for 'reproduction naturalness' Rnat on a graph, and our graph initially looks like picture A. Rnat is necessarily a index of many factors - the graph is actually a 3D matrix so this is a great simplification to give the general idea. The first four materials we tested (shown with Xs) had characteristic that placed them on the Rnat vertical scale as marked. There is nothing to learn from this distribution.

3. One by one as the points start to build up on the Rnat overview graph, we see a pattern emerging. These points are not as random as they appeared at first with just a few sample points. There are two significant clusters groups, and upon examination of the materials in these group, we can find some common technical attributes of the materials. Between P and H, in no-mans-land there is a median area, let's call it F (for fulcrum). Since we now understand the common material attribute present in groups P and H we can conceive of a material that would (theoretically) have characteristic F. Unfortunately, a material possessing the desired attribute F does not exist as an off-the-shelf material, but as with all chemicals, could theoretically be synthesised in the lab. Picture B.

4. Many experimental synthetic materials F1, F2 .... F20 are made by Harbeth, and analysed and plotted. The objective is to hit the green F zone, and to get as far above the A-B performance threshold as possible. This step takes about 18 months. Picture C.

5. Materials which after analysis fall towards or into the P end of the spectrum are invariably thermoplastics. Materials which fall towards the H end are hard materials, such as metal.

6. I made a decision which one of these would become Harbeth's RADIAL cone material. No. 17 I recall. It was the best overall compromise and the most stable in production, which I personally supervised. Bulk quantities of this are made under controlled conditions. (I may share this picture; some thought needed because of commercial secrecy). Picture D.

7. The optimum cone material in region F is neither like pressing the piano sustain pedal when sustain is not required nor like shutting the lid to over-damp the sound. Materials P tend towards over-damping the microtones, and materials H tend towards under-damping.

This Harbeth RADIAL project was part funded by the British Government who have recouped the cost many times over in corporation tax, income tax on employees and export earnings for the UK. The final reports are thousands of pages long.

* There are innumerable potential candidate materials available to industry. However, the list is whittled down rather (too) rapidly when we consider that the cone cannot realistically be too heavy and/or thick. There is a minimum loudness sensitivity which the hifi consumer will tolerate - perhaps around 80dB/1w/1m - and that calculates backwards to a certain maximum cone moving mass for a given magnet/coil motor. That requirement is the killer blow: it eliminated over 50% of potential material solutions even though they had other interesting acoustic properties.

Euler
31-01-2012, 02:57 PM
Great stuff in general and especially recently, Alan!

Alan has stressed Harbeth's reproduction of "microtonality" that reveals "the vital transient information between notes as one note decays into the next." But doesn't this reproduction of "microtonality" also reveal the fine texture, not just between notes, but of sustained sounds themselves when those sounds, like say the human voice and the French horn, have a complex microtonal structure?

Isn't this one reason, along with RADIAL's lack of coloration, that Harbeth's are famous for the accurate reproduction of the human voice and instrumental timbre?

Bruce

A.S.
31-01-2012, 06:36 PM
... Alan has stressed Harbeth's reproduction of "microtonality" that reveals "the vital transient information between notes as one note decays into the next." But doesn't this reproduction of "microtonality" also reveal the fine texture, not just between notes, but of sustained sounds themselves when those sounds, like say the human voice and the French horn, have a complex microtonal structure?

Isn't this one reason, along with RADIAL's lack of coloration, that Harbeth's are famous for the accurate reproduction of the human voice and instrumental timbre?...Yes, absolutely so. But surely the easiest way for me to have demonstrated the microtonality issue is for us to listen to the decay of notes, which we can (or should be able to) separate from the note itself.

You are of course right that the sonic fogging problem, especially from vacuum-formed polypropylene (which accounts for perhaps 85% of all hifi speaker cones), is also obvious 'around' the reproduction of voices. If you only listened to speech not music, now you are alerted to this loss of resolution, you should be able to listen out for it.

I'm reminded of my experiences at the Penta/Ramada hifi shows some time ago. At that time, the P3E/ES/ES2 used a bought-in popypropylene cone (before we invented the 5" RADIAL driver for the P3ESR), and being our entry level product at a attractive price point and size, we made sure it was always available for demo. Being so small, and as our American friends apparently say, cute, it always drew a very good crowd of amazed listeners. As I normally prefer to be at the back of the room, it often became quite difficult to make a path through the entralled listeners to be able to swap the speakers over. Consequently, the (original) P3 always won a goody proportion of the daily demonstration time. I've not mentioned this before out of deference to entirely satisfied customers of the old P3, but after a while I found myself on edge, desperate to swap to any of the RADIAL-coned Harbeths and naturally to put on the best show we could. And when the moment came to switch over to say, the then C7, it was as if a blanket had been lifted off the speakers. Operatic voices and, cello and piano sounded much more open - believeable actually. Yet nobody noticed.

