There is nothing holy about the signal

[Sgt. Ear Ache]

an excellent summary of ... all my looong posts
Just a small addition: that's also my definition of "better"

sure but the sound you like is a pretty useless measure by which anyone other than you can set up a sound system right?

 

[Killingbeans]

To whatever 'bits' are in the recording? Good luck with the accourate reproduction of schitty recordings (and many/most are).

Yes. Warts and all for me please

If you want to convince/move people, tell them something like "try this $500 DAC, you'll save quite a bit of money if it sounds same as good as your $5000 DAC". That's not exactly a succes-guaranteed line, but I bet that your fanclub will grow much faster

That's putting it mildly. Sounds more like a dice throw between god knows how many different kinds of confirmation bias.

 

[Roland]

I assume that double blind tests are also required because expectation bias can occur after room correction, for example, as much it can when getting electronics boxes which measure even “better”.

 

[Sgt. Ear Ache]

I assume that double blind tests are also required because expectation bias can occur after room correction, for example, as much it can when getting electronics boxes which measure even “better”.

which sorts of situations are you referring to with boxes that "measure even better?" In my experience, the DBT thing comes up mostly in response to claims people make about being able to hear differences between devices (usually dacs or amps) that are essentially measurably the same - or at least so close to the same that it becomes very unlikely that any differences could be reliably noted. It's usually got nothing to do with "better." It's about trying to confirm that a person is actually hearing any difference at all...

Of course, blind testing can also be used to establish broad preferences too as in the HK speaker research, but I don't really think that's the primary use case commonly seen on these forums when people ask "did you do a blind test?"

 

[MattHooper]

I assume that double blind tests are also required because expectation bias can occur after room correction, for example, as much it can when getting electronics boxes which measure even “better”.

Fair point, in principle it's possible after putting effort in to measuring/calibrating and getting those numbers "right" you may perceive the room correction sounds better to you, but double blind may not be the case.

 

[lashto]

sure but the sound you like is a pretty useless measure by which anyone other than you can set up a sound system right?

True. But I avoid all 'issues' by not pretending that my sound is "the sound".

And most probably I won't be very happy if everyone had my sound ... all those hours and money and the learning and the tweaking ... just to have the same 'best' sound as everyone else .. grrr

 

[Curvature]

The signal is holy and audibility is divine. Put scientifically validated audibility bands (20 - 60 yr old) on every graph where it is valid and stop the nonsense of things you cannot hear being important.

We can make reasonable guesses for source and signal gear as long as speakers aren't too sensitive.

We have decent guidance for speakers, with many blindspots.

There's a long road ahead.

 

[MattHooper]

And my point is that there is a 'ballpark of accuracy'. So yes there is no perfect reproduction but there is a zone within which it is closer to perfect and beyond that zone it is far from perfect.

It's my view that getting within that zone matters - because otherwise too many recordings begin to sound bad and the system begins to determine what recordings we can listen to with some amount of pleasure.

In other words it's still the wild west but there is a sheriff in town. He can't be everywhere and police everything but just him being there keeps enough of a lid on things that the townsfolk can go about their business without automatically being shot and robbed.

I agree with your general point here.

IF we are talking about the goal of trying deal some way with the circle of confusion:

It may be a mistake to abandon this, by making "perfect" the enemy of "the good."

It's reasonable to affirm these two propositions:

1. Perfect accuracy to what was heard in all the different recording studios is impossible.

but

2. We can still move further or closer to that goal.

To exaggerate: if you've blown both midrange drivers in your speakers, you are most certainly hearing something less like what mixers using functioning speakers heard, than you were before you lost those drivers. There's all sorts of ways in which you could degrade a signal that would clearly move in a direction opposite to what many heard in recording studios. Which means moving in the other direction...fully functioning speakers, ensuring flat frequency response so that IF there is the content in the recording you'll hear it, is at least moving in the right direction.

It's my view that an audiophile can look at the circle of confusion and say "eh, since I'll never really know I'm hearing what they heard in the mixing studio, my approach will be to satisfy my own musical enjoyment."

But that it is ALSO reasonable to take another approach, the one you seem to be taking Mart68, of trying to raise the chances of hearing something like what the mixers heard, and not having a system that is colored enough to routinely distort or lose content in the signal.

