Misleading Measurements

[fas42]

Putting on an objectivist's hat, I would argue that it's a https://en.wikipedia.org/wiki/Factor_of_safety, an engineering safeguard to ensure that in all possible circumstances that the audibility will always lie well below what's borderline audible.

[fas42]

53 minutes ago, The Computer Audiophile said:

I don’t see how this relates to objectivists disliking that which can’t be heard but also publicizing that which can’t be heard when it fits an agenda. 

 

It's also an indicator of quality - that the design and manufacture is to such a high standard that they are well below what's necessary for what they believe to be "the threshold or level of audibility" - implying that other qualities which are not so easily measured are of a commensurate standard.

 

As an objectivist, I would want the best that's possible in performance, for peace of mind - I don't feel that I'm missing out of anything; that there's no chance that there is a better system out there that has better specs, which may give better subjective results, 😉.

[fas42]

There is absolutely nothing "below the level of audibility" that's important, in a direct sense - what it may tell you is how robust the component is, when faced with interference factors, and similar possibly degrading influences.

 

What I'm sure deeply offends many objectivists is that, as claimed by many, the sound of a rig can dramatically change when absolutely nothing is done which will impact all the measurements they like to rely on - many of them resolve this dilemma by insisting that "it's all in your head!!" ... they are not happy that variables that they don't have a firm grasp on are that important - so, sweep it under the carpet is the strategy they will to some degree cling to.

 

Current active speakers I'm playing with are a case in point - I can have the presentation range from "run out of the room with hands over the ears" SQ, up to approaching magic territory, simply by altering the electrical environment external to the units - as has been the case to some degree with every rig I've dealt with, over decades. Now, what the use of measurements of such items, in some laboratory setting, if those behaviours are not tested? This consideration would be highly disturbing to objectivists, so at least some will insist that such things are impossible, and will "over measure" what they think is important, as a type of compensation.

[fas42]

Or ... objectivists are by instinct reductionists, https://en.wikipedia.org/wiki/Reductionism, and the human hearing system is inherently holistic in operation; subjectivists by contrast are attracted to https://en.wikipedia.org/wiki/Antireductionism

[fas42]

42 minutes ago, bluesman said:

If something is inaudible in isolation but has audible effects on other factors, its presence is audible even if it makes no sound of its own.  I think this is directly important.  Look no further than a silent person on a creaky step - you hear the normally silent step because an inaudible person stepped on it.  DC is silent - but a voltage drop that makes its way to an audio signal as a DC offset can affect SQ.

 

Which is a variation on what I'm saying - measuring the output of the system in the electrical sense, and everything there being "below audibility" is what I was referring to - measuring somewhere within a component, or at a link in the chain is entirely different; because there the causal links that you refer to may exist. If the linked to component is extremely sensitive to some parameter that is well within reasonable limits from the preceding component, then everything changes.

 

[fas42]

1 hour ago, JoshM said:

I’m interested in measurements to catch major issues and to understand why we’re hearing what we’re hearing. I’m not interested in them as an arms race with no meaningful purpose. 

 

Now, this is what intrigues me ... why are people interested in measurements "to catch major issues" ?? If the SQ is audibly faulty then you have a significant issue - getting numbers is irrelevant; what needs to be done is to determine what is causing the problem ...if you take a misbehaving TV to a service technician he is not going to waste his time extracting numbers - he can see the picture is not up to scratch, and the only use of numbers is to help diagnose internal to the circuitry - his goal is to locate the fault, and decide whether to chuck the set, or replace some misbehaving parts.

[fas42]

What human hearing is very good at, is being able to focus on fine detail, amidst high levels of background or attendant sound - the cocktail party effect. Live music always has this quality, of there being "tiny things happening", while there is a very intense overall sound level - which we humans have little difficulty picking up. Unfortunately, generally audio rigs are poor at doing this, for a variety of reasons; which is why there is a common belief that playback can't mimic live music making - but, this failing is merely another form of distortion, which can be eliminated.

 

It would not be easy to simulate this, say in Paul's software, because it is a highly non-linear behaviour - being able to measure the levels of this type of distortion would be good, but I can't recall coming across any research looking at this - a common phrase of what people hear when a system does this is that the sound starts to compress; a capable replay doesn't do the latter, a common phrase for that is that the reproduction is effortless.

 

Currently, tweaking is the normal way to deal with this fault in typical audio replay; because the industry as a whole doesn't recognise this misbehaviour.