I vividly recall that after handed over to colleages, taking a break and wandering around the show, I could tell if a RADIAL Harbeth was playing just as I stepped back through the doorway of our room, eyes closed. It still amazes me that anyone is using PP. It makes excellent shampoo bottles, but rather strange speaker cones.

Basically, if you can't see your face reflected in a speaker cone that covers the vital midrange, I do not believe that the loudspeaker is capable of the higest level of fidelity, regardless of the price ticket and fancy cabinetwork.

A.S.
31-01-2012, 07:24 PM
P.S. Hopefully you watched video 17 (mentioned in Step 6 of post #48). Beautifully recorded. I thought I'd call the author, piano restorer Barry Taylor of Taylor Piano Restoration in Virginia and thank him for the clip. I asked him if he had any more footage of the final decay into silence at the end of the clip. Unfortunately, what we hear is all he has, just the time it took him to walk over to the camera and switch off.

I've asked him if he is likely to be able to repeat the exercise, and to leave the camera running for a minute or two after he hammers the peg in, and he's kindly agreed to do so. He has Steinway pianos passing through his workshop continuously.

If you listen carefuly to video 17 again, here (http://www.youtube.com/watch?v=W17xD2p8u0g&feature=related), you may notice that the decay has a particular quality. At first, the sound decays quite rapidly, then the decay slows down and levels out (the sound is being sustained in the piano structure), it is held in that state for many seconds (can you imagine how difficult that is to achieve?), and then decays again. It is hard to believe that mere humans could have manipulated minute details of the structure and materials of the piano to allow such a beautiful, clean decay. Truly a mechanical masterpiece. And asking a loudspeaker to reproduce the complexity of that sound is an impossibility - but we can try our very best to do so.

If you have enough bandwidth you can watch in HD, full screen. Just like being with Barry at the piano. A great loudspeaker, just like a great piano, is entirely about resonance control. Too much and you kill the fresheness of sound: too little and you have ringing which becomes fatiguing.

A.S.
01-02-2012, 10:50 AM
Video 17: are you nit picking the details and missing the big picture? It's pretty obvious that Barry Taylor is talking in a room, adjacent to the piano. But if you, like I, have actually peered into a Steinway (see my post many years ago referring to looking into Peter Katin's Steinway, which led to the C7 cabinet idea) you would recognise the characteristic racetrack-reverb of the instrument has little or nothing to do with the room. No, the room sound is not prominent: the close proximity of the piano to the camera mic substantially drowns out the room. Forget about the room sound. We've been through this close-miking thing recently, and in effect, with the camera right up againt the case we are hearing the nearfield sound of the piano.

I urge you to go out and find a Steinway, lift the lid and ask someone to press the sustain pedal (lifting off all the dampers) and play a compex chord as loud as possible, leave the sustain pedal depressed and count the seconds until the decay has reached inaudibility. I think a good sixty seconds will be needed.

A.S.
02-02-2012, 08:43 AM
I have been advised by that there is an opinion amongst some viewers (who do not have the balls to step forward and say so here) that the claim that the decay sound of the Steinway in video 17 is somehow manipulated or faked by me or the author of the video. As I value my personal integrity I am baffled as to why I and even less so, the piano technician would be motivated to do this. It is exactly how a Steinway sounds when disassembled. Clearly there are some sad souls on the fringes of audiophila who have no contact with live music or instruments.

As a result of this mendaciousness I have written to Barry Taylor of Taylor Piano Restorations asking him if he would be kind enough to make another video, as and when he has a suitable instrument and the time to do so, giving us an even greater insight into the tonality of the various parts of the piano - strings, iron frame, case and perhaps above all, the sound board. He has said that working with these pianos every day, he takes for granted how they sing, and the sound recording in the video was in his opinion a by-product of his demonstration of craft skill. He is delighted that we found the sound side of the video of interest.