 

[majingotan]

So my questions for ASR are: do you think signal integrity is important? Do you avoid all manipulation to the signal? If you did manipulate your signal, how did you choose your target curve? Do you think there is a role for preference when it comes to signal manipulation?

Signal integrity in the digital domain is important for me. That's why I use a 118 dB SINAD DAC without any digital signal manipulation whatsoever (no DSPs, No EQ, purist digital approach.). However, when it comes to analog domain, I'm all in with signal manipulation (add huge 2nd order harmonic distortion reaching THD+N levels of north of 3-5%, yes 5%), but NEVER using directly using any analog tone controls that mucks the inherent phase curve and frequency response of my planar magnetic linear impedance headphone. No target curve for me, just add pleasing large amounts of 2nd harmonic distortion to the sound is all I need.

 

[IAtaman]

Signal integrity in the digital domain is important for me. That's why I use a 118 dB SINAD DAC without any digital signal manipulation whatsoever (no DSPs, No EQ, purist digital approach.). However, when it comes to analog domain, I'm all in with signal manipulation (add huge 2nd order harmonic distortion reaching THD+N levels of north of 3-5%, yes 5%), but NEVER using directly using any analog tone controls that mucks the inherent phase curve and frequency response of my planar magnetic linear impedance headphone. No target curve for me, just add pleasing large amounts of 2nd harmonic distortion to the sound is all I need.

Very interesting - thanks for sharing. Always good to get insights into alternative approaches.

What you are doing sounds like refusing any drink but distilled water, but adding 2 spoons of sugar into it before drinking to me. Out of curiosity, why not add the harmonics in the digital domain - people who mix your music do?

 

[Chr1]

Very interesting - thanks for sharing. Always good to get insights into alternative approaches.

What you are doing sounds like refusing any drink but distilled water, but adding 2 spoons of sugar into it before drinking to me. Out of curiosity, why not add the harmonics in the digital domain - people who mix your music do?

Many producers still add analogue harmonics too. Often both. So not always either or. Again, a matter of personal preference, so no real right or wrong surely. Again however, I think it is important if possible, to keep any additional harmonics optional in your playback system. To me the ability to switch between "flavoured" and "high fidelity" is key.

 

[Multicore]

Signal integrity in the digital domain is important for me. That's why I use a 118 dB SINAD DAC without any digital signal manipulation whatsoever (no DSPs, No EQ, purist digital approach.). However, when it comes to analog domain, I'm all in with signal manipulation (add huge 2nd order harmonic distortion reaching THD+N levels of north of 3-5%, yes 5%), but NEVER using directly using any analog tone controls that mucks the inherent phase curve and frequency response of my planar magnetic linear impedance headphone. No target curve for me, just add pleasing large amounts of 2nd harmonic distortion to the sound is all I need.

If I wanted to add effects to my playback system I would certainly use digital equipment. The algorithms are great and have been for years, as I know from the production side.

How do you know what distortions or coloration (or other effects) are going to be most pleasing to you? Adjustable effects are surely a huge advantage, no? At the very least you can use them decide what effects you want to build into your analog equipment. For example, use something like this to dial in the sound you want and then go shopping and/or modding amplification and speakers that approximate it.

 

[fpitas]

Again however, I think it is important if possible, to keep any additional harmonics optional in your playback system.

A point I'll make (again), is that typically distortion is added track by track in the creation process, which gives very different results than you'll get by distorting the whole mix en masse.

 

[Chr1]

A point I'll make (again), is that typically distortion is added track by track in the creation process, which gives very different results than you'll get by distorting the whole mix en masse.

Agreed. Exactly why I am saying that it is important to be able to switch between "flavoured" and "Hifi" if possible. Though I don't switch track by track...Just like to have both available.

Going back to the old food analogy : It is basically like having seasoning available as an option.

 

[Matt_Holland]

All the preference scores in spinorama.org are moot

On this point I think you’ll struggle to achieve a desired target curve unless your speakers are interacting with the room acoustics in the right way and the preference rating can give help us choose a good speaker that takes EQ well.

Specifically, the speaker should behave correctly off-axis. Then, I suppose it’s on-axis response is moot because you can EQ to your heart’s desire.

