This thread concerns reports from visits to global hifi shows.
This thread concerns reports from visits to global hifi shows.
Alan A. Shaw
Harbeth Audio UK
For family reasons we arranged to be in Bristol during the show. I spent a full day in the company of a BBC sound engineer (a Sound Supervisor in BBC speak) as we toured the show, held in the Mariott Hotel in the city centre. It's been twelve years since I last attended, and many years since I attended any UK show. This Bristol show seemed well organised, well attended and easy to navigate especially using the stairwell.
The hotel is typical of 70s build and as you would expect the hotel bedrooms that have been converted to demo rooms have many acoustic issues in the low frequencies, but at least the rooms are of a common build with common problems. Some exhibitors made an effort to treat the side walls with portable acoustic panels; all had to contend with the large glass window opposite the entrance and behind the speakers. Conference rooms on the lower floors were in use by AV companies.
By the time my friend had joined me I had toured about half the rooms and briefed him with my opinion. To say I was shocked would be an understatement. Audio memory is a fickle friend but I cautioned him that either I was going deaf or many exhibitors had, seemingly reduced, rolled-off, attenuated, subdued or otherwise adjusted the 'top end' loudness downwards. I retraced my steps back to the lobby and together we re-started the tour.
My BBC friend creates audio all day long. He is a highly experienced listener whose opinion I trust and he concurred with me, adding that, this lack of sparkle was not just a relative bass/mid/top balance but that there was all too frequently a lack of 'bite'. What we both heard want plenty of middle and an inadequate amount of high frequency detail. To my ears, even when the relative tweeter signal was tolerably OK (i.e. not too dull) there was the ever present darkness of tone at the top of the bass/midrange, just as the tweeter flared in, which I know to be the tell-tale characteristic of cone coloration the very area where the exclusive Harbeth RADIAL cone material is at maximum advantage. So the years have passed, speaker have innovated all manner of cloaks but still the fundamental limitations of conventional cone materials remain.
I took with me an excellent Olympus VN-5500PC digital voice recorder with an inbuilt mono microphone. Held discretely at waist height when standing (i.e. approx on axis with the tweeter) I made a recording of the music being reproduced in numerous (unnamed) rooms, with the AGC turned on to be sure that the recorder didn't clip. I'll edit that sequence together out of curiosity to see if, on replay, the differences we heard are evidenced when you listen. It may or may not work.
I suggested that what we were hearing was 'the sound of 2011'. My friend thought that widely evidenced rolled-off lacklustre top end may have been a reaction by manufacturers to the 'hard digital sound' and this was agreed by a highly respected technical journalist. If the industry has taken this direction, it has killed the life in music, made it far less engaging and involving. One highly respected British brand had so little top end that we both though that the tweeters had been electrically disconnected or accidentally destroyed. You could not there the high hat at all, nor a small bell known to accompany the music and the violin sheen was completely absent. The bass was good though. Was there something in the air in Bristol which threw a woolen mat over most hifi speakers? Was it the presence of so many bodies in the small rooms? I don't know - but very few rooms sounded anything like real live sound. And the ones that did were often smaller brands and some really innovative speaker enclosures.
And the verdict? A shock. Yes, there were some speakers what sounded great - and some up market iPod docks sounded very well balanced given their size and construction - but it reaffirms what even I took for granted: the Harbeth wide-band, low colouration, open, balanced, revealing sound is something really, really special. If nothing else it explains to me why even though we don't advertise, exhibit or seek reviews in the UK our home sales are rising faster than at any time in our history. It really must be about purity of tone.
I'll see if I can extract usable audio from the Olympus during the next few days.
Alan A. Shaw
Harbeth Audio UK
Hello Alan! It was wonderful to meet you yesterday - and a total surprise. What you say above about the weak high frequencies was definitely in evidence at Bristol. I think that there is a growing sector in the hi-fi community, who are on a misguided misson to remove 'digital hardness' as the main barrier to enjoyable sound. Even if it is not there. This could be a possible reason for the increase in vinyl and valve-based systems at many shows, which a lot of people seem to think take away treble probems by being a bit rolled-off. This is all wrong. There is too much supposition and hearsay in hi-fi which leads to a distorted understanding (of what the important isues are) by people who are not technically-minded. These endevours have no basis in reality. The rational, scientific, reasoned approach is what has led me to Harbeth.
