What is Alan Shaw on about? (is "coloration" unmeasurable?)

[ThoFi]

Allan Shaw is basing his designs on measurements but with a subjective housecurve. And he is using old techniques, but tries to get the maximum out of it within it's limitations that he does not deny. And there is a public for this kind of speakers, that is larger than most here assume. His house curve is made to make speech sound natural, not true to the source. Those are not necesairly the same.

I don't think it's bad, because taste differ, and not everybody wants a superclean neutral speaker. Most want a speaker that is easy to listen to and is fairly close to neutral. And that is what his speakers do. If you are into that or not, is a subjective choice.

He is certainly not into real snakeoil stuff like special cables and so. Nor is he into heavy coloured sounds like tube amps and so. I more like those coloured amps, but i'm aware of it's limitations. Allan says his speakers should work with any amp, and a decent build cheap mainstream amp is often better than those esotheric audiophile amps. Idem with sources. And if people on his forums says it's otherwise he goes against it with real arguments, even if that may imply less sales. He is more on the ASR side than on the audiohile side of the spectrum, but he is not as fanatic and fundamentalistic on that as most here on the forum. And some of his public is into tube amps, even into SET amps (altough it think his speakers are not fit for that, to low sensivity and build for medium to low amplifier output impendance/high damping factor). In one of the discussions he mentioned that he mostly use a Yamaha A/S series amp for his personal setup, and those are fairly cheap and mainstream class AB integrated amps, nothing fancy or audiophile.

but Harbeth, AS, use the snake oil for marketing.
Remember e.g. the SHL5 40th Anniversary model „WBT connectors“ „special inner wiring“…

 

[ThoFi]

 

[tuga]

reality as the yardstick:



tuga,

I agree those are all good points.

On the other hand I think it's fair to recognize that "realistic sound" has been an underlying yardstick for "good sound" for a long time. Not the only one of course, but at least relevant.

In the history of sound reproduction, the starting goal was to recreate the sound of the real thing. That pretty much drove a lot of the sound technology. It's why there'd be live vs reproduced demonstrations (even if they were fudgy), it's why "high fidelity" stereo systems were often sold with the claim of getting people closer to "the sound of the symphony in the comfort of your own living room!"

Once the studio and production effects became a tool of artistic license themselves, then the scope had to widen as to what we were trying to reproduce.
But that still doesn't mean sonic realism isn't a touchstone for good sound. It's been the most common remark when I've demoed my various systems
over the years to guests, how much more 'real' it sounds. Of course it isn't indistinguishable from real live sounds, but relative to what many people are used to, high end equipment can sound "more real" to the degree that is what impresses people.

So I agree sonic realism isn't the whole ball of wax, but it is I think an available yardstick for what many people will experience as "good sound."

(Presumably in the blind testing research for speakers, those that produced vocals with more natural realism were preferred over those where speaker colorations were more obvious intrusions in to the vocals).

"Realism" was one yardstick for BBC testers (together with accurate reproduction) but not for Harman's studies.

Besides, "Realism" is only really a goal with classical music and even that seems it seems to have lost followers; just look at profusion of microphones in photos and videos of many contemporary recording sessions.
Studio mixes are made up in the mixing desk, reality is what comes out of the monitors.

I've come to learn that for audiophiles "realism", like "musicality", is a subjective term which means a sound presentation that a particular person likes; like "good-souding".



The following snippet is from an old EMI leaflet called "The Pursuit of High Fidelity":

 

[Mart68]

It ceases to be the sound of real instruments (or voice) the moment the soundwaves hit the microphone.

Its just not a good benchmark. Plus I note that many seem to confuse 'well recorded vocal' with a vocal that is drenched in artificial reverb.

If it sounds 'right' to you then it is right. If it sounds 'wrong' then you need to change something. This is the extent to which 'Trust your ears' is useful. For everything else - measurements.