This video demonstrates a sound very similar to that of video 17 (http://www.youtube.com/watch?v=W17xD2p8u0g&feature=related), but outside in the open air. Marbles being dropped on the strings. If the body were still fitted, the sound would have been deeper, richer, louder and with longer sustain, as we observed in video 17. Video 20 (http://www.youtube.com/watch?v=fRnqKHHb3Vk&feature=related)

85% of the piano is wood - factory tour Video 21 here (http://www.youtube.com/watch?v=jAInt7hIZlU&feature=related). Note the consideration of grain and the bell-grade iron frame. Bell-grade so it can sing with sustain. Note how the hammers are individually pierced to achieve the desired tonal sound.

Here are some behind-the-scenes videos about the Steinway. Note the comment about 'tonal projection' and the super-critical soundboard. Video 22 (http://www.youtube.com/watch?v=B6gumj9EWWE&feature=related)

Considering the (disputed) body-resonance sound of video 17, this video showing the disassembly of a Steinway for transportation is accompanied by music emphasising exactly that inherent tonality. Video 23 (http://www.youtube.com/watch?v=rBaBtmp4C6k&feature=related). That surely must be obvious to all.

It's alive! A contact microphone on a string ..... video 24 (http://www.youtube.com/watch?v=ML-yJgvJwKA&feature=related). Note how sensitive the entire structure is to acoustic feedback - when controlled that is exactly what gives the piano it's bright tonality.

Very important point demonstrated here video 25 (http://www.youtube.com/watch?v=Dl2lOyhSUg4&feature=related). The sound board in not touching the iron frame. If it were, it would be constrained, and wouldn't be able to sing. And concert-goers recognise this bright, clean sound video 26 (http://www.youtube.com/watch?v=C0scesOGWo4&feature=related) - two examples, somewhat different room acoustic.

Obviously, the nay-sayers would believe all this to be trickery.

thurston
02-02-2012, 09:06 AM
Apart from your integrity (in which I trust) : What interest should you (or Mr Taylor) have to manipulate the sound???

Sometimes people seem to fall into some kind of conspiracy theory trap or something.
Somewhat mad, I think.

keithwwk
02-02-2012, 09:59 AM
This thread is very interesting, informative, true and educate. Viewers, agree or disagree, should grateful and appreciate the effort Alan provided here. Harbeth owner or not, after this and until here, should at least know how to listen the true signature sound of a piano and understand what is superposition really mean but not imaging how it should sound by using hifi magazine or all the audiophile descriptions.

I am truly enjoying and appreciate all the information here.

A.S.
02-02-2012, 10:37 AM
To make it easy for members, I lifted the sound of the marbles being dropped onto the strings of the body-less piano (video 20) and combined it with the sound of the complete piano under restoration (video 17).

First the marbles, then the hammer. Points to note:

1) I applied a little bass-cut in the second clip (complete piano being hammered) because the body makes the sound richer (obviously) and the play-park piano is just sound board, strings and iron frame.

2) The marble clip is recorded outside in the open, so definitely no room wall reflections

3) Despite 1 and 2, the basic quality of the sustained resonance is remarkably similar. They decay at about the same rate with (all things considered) the same sort of tone. If you hadn't seen the videos and just heard the sound (listen to audio clip) you'd be sure to guess that bothw ere some sort of stringed instrument. This must mean that the fundamental sonic character of the piano is not from the rigid, massive case but from the lighter parts: the strings, sound board and perhaps to a lesser degree, the cast iron frame. The volume or loudness of sound is something related to trapping it and amplifying in inside the wooden case. We intuitively know this. All wooden instruments are crafted from very thin wood, moulded and shaped for sonic reasons. We also know that a thin-wall speaker cabinet (like the Harbeths) is easier to tune and damp than a rigid one and during design the sound can be steered from bright and resonant to sweet and damped.


/library/mp3files/inside piano.mp3

STHLS5
02-02-2012, 12:45 PM
,

Obviously, the nay-sayers would believe all this to be trickery.

They may say the same thing here (http://www.youtube.com/watch?v=rj_U4NeS7tE&feature=youtube_gdata_player) at 6:57.

ST

HUG-1
02-02-2012, 01:35 PM
Another obvious fake here (http://www.youtube.com/watch?v=eU2LIQrtDkA&feature=related)?

Or maybe video 17 could even be real? Of course it is. Hope the nay-sayers are ashamed of themselves.

A.S.
02-02-2012, 03:01 PM
Here is another example of the structural resonance of the piano, provided by a member. Sadly, it marks the end of the instrument which can happen for a number of reasons. Woodworm, distortion of the iron frame, cracks in the all-critical sound board. This is on the beach (http://www.youtube.com/watch?v=QQVrwImrvCM&feature=youtube_gdata_player), so again no room reflections.

