Is Audio Science Review going about it all wrong? Or partly wrong? Or all right?

[Blumlein 88]

I think someone alluded to a possible problem with that a few pages back, though: if you start with something distorted in some way (for the sake of argument, let's say 50% THD) and you then add an extra 0.5% of distortion and can't hear the difference, you can't then say that the 0.5% is below the threshold of audibility.

Of course you're talking about less extreme examples than that, but the principle applies: if you are talking about determining absolute transparency, you can't do it using a relative comparison without assuming that you already have transparency in one of the cases. I don't see anything wrong with making such an assumption - but it then prompts the question: if you already know what is transparent and possess it in the form of a 45-year old amplifier available for £100 on eBay and £25 in 'tribute' kit form, why even bother with the test?

There are some leaps of logic in your second paragraph. The assumption we have it in a 45 year old design, and even if we did was it known at the time we did. Certainly the Quad is not unconditionally transparent. The frequency response isn't flat enough. Also the idea we can't get anywhere making relative comparisons without an absolute reference implies knowing we are making progress is impossible. Which isn't the case.

In an absolute state of perfection, an amp would in no way at all alter the waveform. But we know from how they are made, how the devices behave, that some alteration occurs. The question is how much is too much making it audible and how little is enough on one can tell it from the pure original.

So how do we get anywhere. We start with 50% THD for an extreme example. And we build an amp with 10% THD and see if there is a difference. There is so we know we are still above perceptible thresholds somewhere between 10% and 50%. We build an amp at 1% THD. And while there is a difference we can notice that with some music it isn't so clearly different as before when comparing 1% and 10%. So we may be getting close to thresholds. So we build an amp at .1% THD. And with nearly all music no one can hear a difference vs 1%. With some pure test tones there is a perceptible difference. So we build an amp with .01% THD and with no sound we can find does anyone hear a difference vs .1 % THD. So our threshold is somewhere in this range. Yet we never had a pure absolute reference except conceptually.

Of course that all applies to speakers or headphones etc. The series amplifier test does give us something very close to a pure absolute reference in wire vs an active amplifier. The following amplifier would also represent a limit. One can let amplifier B power speakers, and then compare wire feeding B vs amplifier A feeding amplifier B. You can then swap positions of the amplifiers and get some handle on which is closer to wire between the two. Once the two amps are close enough to each other it won't be obvious.

Another approach once our measurements and concepts get things well under control is to add adulteration to signals and see how much it takes to become audible. If we have an amp at -100db THD, and we create a clean signal and a signal with -60 db distortion, the fact the clean signal is not reproduced with absolute perfection isn't a confounding factor should the -60 db signal be perceptibly different. And yet we never had an absolute reference there either.

So we can get in or around real thresholds in these ways.

Now if the 405 mk2 or some other amp is unconditionally transparent vs our hearing thresholds why didn't we quit making new amps? Surely this isn't a serious question. Price, marketing, appearances, convenience, features, a switch from class AB to much more efficient and/or powerful class D amps. Things like that.

 

[amirm]

We seem to be suffering from a touch of looking down the wrong end of the telescope.
The debate regarding ABX is often centered around the equipment. This tends to produce these long threads where various philosophical viewpoints and technical discussions regarding the merits of one design, or another, surface.
Try turning the telescope around and look at the observer, forget about the hardware.
For me the great thing about ABX is it demonstrates the fallibility of the observer, the listener. It's pretty meaningless as a guide to choosing equipment because of all the value judgements the prospective purchaser will apply to their choice. What it does show is a truely fascinating insight into our perception. It's also a lot of fun when conducted in circumstances where 'reputations' are not at stake.

This is a very good point. I often suggest people run a test of MP3 against the original using ABX. If they can't tell that difference, then they should not bother telling me what all they hear that can't be measured. It follows your point that people need to at least once "score" their hearing where they can't know the answer in advance.

 

[amirm]

The Power Cube (or Active Load Box for AP analyzers) is several thousand dollars last I checked (few years ago).

I would get it if it were just a few thousand dollars. Last I checked I think it is $25,000 or some such price!

 

[DonH56]

I would get it if it were just a few thousand dollars. Last I checked I think it is $25,000 or some such price!

That's crazy! Did that include the AP itself? Now that I think about it, it was probably pushing ten years ago when I last got a quote, but I think it was somewhere in the $3k~$5k range for just the cube -- not the big water-cooled version, and not from AP, standalone unit.

Edit: But, since you have a few thousand dollars to throw away, and the cube is out of reach, I'll send you my mailing address...

 

[amirm]

Did that include the AP itself?

No, it is for the cube itself.

 

[DonH56]

No, it is for the cube itself.

I'm in the wrong business... Though I'd guess the demand for them is not all that high. OTOH, that would be a fun retirement project (or whenever I get a life again).