 

[ThoFi]

I'm not spreading lies nor rumours - you forget I was in the UK industry for decades and know some of these people personally, as they used to visit us regularly (or me visiting them in the case of ATC)!!!!!! The Spendor situation is a private group primarily focused on the Spencer-designed era models and both Derek Hughes (the son) and the design chap I refer to, post there from time to time and it was especially useful when it came out that a whole run of models had driver surrounds which hardened badly after ten to fifteen years (Spendor cannot service many of these 80's - 90's drivers I gathered) and how to deal with it, at least temporarily with careful application of brake fluid to soften the compound every few months.

Why *shouldn't* it cost a lot to make the 40.3's? I've no idea of how they schedule production time. They're doing well by UK standards, but they're not a huge company with several 'lines' going at any one time, so I gather it's basically a batch of these today and tomorrow, then a batch of this other model for two days after that and so on. They DO have full order books running months ahead, so they're well able to plan production schedules (I've seen the overall figures and they're a business in good shape it seems and definitely NOT on a shoestring as many boutique audio brands are behind the posh adverts).

I still keep in touch and follow UK trends. 'Retro' doesn't sell here so much, even if it does elsewhere (comments on sales of the Linton came from the rep and not snide comments from a disgruntled dealer!). My comments on popular high end brands here is based on experience, not hearsay and I'm not the only one bemoaning Harman's reluctance (or whatever) to promote the likes of Revel or JBL. Mind you, the products I'm talking about aren't especially cheap and cash strapped dealers will tend to specialise in brands they know they can sell, hence the continuing popularity of Naim (with some Focal off the back if it), PMC (they really are nice people to deal with here) and Danish made Dynaudio, which has a huge manufacturing facility I gather and decent test facilities too.

If you're so suspicious that I'm talking [email protected] that's fine, just please don't come on here telling people who 'know what's what' that they're wrong and spreading guesswork! I do my research and TALK TO PEOPLE I know in th e admittedly small UK industry to get my knowledge. Where you live @ThoFi, it may well be different.

I did not say you are spreading rumors or talking crap.

About the 40.3 price. Is still a speaker similar to the other Harbeth speakers. Nothing fancy (e.g. special material) at the 40.3 I guess. What has the schedule to do with the question?

 

[ThoFi]

On the contrary, sticking to the science is as useful as can be, and I thought part of the ASR ethos. Of course, people are free to prefer whatever they prefer, but blurring the difference between facts and opinions is a dangerous modern trend, in my view.

A dangerous modern trend is that non experts spreading thousands of comments.

 

[Willem]

I like my audio system to produce the closest approach to the orignal sound, as Quad once put it. In my own case, I have opted for Quad electrostats, with additional subwoofers and some room eq to tame the room modes. With well recorded music the result can be impressively life-like. I have recently been listening to the Philips recordings of Haitink's Mahler symphonies with the Concertgebouw Orkest, and the sound quality is remarkable, due, in part I guess, to the relatively small number of microphones used by the Philips recording engineers of the time, and the great dynamic range. I also compared these recordings with a few closed miked recordings of the same works, and those sounded hifi rather than realistic, if I may put it that way. So yes, recording technique makes a real difference. I also admit that with other types of music there is no feasible reality check.
As for what a speaker has to be to sound good, I have Floyd Toole's book on my desk (page 350 sq) where he shows that people prefer a smooth downward sloping response with more bass and less treble (what is now often called the Harman curve). As it so happens, this is not that dissimilar to what Harbeth speakers show.
As for Alan Shaw on things like valve amplifers - he has clearly and repeatedly explained why he thinks they are not a good idea given their high output impedance that affects the frequency response with real speakers. Even so, he does not refuse to sell speakers to those who want to use them with underpowered valve amps. The same applies to some audiophile connectors etc because there was a demand for those. However, he never said they sounded better. It is a difficult market with so many nutcases arguing all sorts of unscientific things. The new line of active speakers should at least solve some of this dilemma.

 

[Waxx]

I did not say you are spreading rumors or talking crap.

About the 40.3 price. Is still a speaker similar to the other Harbeth speakers. Nothing fancy (e.g. special material) at the 40.3 I guess. What has the schedule to do with the question?

The price is a big part markup, but that is not strange if you see the market. It's their top speaker in the lineup, and it does sell a lot.