Interestingly, elsewhere it has been reported that many of the show systems at Bristol were full of exaggerated raised treble to 'give a better soundstage'. This has been widely agreed on another forum. I can't understand that. Unless people are pre-conditioned to think that is what a show will sound like and so when they listen that is what they get!
Thanks again for the photo oppertunities - I will treasure them!
There is a generation coming along that has spent hours listening to iPods. It's not surprising that to sell products, familiar sound (but more of it), is an easy approach.
The past three years or so I have been trying various speakers with a view to taking them on as a dealer. My findings aren't the same as heard at the Bristol show, as I am finding the majority are thin and hard - very bright with no body or warmth.
No idea why this is other than they appear to throw a lot of information at you on first listen and thus may appear to be more exciting in a demo room. However, I find that fatigue sets in very rapidly after the first hit of top end detail.
I have had exactly the experience Alan describes. In between my two pairs of Harbeths (HL-P3ES2 first, then P3ESR) I used a pair of well-regarded compact floorstanders. I thought they were quite good, and not that dissimilar to my first Harbeths in many ways.
After listening to my P3ESRS for a while, for fun I swapped the floorstanders back in, and I couldn't believe what I was hearing. Or more accurately, not hearing. Whereas before they had seemed smooth, refined, easy to listen to, they now sounded muddy, rolled off, and distinctly lacking in upper midrange and treble information. They don't begin to provide the musical information and insight of the P3ESRs - and they're not a bad speaker, as speakers go. Just not comparable. By contrast, the P3ESR just tells you much more about what's happening in the recording - without being in any way harsh or edgy. I appreciated them before, but I appreciate them much more after having done that comparison.
At the Bristol Show over the weekend (where I met Alan), I noticed that some exhibitors gave out branded merchandise like tea/coffee mugs. I've also noticed over the years other hi-fi manufacturers selling branded merchandise like mugs, keyrings, t-shirts etc. I wonder if Harbeth would consider selling a range of this type of thing. I've seen photographs inside the factory of staff wearing Harbeth polo-shirts and I'd really like one! I'd love to sit listening to my system with a Harbeth mug by my side too. What do you think Alan? I bet loads of us would buy them if they were sold direct from the website...
OK, I took my hand held Olympus voice recorder and holding it discretely at waist height made a recording of the sound in various rooms, on different music with different speakers and electronics. I don't think any exhibitors (maybe one or two) were using vinyl - everyone else was using CD/DVD and many even most were playing from some sort of streaming source, be it iPad (several in use as music players in the hands of the sales staff) or similar.
I regret setting the Olympus to AGC-on because that has degraded the usefulness of the recording, but as the rooms were dark and the volume control rather fiddly, I thought it better to have an unclipped recording than take a chance on the record levels. Many recordings are unusable. But here is a selection of what the Olympus heard about 3m away from the loudspeakers in various rooms. Naming the rooms is irrelevant - I have no idea what was playing where - the objective is to see if we can hear any characteristics in the clip that may indicate room/speaker differences or whether this crude recording method merely reduces all sound to the lowest common denominator. In other words, was a complete waste of time. You decide.
The recording is mono.
What I can add having listened to this a few times now is that there is a 'general characteristic' that seems to run through several room recordings despite speakers of very different size, drive unit combinations, aspect ratio and position in the room. Furthermore, some rooms seem to have more 'energy' than others. What I'd ask you do is to imagine how if you were there using your two ears those relatively small subjective differences could magnify themselves up to a marked quality difference. And to consider listening fatigue too.
Alan A. Shaw
Harbeth Audio UK
Alan would it be possible to do a brief recording in whatever demonstration/listening room you have there at Harbeth using the same recorder in the same way?
Just to get some idea of what the recorder is capable of.
The recorder is on the recommended list for Dragon Natural Speak in that its audio record performance is considered good enough to effect speech to text reliably. I did actually make a couple of interviews (although the background noise is rather high) with the intention of providing a reference for sound quality. I'll have a look at those clips and see if I can rustle them up for you.
I have so many ideas for what I'd like to do and write about that I could live three lifetimes. Perhaps I should open an invitation page somewhere to encourage you to drive me with subjects that you find interesting ....