Hopefuly everyone is now convinced, after a huge and needless additional effort, that what Brian Taylor has filmed on video 17 is completely genuine and very beautiful sound of an entire piano structure in glorious resonance. Blindingly obvious to anyone who turns off the hifi and actully goes out listening to live music in the real world. Surely only such a person is qualified to preach to others about 'high fidelity' and whether this or that equipment, recording, speaker, room is more natural than another.

EricW
03-02-2012, 01:24 AM
I have been advised by that there is an opinion amongst some viewers (who do not have the balls to step forward and say so here) that the claim that the decay sound of the Steinway in video 17 is somehow manipulated or faked by me or the author of the video.

I have no idea who these viewers are but the very thought that you would fake something like this is utterly mad.

Not wanting to appear sycophantic, I and maybe also others don't perhaps express appreciation as much as we should for all the time and effort poured into what is almost a free course on acoustics and speaker design. Educational, informative, interesting and very much appreciated. Thank you.

The thread has moved into the direction of a focus on the critical importance of cone material, and how RADIAL is different from anything else on the market. There's also been an introduction of the idea that the real progress since the 1970s has been in terms of cone material advances (especially RADIAL, of course).

That raises for me this question. It seems to me that the Harbeth recipe has, broadly speaking, two key ingredients: (1) BBC design principles - e.g. damped thin-walled cabinets, and (2) RADIAL drivers. If one were to take a RADIAL driver and mount it in a "standard" type of speaker cabinet - thickwalled, mitred, glued, or even braced or metallic and designed to be as rigid as possible - how would the combination sound? I realize that, unless Harbeth were ever to choose the route of licencing RADIAL, we are not likely ever to know. But from a purely scientific, non-commercial perspective, wouldn't this be the ideal way to isolate and separate the two variables, in order to get a better handle on what each does in fact contribute to the end result?

STHLS5
03-02-2012, 09:39 AM
...

That raises for me this question. It seems to me that the Harbeth recipe has, broadly speaking, two key ingredients: (1) BBC design principles - e.g. damped thin-walled cabinets, and (2) RADIAL drivers. If one were to take a RADIAL driver and mount it in a "standard" type of speaker cabinet - thickwalled, mitred, glued, or even braced or metallic and designed to be as rigid as possible - how would the combination sound? ..

If you really want to separate RADIAL from the thin walled cabinet then wouldn’t it be much simpler just to use the RADIAL without any cabinets? I thought question 14 of Harbeth’s FAQ explained very well about the role of a cabinet and the reasons for choosing the more expensive thin- walled cabinets that are being used by Harbeth. But it was an interesting question that would have crossed the designer's mind during the development stage, wouldn't it?

ST

keithwwk
04-02-2012, 04:00 AM
It seem to me Eric was curious how a RADIAL driver sound on a "non thin wall BBC cabinet" instead of how it sound without a cabinet. If I am not wrong the thin wall concept is used before RADIAL.

denjo
25-02-2012, 09:08 AM
I am an acoustic guitarist and this gives me a good idea of the timbral qualities of 'live' music!

Macjager
28-03-2012, 11:30 AM
So, last evening we went to the Papp Laszlo Sport Arena here in Budapest to hear Lorraina McKennitt and here band play (set up like a concert bowl, with the stage being at "centre ice" facing the long side of the arena, black curtains blocking off the other side).

One would think that the acoustics would be horrible, but to our amazement, the music and voices sounded very clear. I can say that at no time could I pick out any particular instrument in its location on stage, as the mixing was such that all instruments came from the wall of sound...even when an instrument was soloing. So, it proves once again, when amplified, a live show usually will just be a wall of sound, rather than placing the player in a particular "spot" in space. What we could notice was when an instrument was louder than the others, but again, this was due to mixing and not some acoustic trick of the venue.

The "illusion" of hearing an insturment in its "specific spot on stage" is really a manipulation of the engineer, rather than some magically transmorgifiration of the sound by the audiophile system...the crazed arguments of cables/tubes/marble/steel/unobtanium constructed support making the sound do things is unsupported nonsense...

Conclusion, go to live shows, listen occassionally to where the sound is coming from, acknowldge the obvious, then just sit back and enjoy.

Imagine how much show tickets would cost if each performer insisted on their speaker and microphone cables costing $1000/m to get that "perfect" sound...

cheers

George

royals1871
08-03-2013, 11:42 AM
Rarely post on here, but would like to offer some opinion.