The very popular (but bad designed) Devore O96 uses very standard parts, (Seas A26RE4 woofer with a phase plug added and a Morel CAT378 tweeter) in a very basic but fancy finshed reflex made of MDF with a plywood front and a very basic passive crossover (basicly only a 1st order filter on the tweeter) and still costs similar prices. Harbeth makes their own woofer and mid, and use a tweaked Seas tweeter and a well studied crossover and sounds way better. Just like the Seas A26 kit with the same Seas woofer as the O96 and a Seas tweeter sounds way better and cost a fraction of the O96. Clones of the O96 with better crossovers were build on DIYaudio.com for less than 1K$... But the O96 still sells a lot.

And it's very similar with the high end car or watch market. If there is a lot of demand, prices can go high without stopping the sales. You don't want to know how much the markup is for a Ferrari or a Lamborghini, or a Rolex or a Cartier watch...

 

[DSJR]

I did not say you are spreading rumors or talking crap.

About the 40.3 price. Is still a speaker similar to the other Harbeth speakers. Nothing fancy (e.g. special material) at the 40.3 I guess. What has the schedule to do with the question?

I still think it's a matter of production scale as much as anything. The 40 series is a big box (far more costly to supply in the inevitable more limited quantities - I suspect but admittedly with no proof). the midrange unit sits in its own sealed enclosure and I wonder if Harbeth don't make and fit this part themselves? Again guesswork, but I bet those bass drivers aren't made by the thousand even if the cones themselves are? I doubt we'll ever know exactly how many 40.3's are made each year, but my *guess* would be hundreds rather than thousands. The claim is that many man-hours are needed to build and test each pair, I can vouch for the economic but clumsy packaging used to box them up (I helped to pack a pair of 40.2's last year) and I gather they (used to be) delivered on a pallet if my memory is correct. As with top model cars, the profits on these must be the highest percentage of all and do remember how many of their passive domestic competitors are/were far eastern made at much lower labour rates (?)...

Having tried to defend the 40.3, I've commented here and elsewhere how close they are to the ATC 100ASL Classic, this latter with solid durable amp packs and stands I think. Harbeth are a family business now if the list of directors is anything to go by, I suspect the tech staff now there like to be paid well for their undoubted skills and although AS is most definitely the figurehead and guiding light still, I suspect he must be realising his increasing age and the fact that if the brand continues into future decades, it must move forward with a view that he may not always be there (much as I love ATC, I suspect it's now a smaller concern and I was told recently that BW's son is a lawyer by trade, not an electronics or acoustics disciplined chap as Tim and Billy were...).


I've said all along that I'm no marketing chap and had I started my own retail business as was considered seriously twenty years ago, I'd have failed miserably and probably taken my family with me. AS is great at marketing his brand. The new higher pricing (already discussed as truthfully as I can be) is attracting a different tier of audiophile to the brand, one where (higher, status enhancing) price is important and it's these largely tech-ignorant people that the HUG tries to assist. Yes, many of them use valve gear and I have to say that @Willem and I among others, have been 'edited' or PM'd regarding going too far in criticising some of these amps on the HUG, as all it does is turn these gear enthusiasts away, rather than educating them. If they love a dire-measuring and technically inappropriate as regards output characteristics PrimaLuna amp into the rollercoaster Harbeth impedance loading, then my suggestion is to use that combo while learning what that amp is doing. My continued suggestions to start with something like a Quad Artera Stereo power amp (based on the eternally durable 606 original) are almost totally ignored while they agonise over and defend to the death their Pass 25WPC or similar confection...

@ThoFi, all I can do is suggest you (and I) keep an open mind without hostility to this brand. The products may be 'old school' these days, but there's more modern thinking going on there now, the market for them is established and it'll be interesting to see what they come up with in the next few years - AS is VERY well aware he can't stay with this range forever and that in current form, it's gone about as far as it can without radical re-design. I doubt that any new models will give Neumann or Genelec any headaches, but they may get Dynaudio to look over their shoulders, as this brand is moving some quite serious and good looking (conventional with grilles on) DSP-Dirac/wireless capable actives into the domestic arena and at not stupid prices either when taking dealer margins into account, the £8600pr model falling easily into the SHL5+XD plus half decent posh-amp price category.