[fas42]

1 hour ago, ray-dude said:

 

To the ITD, you are certainly better read on this than I am, but isn't that related to localization of sound source by the time difference it takes for sound to get to each ear?  With stereo music reproduction, we actually don't want to hear the speaker driver, we want a sound stage projected before us.  My phase analogies were related to that sound stage projection, not localizing where a speaker may be.  

 

That being said, I am google-level ignorant on the ITD measures.  I have no idea if the psycho acoustics are the same mechanism between ITD for sound source localization, and reconstructing a sound scape from the aggregate phases of the sounds we are hearing.  I suspect our brains are doing a lot of interpolation/projection for the later functions, just because our brains are really good at casting things in a way where it is easier to digest/interpret (Coltrane obviously isn't standing in front of me, but damn does my brain gets a lot of juice when it sounds like he is...sign me up for more of that kind of self-delusion!).

 

 

 

One area where misleading measurements certainly figure is that the quality of the audio reproduction is largely ignored, say in worrying about the precision of phase alignment. The rule in my book is very simple: the lower the overall, subjective, accuracy of the replay, the more factors like perfect phase alignment will matter, as regards getting impressive soundstaging happening - this was something I learned over 3 decades ago; if the SQ falls below a certain, quite precise standard, then the holographic qualities of the presentation evaporate - the mind no longer has enough information to 'interpolate', to maintain the reconstruction within one's head. It's quite staggering how capable the brain is in reconciling "what it all means" - I was amazed at the time at how powerful the illusion was - this can be pushed to the point that it becomes completely impossible to localise the speaker, no matter how ridiculous a place you move your head in an attempt to "hear the drivers working".

 

The simplistic measuring of our ears responding to phase anomalies completely misses the point that the brain has remarkable ability to interpret the data, when it has meaning - that is, it's a reproduction of music being played. With the obvious corollary ... surely this behaviour of human hearing should be exploited, to achieve maximum value in the situation?

[fas42]

16 minutes ago, ray-dude said:

This is very true.  I experience a very abrupt phase transition where some recordings suddenly seem very real.  Some recordings never get there, others get on the "good side" of reality mountain after some significant effort or enhancement.  I've attributed this to my brain getting better at fabricating the illusion for me, once appropriate auditory cues are there.  Conversely, for recordings on edge of that transition, it is a very sensitive "tell" when I've done something to disrupt some aspect of music reproduction.   In a way, much of my audio optimization journey has been about getting more and more of my music library to feel like it is "real" and in the room.

 

 

Spot on. You're following the very methodology that I've been using for decades - nice to come across fellow travellers, 😁.

[fas42]

4 minutes ago, pkane2001 said:

 

Realistic reconstruction of a complex soundscape with precise object placement is a much more complex topic that involves head tracking, head-related transfer function (HRTF), reflections/reverb, dynamic frequency effects, as well as ITD and ILD. Simple two channel/two speaker stereo can't reproduce a complex soundscape in a truly believable fashion.

 

 

 

That's the book learnin' explanation - but is not the reality. Anyone who has managed to evolve audio playback to the necessary quality knows what happens - it's quite trivial for an immensely complex soundscape to be perceived as being truly believable, at this standard of SQ.

[fas42]

38 minutes ago, pkane2001 said:

 

Sorry, Frank, but that's false. Not just from books, but from my own experience. If you've not heard a properly instrumented 3-D audio with HRTF adjustment and head tracking, you should try it. There's no comparison to your "imagined" soundscape.

 

Don't you appreciate that there are multiple ways "for the mind to be tricked"? It can be done with the deliberately manipulated process you mentioned - or by having the clues available by a very high standard of playback. The soundscapes "inside the skull" are the same - if your mind decides that what you hear is the "real thing" then it hangs on to the illusion; and won't give it up readily ... you see, everything is "imagined" - our minds, internally, don't put instances "created by science" on a pedestal.

[fas42]

32 minutes ago, pkane2001 said:

 

I do, and I also know that each mind and ear combination is different. When you claim you hear "realistic" soundscape it means absolutely nothing to me, since I don't know what it takes to trick your mind into believing something like this. Maybe your imagination is much better than mine. Or maybe your definition of realistic is not the same as mine. Or maybe your brain never adjusted to use your HRTF for sound location and compensated by using some other method, maybe ITD/ILD only. Or...? I can't see (or hear) into your mind Frank, so I can't tell what you can hear and what is enough to fool you.

 

Why I'm certain that it has nothing to do with 'conscious imagining' is because of the behaviour of that first good rig, nearly 35 years ago. I could easily have the system be in, and out, of the right quality zone, any number of times during the day; I was in full control of a sequence of actions that allowed this to happen - when "out" of the necessary quality zone, all I got was quite decent stereo, the same thing that most people get ...