I am not an audio engineer, a scientist, a salesman, an audiophile (I still don't know what that actually means). I'm a musician, keyboards mostly but I've messed around with nearly all of the instruments you would see in an orchestra. Never got my hands on a double bassoon mind. I've also tuned Keyboards from Harpsichords to Rhodes.

I guarantee that anyone who claimed that 'video 17' was intentionally manipulated has not opened the lid, played a piano, or slammed the lid of the piano in a fit of pique! Either that or they are joking. Next time you're in a music store, go up to a piano lift the lid, and gently run your fingers over the strings, flick your finger on the bracing, then play a note on the keyboard. You have achieved sound from a piano in four different ways, with 4 different results. The Piano is a severely manipulated, severely dynamic harp after all.

Below is chap who dissects an upright (not grand) piano, same effects exist fundamentally. You will see the full insides of a piano being removed - how piano keys are struck, YOU can see how the sustain pedal would manipulates the sound/tone etc when the hammer of the key hits the TAUT string. I hear the TAUT strings vibrating, echoing, rumbling when indirectly knocks the exposed piano.

If you're interested in how it's all made up, this is a very interesting section, grab a sandwich!

Piano dissection/disassembly/workings. 1906 Ludwig & co.

It's brutal really, but informative.

http://www.youtube.com/watch?v=baau5YYiIXk

What is the oldest stringed instrument keyboard we know of? It's the Clavichord! Below clips show a progression from Clavichord to Harpsichord to the modern PianoForte or Piano.

Lots of different tones, vibrations and manipulations (some you may not exepct) performed right here!

Very interesting:

Bravo to the moustache'd demonstrator!

Part 1: http://www.youtube.com/watch?v=4uCCw_hmILA
Part 2: http://www.youtube.com/watch?v=a9IaE2i-DmA

More dynamic, more manipulation, more excitement! Wow, sounds like a lot of instruments, very clever!

A very, big deal.

Pianoforte, which means LoudSoft in Italian.

The Italians had most musical descriptions (be it the pace of the music being played by voice/instrument eg Adagio, the volume of the music being played by voice/instrument eg Forte, or even the names of the instruments being developed) sewn up by then.

From Wikipedia, but this is a decent bullet point explanation showing the progression for a stringed keyboard to combine expression of control, volume and sustain.

"While the clavichord allowed expressive control of volume and sustain, it was too quiet for large performances. The harpsichord produced a sufficiently loud sound, but had little expressive control over each note. The piano was likely formed as an attempt to combine loudness with control, avoiding the trade-offs of available instruments."

A.S.
09-03-2013, 06:33 PM
... Bravo to the moustache'd demonstrator!

Part 1: http://www.youtube.com/watch?v=4uCCw_hmILA
Part 2: http://www.youtube.com/watch?v=a9IaE2i-DmA...I agree, two absorbing videos. You've got me interested in pianos again!

Did you notice the fabulous acoustics of the hall in which the video was recorded? I calculate the reverberation as about one second (inspired guesstimate). It's the cleanness of the decay after his voice which is so impressively even in tonality. I made a short excerpt from the video here.

/library/mp3files/acoustic_space1.mp3
When the recording engineer sets just the optimal amount of reverb for the piece, and the reverberation is sonically characterless, it brightness and loudness to instrumental tone: it brings out the 'flavour' and sparkle which is what holds the listener's attention. In our home listening environments we too need to aim for characterless reverberation. But no reverberation at all - like listening in an anechoic chamber - kills the life in music as we expect it to sound.

Accurately replaying this reverberation 'air' on a loudspeaker is one of the most difficult - perhaps the most difficult - aspects of the entire design. Contrast what you can hear in the reverb on a conventional speaker with a polypropylene cone with what you can hear on a Harbeth RADIAL cone. The reason I developed RADIAL was precisely to reproduce that microtonal decay, which polypropylene just cannot do accurately due to its molecular structure.

A.S.
10-03-2013, 09:38 PM
The grand piano is the most incredibly wonderful machine. It is hard to even begin to describe the subtlety of the construction and the exceedingly well balanced forces harnessed for our pleasure. The taut strings exert a pull of some 60 tonnes and would kill if the iron frame failed and the instrument imploded. No other instrument of the orchestra harnesses pure energy in such a way. The details of the engineering leave one speechless.