Alan A. Shaw
Harbeth Audio UK
Your comments above about other visitor's feedback simply baffles me too. We recognised certain recordings being played in various rooms and we strained good and hard to hear the upper detail. It was rarely there. Was it the absorption of human bodies in the cramped rooms? Maybe it was. Yes, there were as you would expect all sorts of coloration/projection/distortion/balance issues in the middle and lower frequencies from various systems, some benign, some rather fatiguing. But I was shocked to find that one iPod docking unit had, in those conditions, what I considered to be a better balance than almost all separate speakers I heard. Many exhibitors were using iPads for convenience*.
Maybe there is something to do not only with HF level but HF distribution through the room. I had the feeling that if the SHL5 (just an example) had been there it would have been in a comletely different class for calrity. Maybe we should think about exhibiting - but first we have to increase production capacity.
*After a careful consideration of features/costs/size I recently bought a wonderful Dell Streak. It has the same usability as the iPad, can be carried in a pocket (5 inch screen) and unlike the iPad doubles up as a phone. It's a wonderful machine. But ..... I put it under the driver's armrest when I left home, and when I arrived in Bristol the screen had been destroyed by the downward pressure of my arm on the rest*. And it was being "protected" in a leather pouch. Pity - I'd loaded an RTA into it to check the frequency response of the various replay systems. Beware then, these first generation pads are extremely fragile.
*If you do buy the Dell Steak, beware. Dell are unable to provide service it seems ....
Alan A. Shaw
Harbeth Audio UK
I've had a few listens to your recording Alan, using my laptop-Nagra preamp-Sennheiser setup and I have a few comments. I'm not a technical critical listening expert at all, so bare that in mind - I could be way off.
I noticed that with the Kind of Blue and the Ella Fitzgerald tracks the drum kit sounds seemed to be much quieter compared with the midrange saxophone and Ella's voice (treble roll-off?). It was as if the drums were playing in another room much further away. Also there seemed to be an obscuring drone lower down that made listening to upper bass, lower mid uninformative and unpleasant.
I also thought that Ella's vocal was not even-sounding. It was raised near the higher notes she sang. The saxophone (Kind of Blue) was forced-sounding too.
There was a rock track with, I think Elton John, that to me seemed to have NO TREBLE?! What did you think of that one?
Generally, below the midrange I thought there was a drone-ing indistinctness and one-note bass that sounded like an industrial machine and someone knocking on an old cottage front door, respectively. That was pretty much accross the board, I wonder if that shows a limit in the microphone? However, I did experience fuzzy indistinct bass and blurry lower mids all day at the show - I often walked out within 5 seconds of entering a room.
I think the various tracks do sound different, but it's hard to know whether the running characteristics I hear are evidence for a trend at the show, or a recorder that is not useful for this purpose.
Hope my very amateur opinion is of some use to Alan and people who missed the show.
Interesting feedback. I think we can probably agree that here in 2011, fifty+ years after those hugely well attended public hifi shows that we covered here, that there are still significant differences between the sound of various hifi systems as heard side by side at a modern hifi show. And that is indeed a very curious situation. That half century has witnessed the end of the 78 era, the arrival of the vinyl LP, then stereo LP, FM stereo radio, CD, DAT, DVD, MP3, 192kb sampling systems yet still we have the ludicrous sonic variability between systems. There would be no market today for a car designed in the 50s, nor a TV, radio, or typewriter. In my opinion, this audio stagnation can only be due to three primary factors ....
- Human preference for one sound over another is highly personal, with no absolutes and/or the importance of speech as a quality arbiter has been largely overlooked and the public are apathetic
- Electro-mechanical loudspeakers are by far the most variable, least faithful and challenging component to design in the audio chain and have been so for at least fifty years (today's electro-mechanical microphones have also little changed since the 60s) and will continue to be so
- The room's influence is as great as it ever was and dominates ultimate fidelity
Here is a short commentary that was made about half way through our day.
As you may hear, there are some higher frequency digital burblings due to the highly optimised WMA encoding system used in the Olympus. The lower registers are OK though.
Alan A. Shaw
Harbeth Audio UK
Interesting to hear P speak about the show sound Alan. That recording of your voices shows me that your recorder is quite accurate. It puts the show recordings in sharper relief.