 

Other people around me have appreciated the difference it made - which confirmed that I was not alone in hearing this ... people who couldn't give a damn about audiophile concerns just enjoyed listening to it; it "ticked the boxes".

 

32 minutes ago, pkane2001 said:

 

With my system, I hear a very good approximation to a soundstage with very good tonal reproduction, with excellent dynamics. I hear depth and position of instruments, and I enjoy it immensely,  and have for about 20 years with only minor changes. And yet, it's not producing a real soundscape. I only occasionally (and mostly on binaural content) hear voices and noises that startle me, surprise me to the point of thinking it's coming from inside the house rather than from the audio system. I almost never hear the sound coming from my speakers, unless the sound is panned to be exactly at the speaker position. Most of the left-to-right sounds come from in between, with some coming outside the speakers. The soundstage depth is amazing on some content, and goes a bit in front of the speakers on other. But sounds never get close to me. They never envelope me to the degree that I can move my head around and feel like I'm there. That's what proper 3-D soundscape reconstruction does. It's the difference between a very good 2-D photograph and the 3-D world. At least to me.

 

 

 

This clarifies to some degree ... direct sounds, the actual noise makers, never get close to one; they always remain beyond the vertical plane that the speakers lie in. What envelopes one is the reverberation of that direct sound from other surfaces in the listening area - exactly the same as listening in a concert hall; you never feel that you are in the middle of the orchestra - that's always in front of you, but the sound of the music fills the space you're in ... it's "immersive". I have no interest in having some instrument sounding like its coming from behind me; that's gimmicky, and has nothing to do with what listening to live music is like.

 

An almost perfect analogy is slicing off your listening room vertically at the plane of the speakers; and moving the remaining area, where the listeners are, to the place where the performance was being recorded. One can move around in the listening area as much as you like, and one always hears what an actual real world version of building something like this would be like.

 

 

[fas42]

4 minutes ago, John Dyson said:

Tweak-tweak-tweak works okay in primitive situations -- but in more complex multi-dimensional situations (almost all EE/DSP stuff is like that), can become a misguided allocation of time.  A little learning/planning ahead of time, in many simple applied technology cases (home audio systems) are well worth the investment.

 

Many of you know that I am in tweak tweak tweak hell right now -- and I would never counsel any friend of mine to use the design-by-tweak technique for any complex project at all.

 

John

 

 

Why the "tweak-tweak-tweak" method works in the audio game is because the core engineering is good - to use your analogy, Tesla did all the crucial research as well as what was needed - but, he was a bad 'finisher' ...it needed an Edison to come along and finesse the solid design elements already in place; tidy up the loose ends.

 

Which is why "turds can be polished" - the keys parts are good enough to do the job, but require extra, additional, knowledge driven refinement to bring out what is inherently possible - the end result can easily outperform, subjectively, an unfinished "Tesla effort" ...

[fas42]

6 hours ago, Jud said:

 

Though there are plenty of opportunities to introduce noise into a system involving a number of boxes and cables that include both of these. I would love to see more work done on system topology, components and cabling and how all of these affect system noise. I've seen a few articles about topics like grounding, but I haven't had good luck finding detailed and precise information on how to configure a system and then how to measure the results in easy to digest form for the audio layperson (i.e., non-electrician).

 

I would suggest looking at posts by member jneutron in the diyAudio forum for some good insights, if you are not familiar with him  - his bread and butter is making circuits in research facilties where huge currents are involved work properly, so has real world knowledge of what is needed. An example of his thinking, https://www.diyaudio.com/forums/the-lounge/348308-john-curls-blowtorch-preamplifier-iv-post6086931.html

 

[fas42]

6 hours ago, pkane2001 said:

 

If you want to know how anything in the chain is affecting noise in the audio system -- measure the output of the system. The output of a DAC is what is fed into an amp. If the noise is not detectable there, then it's either non-existent, or it's properly filtered out by the receiver circuits.

 

How I detect noise is by listening ... it's trivially easy for me to make a tiny adjustment to the electrical environment of the home in which one is listening, and hear the variation in SQ. Whether one wishes to call this noise is up to the individual, but what it really says is that the playback chain is not sufficiently robust to reject this input - why you should want to measure such I don't quite see; but if you want to make it really obvious, in some numbers, just hook up, say, a working arc welder into a nearby socket - that will give you plenty of juicy data to work with, 😁.