Interesting times in cash-strapped Europe methinks, where most of us are stung with ever higher fuel and energy bills, yet the Paris show goes ahead, well attended (by what we call tyre kickers?) and full of expensive bling-boxes and lavish displays. Apologies to US based readers, as I don't know how the stuff above compares - you have speakers available we've barely heard of and I weep when I see how much dearer US imports of our gear is and how much the likes of Schiit has gone up in recent times, assuming it's in stock of course ..

P.S. I wonder how much the PMC 'Fact' models actually cost to make, let alone the 20-25 range. They don't seem to perform that well either, although the sins are of omission (deep crossover region suckouts followed by a tinsel peak further up) rather than upper midrange edginess.

 

[Willem]

I cannot quote the line in Tuga's post on the importance of scale, but I could not agree more. Over the years I have learned that one needs both a large listening room and serious low frequency extension, both of which I now have in my main system. The desktop system in my 18 m2 study (just Harbeth P3 ESR speakers) is fine for small scale music, but unevitably unconvincing with e.g. symphonic repertoire.

 

[killdozzer]

Just 2 cents, no big expectations...

Once I tried to explain how I see the recording process as a one way valve - it delivers music to you, but you can't peak behind it to reach actual performance and see what it looked like. All you get is what is recorded and this is why I think the recording is the blueprint for reproduction and not the original performance. I don't see the recording as a gateway to the original performance. It only shows how it has been recorded. But, since I imagine there was some sort of agreement/consensus about what artist and the rest found to be pleasing, the recording is what it's about for me. Furthermore, if we talk about electro music, there's no real archetype of those sounds. It's what artist and all involved decide it is.

Another topic in this thread - "measurability" of sound waves; what I found strange about it is how rarely people see the logical fallacy in thinking "we can't measure everything". Those features they deem "unmeasurable" WERE recorded at one point. They found their place on the recording by the virtue of the mic. This would literally mean that there is a device that reacts to all those properties they think we can't measure. It is the microphone. If microphone didn't react to those, you wouldn't have them recorded. Timbre or the rest of them (space, reflections...), they are all registered by the mic. Microphone is man made. Isn't it funny how we were able to produce a capturing device, but not a measuring device?

Mr. Shaw didn't sway me, I'm still in the "all features of sound can be measured" camp. I could be proven wrong although not by someones' lifestyle or worldview. what I think we still can't fully measure is the emotional response to sound in its full capacity, but I don't think it will necessarily stay that way.

 

[Phoney]

He posted my last, and responded in two posts. I'm considering my response. I think this is evasive on the subject of measurement. He says "not really" which *might* be in response to my question about "are you asserting that coloration is too subtle..." question. He points out that people are sensitive within a certain band, but doesn't allow that we can still measure the 'coloration' within it. He doesn't really answer my questions. Thoughts on polite responses?

I'd like to ask him what is missing from a CEA-2034 and multi-level distortion panel, in terms of catching 'coloration'.

View attachment 192176

View attachment 192178

How does his speaker have low coloration in general if its not about the physical gear being colored objectively, but about our brains reacting differently to sound and our ears having different sensitivity across the bands? Does his speakers magically adjust to the listeners brain and ears?

 

[ThoFi]

I still think it's a matter of production scale as much as anything. The 40 series is a big box (far more costly to supply in the inevitable more limited quantities - I suspect but admittedly with no proof). the midrange unit sits in its own sealed enclosure and I wonder if Harbeth don't make and fit this part themselves? Again guesswork, but I bet those bass drivers aren't made by the thousand even if the cones themselves are? I doubt we'll ever know exactly how many 40.3's are made each year, but my *guess* would be hundreds rather than thousands. The claim is that many man-hours are needed to build and test each pair, I can vouch for the economic but clumsy packaging used to box them up (I helped to pack a pair of 40.2's last year) and I gather they (used to be) delivered on a pallet if my memory is correct. As with top model cars, the profits on these must be the highest percentage of all and do remember how many of their passive domestic competitors are/were far eastern made at much lower labour rates (?)...