One thing that's been bugging me at hi-fi shows this and last year, is how frustrating it is to be sitting in a demo room 6ft from the component you are keen to buy and hearing it playing the 'music', but seemingly because the room is so unfamiliar to you, all you hear is a pretty unpleasant mess that tells you nothing about the (possible!) sound of the component. Our brain seems to adjust to our room over time and that feels like 'reality' to us eventually. Hear a system in a different room we're not used to and (I find) I cannot really make any realistic judgement. Yet take that same component home and put it in your own system and all is revealed. This affects speakers less than amplifiers of course, because amplifiers are less different anyway. There was a no-nonsense integrated amplifer I saw for the first time in the metal at Bristol, but I had no idea what it may have sounded like because the room, speakers, phonostage and turntable were all unfamiliar to me. Very frustrating.
Every speaker is different and the designer rarely knows what room it will be used in. This must lead to many different presentations when a speaker is used in all sorts of rooms. People then get used to the sound of that speaker in their room and gradually accept that as good enough. It occurs to me that Harbeth speakers are designed with a known room in mind during the design process - a BBC control room. And an intended usage volume level (about 85dB) as you explained to me at Bristol. This means we have a much better idea as customers how to hear the intended performance from them, some of the guess work for the designer and customer is reduced. No wonder Harbeths are considered so natural-sounding by their users. I look forward to my first hearing of your speakers Alan!
Let's take a step backwards and consider what we're trying to achieve when listening at home. The objective we audio fans promote is of 'closing the gap', 'the closest approach to reality', 'concert hall realism' and similar. And how about the recording end of the chain? Physically what recording engineers do is to arrange a few (perhaps only two!) microphones in front of the performers and press the record button. You know yourself that when you attend a live (unamplified, natural sound) concert you can choose to sit anywhere from the first row ground floor right up to the furthest corner in the top row (what we call 'sitting in the gods'). Wherever you sit the sound will be a little or a lot different. And if you are a musician or a stage hand, you'll be in or behind the orchestra hearing another different sound as will the conductor immersed right at the front of the orchestra.
Imagine you attended the rehearsals and the orchestra invited you to move freely around in the hall as they played, seeking what you consider to be the best seat. You'd probably start somewhere near the front - perhaps five or six rows back - and then just for curiosity see what it sound like near the back and then to the side and upstairs. You would be sampling the sound of the orchestra in the hall. What does sampling imply? It implies that you are going to take a guess that experiencing the sound in just a few semi-randomly selected seats that you could draw valid conclusions about the characteristics of a larger group - in this case, the sonic signature of the entire hall. For example, if you first picked seat 5K you could probably assume that 5F to 5P would have a very similar sound.... and possible even deduce that 6F-6P or even 7F to 7P would sound broadly the same. So, out of all the hundreds of seats in the hall, you could probably make life easy for yourself and only experience or sample maybe 20 seats from which you would draw conclusions of the whole hall.
Now let's think about the microphones themselves. As noted above, as you moved around the hall you were sampling the sound using your own ears and brain. The recording engineer picks a handful of microphones and places them at various distances and heights from the musicians to capture the sound at those points in 3D space. The microphone is a pressure sensitive device with typically a diaphragm (the sound sensing part) of about 25sq mm (1 sq. inch). The engineer places those few microphones to sample the entire 3D sound output generated by the performers. It's surely obvious that the selection and positioning of the mics is critical and at best yields nothing more than an impression of the sound waves at just a few square inches of acoustic space over the surface of a huge expanding sonic sound bubble that radiates away from the performers. The resulting microphones therefore capture a woefully inadequate microcosm of the entire performance + acoustic, but it's the best we can do with current technology.- and it's cheap and simple.
So that recording is then replayed in a room. Let's assume that just two microphones were used (left and right channels) and that they drive the left and right speakers. The speaker cones act like microphones in reverse and they pump out sound into the room, reversing the process by which sound from the musicians pressed onto the microphone diaphragms. It must be obvious that the resulting sound as it fills the room is merely what those two, one square inch diaphragms detected at two, one square inch points in space, only a minute sample of the entire balloon of sound generated by the musicians. And yet, that tiny sample is blown-up by the speakers and fill the listening room in three dimensions with sound. But the listening room full of sound bears no resemblance to the 3D sound stage of the performance - it can't be expected to as all the speakers are given to reproduce is a tiny sample of the sound, at the microphone diaphragms, in the recording venue.