 

[jsrtheta]

That's crazy! Did that include the AP itself? Now that I think about it, it was probably pushing ten years ago when I last got a quote, but I think it was somewhere in the $3k~$5k range for just the cube -- not the big water-cooled version, and not from AP, standalone unit.

Edit: But, since you have a few thousand dollars to throw away, and the cube is out of reach, I'll send you my mailing address...

Peter Aczel used to use the Power cube. I think he used AP as well.

 

[FrantzM]

Amir

You should sell one the 2 pair of speaker cables you use for bi-wiring perhaps one of your USB cable to buy a Power Cube. You get to keep your interconnects, the other speaker cable , the Cable elevators and Power Purifier. You may lose some PRATT in your system but would do us a lot of good.
Thanks in advance

 

[DDF]

Ken Kantor's recommended test load (https://www.stereophile.com/reference/60/index.html ) is (cheap &) suitable for testing amps for variances into complex impedances, while paralleled high power resistors on a suitable heatsink nails it for low impedances. Take the savings and buy a new car to drive to live concerts.

 

[Cosmik]

Certainly the Quad is not unconditionally transparent. The frequency response isn't flat enough.

Evidence? I'm having trouble finding any frequency response measurement for it, but I wouldn't have expected anything but a flat frequency response. Quad deliberately rolled-off the ultrasonic and infrasonic frequencies, but I wouldn't bet anything on anyone here hearing the difference, unless it is because their speakers aren't producing as much IMD in the audible range as a result.

But if you don't like the Quad, then for the thought experiment use some other exemplary-measuring-but-non-pretentious amp from anywhere in the intervening 45 years. The point is that we already have amps that measure well and, as you did above, make a judgement as to their transparency based on measurements alone. Why do you want to do more listening tests if you have already decided what is transparent and what isn't? And, even more logically, that you already have the best amplifiers available for no more cost than worse-measuring boutique amplifiers.

If that latter point wasn't the case, then there might be a logical reason for continuing with the audiophile pastime of listening tests. But if you can't do better than a certain design and its measurements then whether the amp is truly transparent or not is beside the point: it is the best you have got.

 

[Blumlein 88]

Evidence? I'm having trouble finding any frequency response measurement for it, but I wouldn't have expected anything but a flat frequency response. Quad deliberately rolled-off the ultrasonic and infrasonic frequencies, but I wouldn't bet anything on anyone here hearing the difference, unless it is because their speakers aren't producing as much IMD in the audible range as a result.

But if you don't like the Quad, then for the thought experiment use some other exemplary-measuring-but-non-pretentious amp from anywhere in the intervening 45 years. The point is that we already have amps that measure well and, as you did above, make a judgement as to their transparency based on measurements alone. Why do you want to do more listening tests if you have already decided what is transparent and what isn't? And, even more logically, that you already have the best amplifiers available for no more cost than worse-measuring boutique amplifiers.

If that latter point wasn't the case, then there might be a logical reason for continuing with the audiophile pastime of listening tests. But if you can't do better than a certain design and its measurements then whether the amp is truly transparent or not is beside the point: it is the best you have got.

I'm not in disagreement with what you say here. It seemed to me, perhaps misunderstanding you, that you were pointing toward our inability to know what is transparent or at least close enough without an absolute reference. So the post about how we can approach that without such a reference starting out.

Now otherwise, some listening tests help us convince ourselves. Doing some tests just makes it clear in a way reading about it does not. So easy to fool yourself on non-audible things or even audible issues that are thought to be from something they aren't. I'm not like Albert Einstein. He developed theories that predicted bizarre things for the time, and followed them to a reasonable conclusion. His theory hasn't been found wanting over 100 years later. I can do such things sometimes, but I'm more someone who wants to see it corroborated somehow more personally. In the field I once worked in I could do like Einstein and follow the principles where they pointed (others couldn't). Audio and electronics I should and currently do, but I was infected with audiophillia early on, and it takes time to recover.

In fact, since convincing myself, I happily purchase on specs, features etc. You do need to learn how makers game the spec sheet. You do need some reliable measurements from 3rd parties. Or do your own. The only things not transparent now are transducers and sometimes power amps. Excepting gear which has simply been poorly designed and manufactured.

The Quad 405 mk2 does roll-off both ends. There was a time back in that amp's heyday, when I could hear high enough to spot the difference. Not anymore.

I do find the series amp testing interesting precisely for boutique amplifiers that do have a sound that isn't transparency. It is easy to hear some of them have a identifiable signature. Which is good for showing such designs have a coloration instead of higher fidelity. Just as the measurements predict. It wouldn't matter if there were no social aspect to being an audiophile. But there is one, so unless you want to know what you know and not interact with others it is sometimes useful to demonstrate principles that others scoff at.