I’d aim for accuracy and low distortion in all system components. Then season to taste.

However, I also understand why subjectivists get pleasure from listening to different kit and “hearing” different attributes. If anything, it’s the whole re-discovering your music collection thing that is enjoyable enough to keep that faction of the industry going.

 

[Anton D]

I assume that double blind tests are also required because expectation bias can occur after room correction, for example, as much it can when getting electronics boxes which measure even “better”.

General question...

How long does expectation bias in listening last?

If someone buys something based on expectation bias, is there an average length of time before the bias wears off and they 'hear the device for what it is?' (That would seem to account for 'audiophilia nervosa' and a high rate of changing gear.)

For some lucky/unlucky people, could it be permanent?

 

[majingotan]

Very interesting - thanks for sharing. Always good to get insights into alternative approaches.

What you are doing sounds like refusing any drink but distilled water, but adding 2 spoons of sugar into it before drinking to me. Out of curiosity, why not add the harmonics in the digital domain - people who mix your music do?

Thanks! I'm gonna be tied to the source with the capability of adding harmonics to the digital domain. What if I want to watch Blu-ray movies then my Blu-ray player must have a harmonic distortion filter in the digital domain

If I wanted to add effects to my playback system I would certainly use digital equipment. The algorithms are great and have been for years, as I know from the production side.

How do you know what distortions or coloration (or other effects) are going to be most pleasing to you? Adjustable effects are surely a huge advantage, no? At the very least you can use them decide what effects you want to build into your analog equipment. For example, use something like this to dial in the sound you want and then go shopping and/or modding amplification and speakers that approximate it.

Glowing tubes and rolling tubes from El34, 6L6, KT66, KT77, KT88, 6550WE, El84 and driver tubes from 6SN7, 6SL7 and 12XA7, 12AU7 is more fun way of tweaking for my subjective taste. It's easier for me to find the pleasing distortion effect that I want for my headphone of choice. For Susvara headphones, I find the combination of 6SL7 and KT88 gives me pleasing sound with it over the less harmonic sounding Schiit Midgard that I use at my work desk. For Focal Utopia original, I swap the tubes to a 6L6GC and 6SN7 (lower gain, less tube noise floor) and get a thicker sounding yet pleasing distortions Utopia. KT88 has way too much gain, and it sounds too bright with Utopia for my taste.

At my workplace, I use the better performing Schiit Midgard since it sounds great both objectively and subjectively to my taste without any adulteration to the amplification of the signal. They're my workhorse for telling me what the source file actually sounds like and I rotate between home enjoyment with pleasing distortions from my SET amp and for critical listening sessions at work.

 

[MattHooper]

I prefer to have as accurate a system as possible to provide a good basis or starting point. I do not want my system adding distortion or unknown frequency response aberrations. From that (hopefully) uncolored start I manipulate to my preference. I can listen as the artist intended, remembering it is not just the artist but also the mixing, mastering, and perhaps cooperate input driving a frequency response that may or may not be what the artist actually intended. Decades ago when I had some (very limited) involvement in the process, mastering was a fuzzy target given the recording had to sound "ok" in systems from high-end home stereos to 8-track car players. It was my exposure to the circle of confusion long before I knew that term. It did teach me that adjusting the sound in my system to sound what I liked was not some sort of sin against the recording gods. In fact, most recordings were created IME/IMO to sound good on more modest systems and sound OK in the car, and were not EQ'd to sound great on a more accurate system.

My curve is close to the Harman curve with some tweaks. I do not listen as loudly as most mixers (or so I have been told) so have implemented the bass boost, but since my room is fairly well treated my high-frequency roll-off is milder than the basic Harman curve. From there it is up to the recordings, which I found to be all over the map, at least to my ears. Some seem way too "piercing" in the highs, others have very "boomy" bass. I'd love a simple bass/mid/treble set of tone controls I could easily change from the listening position. One of my frustrations with modern processors is that they have gone all-in on room correction and target curves, exhibiting tremendous processing capability, but don't offer a simple way for me to do just the basic "tweaks" I used to do all the time with my tone controls.

I appreciate how you phrased your view...as your preference and goal.