Having tried to defend the 40.3, I've commented here and elsewhere how close they are to the ATC 100ASL Classic, this latter with solid durable amp packs and stands I think. Harbeth are a family business now if the list of directors is anything to go by, I suspect the tech staff now there like to be paid well for their undoubted skills and although AS is most definitely the figurehead and guiding light still, I suspect he must be realising his increasing age and the fact that if the brand continues into future decades, it must move forward with a view that he may not always be there (much as I love ATC, I suspect it's now a smaller concern and I was told recently that BW's son is a lawyer by trade, not an electronics or acoustics disciplined chap as Tim and Billy were...).


I've said all along that I'm no marketing chap and had I started my own retail business as was considered seriously twenty years ago, I'd have failed miserably and probably taken my family with me. AS is great at marketing his brand. The new higher pricing (already discussed as truthfully as I can be) is attracting a different tier of audiophile to the brand, one where (higher, status enhancing) price is important and it's these largely tech-ignorant people that the HUG tries to assist. Yes, many of them use valve gear and I have to say that @Willem and I among others, have been 'edited' or PM'd regarding going too far in criticising some of these amps on the HUG, as all it does is turn these gear enthusiasts away, rather than educating them. If they love a dire-measuring and technically inappropriate as regards output characteristics PrimaLuna amp into the rollercoaster Harbeth impedance loading, then my suggestion is to use that combo while learning what that amp is doing. My continued suggestions to start with something like a Quad Artera Stereo power amp (based on the eternally durable 606 original) are almost totally ignored while they agonise over and defend to the death their Pass 25WPC or similar confection...

@ThoFi, all I can do is suggest you (and I) keep an open mind without hostility to this brand. The products may be 'old school' these days, but there's more modern thinking going on there now, the market for them is established and it'll be interesting to see what they come up with in the next few years - AS is VERY well aware he can't stay with this range forever and that in current form, it's gone about as far as it can without radical re-design. I doubt that any new models will give Neumann or Genelec any headaches, but they may get Dynaudio to look over their shoulders, as this brand is moving some quite serious and good looking (conventional with grilles on) DSP-Dirac/wireless capable actives into the domestic arena and at not stupid prices either when taking dealer margins into account, the £8600pr model falling easily into the SHL5+XD plus half decent posh-amp price category.

Interesting times in cash-strapped Europe methinks, where most of us are stung with ever higher fuel and energy bills, yet the Paris show goes ahead, well attended (by what we call tyre kickers?) and full of expensive bling-boxes and lavish displays. Apologies to US based readers, as I don't know how the stuff above compares - you have speakers available we've barely heard of and I weep when I see how much dearer US imports of our gear is and how much the likes of Schiit has gone up in recent times, assuming it's in stock of course ..

P.S. I wonder how much the PMC 'Fact' models actually cost to make, let alone the 20-25 range. They don't seem to perform that well either, although the sins are of omission (deep crossover region suckouts followed by a tinsel peak further up) rather than upper midrange edginess.

Speakers are luxury products and to me the the price of the 40.3 brings the highest margin. (analogy to cars)
And again, and to me, it is absolutely not correct to say such things like „oh it needs so much effort to produce the 40.3……“.

I do not understand why some people are freaking out about valve amps and their effect on the FR.
Yes we do know that the valve amps do „follow“ the speakers impedance curve And this results in little different FR of the speaker. But hey! Every time when I look at my extremly wobby in-room FR measurements I am no longer afraid of the influence of an valve amp.

 

[DSJR]

Surely we need to keep the variables in the playback down so the only major effect on presentation/reproduction via a loudpsealker is the room interface? Adding in a fixed-eq graphic equliser via a *typical* domestic enthusiast valve amp is not a good idea. been there, done that, had the charm and grew bored of it all sounding much the same,,,

The onus now is on YOU to prove the 40.3 is not a labour intensive or expensive speaker to make! Go on, you make your own and tell how easy or not it is. Sorry to be excessively blunt, but at least I tried to mention an alternative and roughly similar price which I've personally seen being made and tested, albeit in slightly earlier form! Maybe AS doesn't want to do the 40.3 any more? I mean, the originals were custom installed into the professional rooms personally I seem to remember and it was only really the 40.1 that was released domestically (they cost twelve grand or so in the UK ten years or so back).