It should be clear that the entire business of reproducing stereophonic sound at home is a psychoacoustic trick. And a trick that luckily humans fall willingly into. Ideally, there would be more microphones sampling more points (ideally an infinite number of points) in and around the performers in 3D. And ideally each mic would record a discrete channel of sound to be reproduced over its own dedicated loudspeaker. But if the room so readily screws-up the sound of two speakers struggling to faithfully reproduce the sound of just two points in space, can you imagine the acoustic mess that multiple speakers would generate in the untreated room?
A few posts ago we heard what the mono microphone sampled as the sound of speakers in the untreated listening room - and we have to be very realistic about the degree of fidelity that can be achieved at home without attention to the room treatment. So in my view, to spend serious money on tweaks, cables and the like but to leave the room untreated is chasing the wrong objective.
See for yourself how a handful of microphones are expected to sample the entire acoustic space of EMI's Abbey Road studio here.
Alan A. Shaw
Harbeth Audio UK
Soundfield Microphone. While this system isn't entirely the total answer to all recording problems, it comes close! Of course, like any microphone, the skill is in placing it correctly to obtain the required result. But having done so, it offers a flexibility and degree of control like no other. See this page for more details of the Soundfield Mic and what it can offer.
Have we ever discussed Ambisonics here? Suffice it to say that this is the technique that should have been adopted for surround sound instead of the 5.1 nonsense that Hollywood (and its implicit partner, Dolby) foisted upon us. Tetrahedral Ambiophony, as Ambisonics is sometimes known, is a surround technique which endeavours to reproduce the entire sound field at the point of capture, in all three dimensions. A minimum of four loudspeakers (but ideally rather more) are required, of which some need to be elevated so that the third dimension can be reproduced; a technique termed periphony. Take a good look at this BBC R&D page and watch the truly excellent video.
Would I be correct in saying that the last time Harbeth exhibited at the Bristol show was 2005?
I remember it particularly because it was the first hifi show of any kind that I had been to. The Harbeth room was one of the first we visited and after the 2 or 3 people who were there had finished listening I asked Alan whether he would play some selections from the CDs that I had brought. He duly went and shut the door (as the corridor was already noisy even early on the Friday) and we spent the next 15-20 minutes listening to what I had brought.
The system was, I believe, M30s with Sugden CDP and amp.
My memory is that I could have quite happily listened to it all day - but that it didn't stand out in any particular way.
Do I have any memory of what else I heard on that day? - none at all; and six years later what am I listening to? - Harbeths. So they must have made some kind of positive impression!
Several questions proceed from that and what has been said in this thread:
The most common reason for poor sound at shows that I see put forward is to blame the rooms; listening to the commentary above this suggestion is indeed put forward but then refuted and reasons given.
Going back to 2005, I don't recall a significant amount of room treatment in the Harbeth room, would Alan remember if much was done to the room and further whether he found it particularly difficult to achieve a sound that he was happy with?
Secondly, this thread started with a suggestion about what may be 'the sound of 2011' - that there may be a pervasive trend towards toning down the top end of systems and that in doing so the 'sparkle' and 'musical engagement' has been diminished. The view put forward in the commentary is that this is not down to the rooms at Bristol.
Another request of Alan I'm afraid: if you were given a room at the show today could you take along a pair of Harbeths and doctor them such that they resembled what you heard last week? Without actually substituting different drivers would it be possible to adjust the crossover so that the sound fell in line with the 'sound of 2011'.
The reason I ask is simply to get a better understanding of where the the 'sparkle' and 'musical engagement' are located sonically.
Having been to this show a number of times since 2005 I have some idea of it's overall sound (generally too loud for a start) but not the particular sound this year as I didn't go.
Sorry no, that was not the point of my question.
Alan has, for example, described how it is possible to make a speaker sound 'impressive' in an A-B demo; I was simply asking if he could explain in a similar way how (or even if) a speaker could be 'dumbed down' - I wasn't for one moment suggesting it as a commercial proposition!
Last edited by weaver; 02-03-2011 at 05:54 PM. Reason: sp