I agree that the approach "I do not want my system adding distortion or unknown frequency response aberrations. From that (hopefully) uncolored start I manipulate to my preference." Makes good sense.

I would also understand if someone said: And if I were advising a new audiophile which approach to take, I'd recommend the above, and the reasons why.

It's only when it becomes more of a "my way or the highway" attitude, or presented as The Only Sensible Approach Someone Could Take, that I start to object.

 

[MattHooper]

General question...

How long does expectation bias in listening last?

If someone buys something based on expectation bias, is there an average length of time before the bias wears off and they 'hear the device for what it is?' (That would seem to account for 'audiophilia nervosa' and a high rate of changing gear.)

For some lucky/unlucky people, could it be permanent?

That's an interesting question. I'm not aware of studies on that exact issue.

One of the knocks made here against buying on sighted biases is that it would be unstable. That bias may work for a while, or on some days, but at some point maybe it wears off or goes off and on, and so your system doesn't sound reliably the same to you. Therefore: buy a system that is neutral in actual objective performance, and also doesn't have snake oil claims attached trying to bias you.

But, for one thing, the problem of bias in our listening doesn't go away. Everyone is affected by bias, no matter what equipment you might buy, and everyone's perception will vary at times. It's not the equipment that solves this. Ultimately it's someone's attitude. Someone who bought high performance gear based on objective tests may find some variation in how they think music sounds through their system, but they can say "Well, I bought accurate gear, and that allowed me to make my peace with the result, even on the days I think it's not sounding as good." But likewise the person who bought something more colored, or who bought something on impressions infected by expectation bias, can say something similar. They too can make their peace with the idea that, well, some days the system will not seem to be as compelling as other days. That's being human. And it's a spectrum, because some who buy objectively-based will rarely or never think their system sounds bad, and the same goes for some people who bought on expectation bias.

As to how long it can last, as I've said, I've been absolutely thrilled by the characteristics I at least seem to hear from my tube amps. This has lasted for almost 25 years! I don't see any reason to think it could not have been an expectation bias all this time. But as I've said, even if so, I'm happy to take advantage of it, given this consistency.

Glowing tubes and rolling tubes from El34, 6L6, KT66, KT77, KT88, 6550WE, El84 and driver tubes from 6SN7, 6SL7 and 12XA7, 12AU7 is more fun way of tweaking for my subjective taste. I

I agree. The idea of doing it via digital modelling takes the fun and romance out of it, for me. I really got a kick out of getting in to tube rolling a few years ago. Am I just imagining things? Could be. But it's fun, and I just find different tubes kinda neat, and tactile, and I love the visuals.

I get why others would go the digital modelling route.

 

[Multicore]

Thanks! I'm gonna be tied to the source with the capability of adding harmonics to the digital domain. What if I want to watch Blu-ray movies then my Blu-ray player must have a harmonic distortion filter in the digital domain



Glowing tubes and rolling tubes from El34, 6L6, KT66, KT77, KT88, 6550WE, El84 and driver tubes from 6SN7, 6SL7 and 12XA7, 12AU7 is more fun way of tweaking for my subjective taste. It's easier for me to find the pleasing distortion effect that I want for my headphone of choice. For Susvara headphones, I find the combination of 6SL7 and KT88 gives me pleasing sound with it over the less harmonic sounding Schiit Midgard that I use at my work desk. For Focal Utopia original, I swap the tubes to a 6L6GC and 6SN7 (lower gain, less tube noise floor) and get a thicker sounding yet pleasing distortions Utopia. KT88 has way too much gain, and it sounds too bright with Utopia for my taste.

At my workplace, I use the better performing Schiit Midgard since it sounds great both objectively and subjectively to my taste without any adulteration to the amplification of the signal. They're my workhorse for telling me what the source file actually sounds like and I rotate between home enjoyment with pleasing distortions from my SET amp and for critical listening sessions at work.View attachment 333734

It seems to me a strange thing

Mystifying

That this holy, immaculate signal is so carefully taken care of and then sent to that musty obsolete tech and then into high fidelity headphones for evaluation of how well the tubes mussed it up

And that this practice is

A fashionable hobby?

[Don't mind me, I'm just old and completely out of touch with what cool people do these days.]