Seriously sir, you don't seem to have much good to say about the brand, so why not leave it and us users alone and concentrate on cheaper, more ASR-friendly prosumer speaker brands which often cost rather less as well. I mean, I'm hankering after returning to a good active model at less money than an ATC for example, but I have an awful load of old 'stuff' to sell off first to make that possible and I'm too damned well attached to most of it to make that change currently! (we hoarders are very odd people and tend to 'bond' with the stuff we have, something my wife doesn't understand at all - I've told her to have a bloody huge bonfire after I've gone )

 

[Elmar.F]

What an odd thread. Meanwhile it should be common knowledge that Harbeth is one of the few manufacturers, if not the one and only, whose speakers reproduce recorded speech in the best and truest sense. Not one speaker brand matches Harbeth in this regard probably.

 

[ahofer]

What an odd thread. Meanwhile it should be common knowledge that Harbeth is one of the few manufacturers, if not the one and only, whose speakers reproduce recorded speech in the best and truest sense. Not one speaker brand matches Harbeth in this regard probably.

That is not an objectively supported fact, and there’s no particular reason it ”should be common knowledge” unless you are in Harbeth marketing. I’m a big fan of Harbeth and have them at home. But I also own a pair of Revel 228Be and I don’t think Harbeth has any edge on Revel for reproducing voice. In fact, if you are out of the sweet spot, you might find the Revels quite a bit better. My experience with KEF Ref 5 was similar.

What IS fact is that Alan Shaw prioritizes voice reproduction in his design process, and advocates that vociferously. That’s different from being the sole leader in execution.

This thread has been about separating Shaw’s marketing talk from his design achievements, and this is a good example.

 

[ahofer]

How does his speaker have low coloration in general if its not about the physical gear being colored objectively, but about our brains reacting differently to sound and our ears having different sensitivity across the bands? Does his speakers magically adjust to the listeners brain and ears?

I’m just going to be blunt and say (my supposition is) he knows he can’t win a measurement race, so he is adding some branding fiction to make his product seem unique.

Harbeths are great speakers, certainly the best implementation of the lossy-cabinet/wide baffle speaker I’ve heard. This design has its pluses and minuses, but he’s got it to the point where it competes with a select few designs in terms of a linear on-axis presentation and low distortion. His focus on dealing with *audible* problems is laudable, and allows him to make some very effective compromises with the old design. I think his concept of “real speakers in real rooms” also ends up in a product that is satisfying at home and sounds good with many recordings. Notice how he never talks about dispersion in his own speakers though? Because he can’t win that one on measurements. And it is an issue. My SHL5+ compete pretty well with my Revels on axis, but not in terms of placement and seating freedom.

 

[Elmar.F]

What IS fact is that Alan Shaw prioritizes voice reproduction in his design process, and advocates that vociferously. That’s different from being the sole leader in execution.

Do you know of a speaker brand other than Harbeth who prioritize voice reproduction in the designing process?

 

[ahofer]

Do you know of a speaker brand other than Harbeth who prioritize voice reproduction in the designing process?

I’ve seen Focal and Wilson brag about it. But again, let’s keep the company’s statements separate from established facts. After all, Wilson has to be one of the most over-hyped brands out there.

https://www.wilsonaudio.com/pdf/brochures/wilson-audio-full-line.pdf

 

[fpitas]

I’ve seen Focal and Wilson brag about it. But again, let’s keep the company’s statements separate from established facts. After all, Wilson has to be one of the most over-hyped brands out there.

https://www.wilsonaudio.com/pdf/brochures/wilson-audio-full-line.pdf

Yes. It's one thing to like or prefer a speaker tuning, quite another to buy into rather cynical marketing.