[fas42]

18 minutes ago, John Dyson said:

Since my hearing has never worked beyond perhaps 22kHz, and currently isn't even that good, I find that using only my hearing to detect noise problems - (such problems not always being in the audible frequency range), isn't adequate in a lot of cases.   Being able to use an objective measurement of some kind -- often the direct measurement being further processed so that the details are more clear seems to be more effective when available.  Properly presented details about the 'noise' can be helpful to pinpoint even unforseen problems.   Sure, there are cases where a direct measurement with technology might be too difficult, but having objective/technologically aided measurements available as a primary means can eliminate missing lots of potentially non-audible impairments.   Out of band impairments can indirectly cause in-band audible problems.

 

If the perceived sound changes, then there is an issue - the mechanism that allows this to happen may be hard to track down, and if the primary goal is to improve the SQ then IMO resolving that should be the first focus. In the long run a full understanding will of course be highly desirable - but a lot of energy may be wasted prior to that dealing with red herrings, simply because the latter may be easier to get numbers for.

 

Ultimately, what's going on when people complain that the SQ is not good enough will totally measurable ... we just haven't got there, yet 🙂.

[fas42]

Typical measurements are misleading, because they don't point out the weaknesses of the product - those put out by the manufacturer are a joke, because they merely form part of the marketing ... my 35 year old power amp had 10 times as much information printed in the included manual as compared to current units held in high regard - the word "pathetic" comes to mind.

 

No-one measures what really matters when the 'transparency' of a system reaches a certain level - things like, the ability to reject external interference - the consumer is completely blind to how it will really behave, until he has it in situ. Quite absurd, so no wonder the industry is such a mess,when looked at by an outsider ...

 

 

[fas42]

For some bizarre reason the concept of weighting barely registers in the audio world - in many fields this is a vital component for assessing the relative importance of various factors; a brilliant number in some area is almost irrelevant, because it has a low weighting in the overall. But audio is caught up in the 'purity' of specific numbers, in themselves - no wonder people have no idea how good some component will sound; when starting from this low base in measuring technique.

[fas42]

6 hours ago, pkane2001 said:

This forum isn’t big enough for me to describe all my equipment

 

Three systems (one is 5.1) plus a ton of DACs audio interfaces, headamps, amps, headphones.
 

WiFi, Ethernet and USB are all in the digital audio path between the three floors. No PC in the main listening room.

 

Which tells me a thing or two ... 😁.

 

First thing I would do if I visited you to listen, seriously, to your "best rig", would be to go throughout your home, and pull the plug out on every single item that wasn't actually needed to make that primary rig work - and see what that gave me ... that would give me a baseline for assessing what the potential in SQ was, 🙂.

[fas42]

2 minutes ago, pkane2001 said:

I’m afraid that’s quite impossible. All the plugs are removed and the bare wire soldered in. It cannot be pulled out... you know... to improve SQ

 

 

... you do have those panzerholz headvices concreted in place, to guarantee the perfect alignment of the ear canals to the sound waves, of course ... ?

[fas42]

5 hours ago, pkane2001 said:

 

The DAC is Holo Spring, balanced interconnects to Pass Aleph monoblocks leading to PSB Stratus Gold speakers. Room correction curve applied.

 

Oh dear ... the Thiels, which were a match made in heaven for the Pass, are consigned to the darker regions ... ?

[fas42]

4 hours ago, pkane2001 said:

 

Bought them when I was younger and more suggestible  I've had the monoblocks for about 25 years, and still like them. I do have a few other amps that are sometimes swapped in, when I want better measurements

 

But they don't sound as good, so they then get swapped out again ... 😁

[fas42]

4 hours ago, manisandher said:

 

 

Is it that our threshold of audibility is below the level to which we can measure THD/SINAD, etc., or... are we simply measuring the wrong things?

 

Mani.

 

The latter.

[fas42]

4 hours ago, pkane2001 said:

 

Did you participate in Archimago's THD blind test? THD/SINAD are not the wrong thing to measure, they are a very simplified, average number approximating a much more complex non-linear behavior. As a first order approximation, these might be useful, but certainly not enough for a careful analysis of a device. Other measurements are required for that.

 

As is obvious to everyone who takes subjective SQ seriously - my current budget active speakers' biggest issue, so far, is their sensitivity to the mains quality - which has been a major part of the optimisation of every rig I've had ... something which no-one ever, ever measures  ...

[fas42]

14 minutes ago, pkane2001 said:

 

Really? Mains interference is never measured? Maybe it's not something a THD or SINAD number by itself will reveal, but certainly distortion and noise floor plots will show it.

 

 

Of course it can be measured ... but no-one actually does it ... give me a single link, to any online document, that has a specific measure of a particular component's performance in this one thing - no, "part of the overall picture" stuff, please!