On other clips here on HUG I've mentioned the importance of a quality audio system trying to regenerate the 'air' around an instrument, the sense that it is placed in an acoustic space and to give the listener a sense of how large that space is. Is it a small church or a cavernous international concert hall with beautiful, well controlled but lively acoustics. I know that I'm listening to a piano - that's taken for granted - what I really want to hear is how that instrument interacts with the hall around it. To be honest, I'm often more interested in that interplay, especially between hall and piano, than the music itself: it's that important to me. My deveopment of the Harbeth RADIAL™ cone is a direct result of wanting to release more of the masked sound of the hall which we concert goers appreciate when the pianists fingers leave the keyboard and the sound decays into silence - perhaps over several seconds. Of course, if the recording was made in a studio with the microphone jammed under the lid - the way much jazz is recorded - there will be no ambience; we can't create ambience out of nothing - all we can do is release it from a recording that already contains it.

I've commented before that there is a certain 'sourness' of tone which I find very attractive from the piano. Well, I've just discovered a video which perfectly and accurately describes what I hear live, and what I want to hear from a well recorded piano. It is, in fact, nothing more (! - the engineering challenges must be horrendous) than extremely subtle beat frequencies of strings in motion. I hear them, and I really like them; they define the concert grand piano, especially the Steinway D. What I didn't appreciate until I watched this video is that the beat products are not the result of played strings, but open, undamped strings that have not been played (hit).

This really is astonishing. I hope you enjoy and appreciate the significance of this engineering miracle. The key point is that we hear cyclic microtones not from the played strings, but from the unplayed string. The 'sourness' I describe is illustrated as the 3 or 7 times a second beats. Reproducing these beats and hearing them clearly when the fingers lift away is damned difficult for a loudspeaker. Speakers don't 'do' subtlety very convincingly.

Video here (https://www.youtube.com/watch?v=-WajSWTLtqE). Full screen is best.

Now example recordings that illustrated the subtleties of tone which are really brought out when you let the piano breath into a big open space .... (this is the primary reason I became a speaker designer, to get hi-fi speaker sound as close as possible to the real Steinway which I adore)

Hop over to here... (http://www.harbeth.co.uk/usergroup/showthread.php?1832-Demonstration-of-the-unique-Harbeth-RADIAL%99-cone-material-v-other-materials&p=22516#post22516)

Pluto
11-03-2013, 12:00 PM
Well thanks Alan, for causing me to use the best part of an entire morning looking at videos of piano technicalities ;)

Two I found of particular interest:

This one (https://www.youtube.com/watch?v=7uCu5pTBzbU) shot at a lecture about the challenge of equal temperament and changing approaches to it over the past four hundred years. Unfortunately, there is quite a lot of audible analogue flutter which disguises some of the more subtle examples, but an interesting video nonetheless.

This video (https://www.youtube.com/watch?v=I5pFx3XJQdI) shows a top class piano tuner at work. To call the guy merely a 'tuner' is doing him down as you will see. Piano maintenance at the highest level involves a lot of work on the mechanism and this film demonstrates some of the repairs done as part of the piano technicians routine.

royals1871
14-03-2013, 03:06 PM
I agree, two absorbing videos. You've got me interested in pianos again!

Did you notice the fabulous acoustics of the hall in which the video was recorded? I calculate the reverberation as about one second (inspired guesstimate). It's the cleanness of the decay after his voice which is so impressively even in tonality. I made a short excerpt from the video here.

/library/mp3files/acoustic_space1.mp3
When the recording engineer sets just the optimal amount of reverb for the piece, and the reverberation is sonically characterless, it brightness and loudness to instrumental tone: it brings out the 'flavour' and sparkle which is what holds the listener's attention. In our home listening environments we too need to aim for characterless reverberation. But no reverberation at all - like listening in an anechoic chamber - kills the life in music as we expect it to sound.

Accurately replaying this reverberation 'air' on a loudspeaker is one of the most difficult - perhaps the most difficult - aspects of the entire design. Contrast what you can hear in the reverb on a conventional speaker with a polypropylene cone with what you can hear on a Harbeth RADIAL cone. The reason I developed RADIAL was precisely to reproduce that microtonal decay, which polypropylene just cannot do accurately due to its molecular structure.
Yes, I did, fabulous acoustics indeed. An attacking acoustic.

Acoustics in halls such as these can break many a performance, lumpen and leaden. The Royal Festival Hall has an awful reputation with their acoustics. I haven't met anyone who enjoys performing or listening in that environment.

Acoustics in performance halls should be an accomplice to the performer, if you can control it through your performance to me that is a good acoustic, if you can't it is a bad acoustic.

Bit black and white, but...