 

[solderdude]

Quad 405 measurements

Frequency response was flat within ±1.0 dB from 19 Hz to 26 kHz, while the -3 dB roll-off points were observed at 13.5 Hz and 48 kHz. The amplifier has a built-in, sub-sonic filter rolling off low frequencies

 

[Blumlein 88]

Quad 405 measurements

Frequency response was flat within ±1.0 dB from 19 Hz to 26 kHz, while the -3 dB roll-off points were observed at 13.5 Hz and 48 kHz. The amplifier has a built-in, sub-sonic filter rolling off low frequencies

Yes, we need -3 db roll-offs at 2 hz and 200khz respectively. This with your normal class A or A/B amps. The real objective is 20hz-20khz +/- .1 db.

But oh how I wish the review in your link was the norm for gear we read about now. It actually is pretty short and packed with useful and meaningful information.

 

[watchnerd]

For me the great thing about ABX is it demonstrates the fallibility of the observer, the listener. It's pretty meaningless as a guide to choosing equipment because of all the value judgements the prospective purchaser will apply to their choice. What it does show is a truely fascinating insight into our perception.

One of the implications of ABX is its ability to show how powerful sighted / informed perception is.

So if our cognition is such a strong influence on what we hear, then maybe fancy audio jewelry is worth paying for because you think it must be better, because it *looks* better (not because it *is* better) and therefore it really does sound better because you think it ought to.

 

[Shadrach]

One of the implications of ABX is its ability to show how powerful sighted / informed perception is.

So if our cognition is such a strong influence on what we hear, then maybe fancy audio jewelry is worth paying for because you think it must be better, because it *looks* better (not because it *is* better) and therefore it really does sound better because you think it ought to.

That unfortunately is exactly how it works. It's a human condition that doesn't just apply to audio, 'I believe it, so it is so'.

 

[amirm]

So if our cognition is such a strong influence on what we hear, then maybe fancy audio jewelry is worth paying for because you think it must be better, because it *looks* better (not because it *is* better) and therefore it really does sound better because you think it ought to.

Just like medical placebo, the effect is not permanent. Which explains why subjectivists audiophiles are constantly upgrading and changing their expensive gear. The false observation they had during initial evaluation evaporates causing them to seek the next high (testing and believing the next change).

 

[MattHooper]

Just like medical placebo, the effect is not permanent. Which explains why subjectivists audiophiles are constantly upgrading and changing their expensive gear. The false observation they had during initial evaluation evaporates causing them to seek the next high (testing and believing the next change).

I think that's a good point. And I do think that issue is operational to some degree in how some audiophiles switch equipment.

But even so, I have also noticed that it's far from necessarily the case.

I say this because, having been a citizen of many different audio forums over decades, I've seen tons of threads asking about people's speaker history and what they'd settled on as their "keeper/end game" speaker. (I don't pay attention btw to people who are just currently infatuate with their new speaker thinking they have "found the one").

And when those threads/surveys come up the speakers that audiophiles have settled with, expressing long term satisfaction, have been all over the map - from original Quads, to varieties of horn speakers, to omnis, to dipoles, to wide dispersion, narrow dispersion, you name it.

As almost none of these people had the luxury of double-blind testing their speakers, surely some level of sighted (and other) bias may be in play.
But if so, whatever combination of sound/bias was in play, it still worked out to many people's satisfaction in settling on a speaker.

 

[watchnerd]

Just like medical placebo, the effect is not permanent.

As an aside, this isn't exactly universally true.....they're now finding several anti-depressants are equally effective as placebo, i.e. no better than, on an on-going basis.

Probably doesn't work for cancer, but seems to apply to certain drugs used to treat mental health.

Or in other words, where cognition may play an important role, much like in audio.

 

[watchnerd]

That unfortunately is exactly how it works. It's a human condition that doesn't just apply to audio, 'I believe it, so it is so'.

Yes, see above reference to anti-depressants and other mental health drugs.

 

[Bliman]

Just like medical placebo, the effect is not permanent. Which explains why subjectivists audiophiles are constantly upgrading and changing their expensive gear. The false observation they had during initial evaluation evaporates causing them to seek the next high (testing and believing the next change).

Why do you say that medical placebo is not permanent? I have read this book a while back about the subject https://www.amazon.de/gp/product/0857868853/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1 .
And the effect is in many instances permanent in the medical field and is even greater than some regular medical drugs.
With that, I didn't say that you have to throw away your medical drugs, it is a synergy thing. I am saying that the mind is much more complex than first thought. It is a very interesting book.
And there are many things why you can like a system. It can't be so easily disregarded. That's why for example when you eat a strawberry in a field with warm weather and with good friends around will taste sweeter and better in your mind then you ate the same one on a cold day in the winter with bad company around.
Although if you would measure the two, they would be identical. Your mind fills in the rest and that is super important. That is part of life.
And there are many many examples like that.