accuracy is subjective

[johncarm]

I argue that accuracy is subjective. First of all we have to be in a context where accuracy is meaningful.

An example is a performance of acoustic music in a hall, then listening in the control room to see how well the sound is captured.

As each listener, say the conductor or the sound engineer, compares the recording to the live sound, they are listening for different things. Some might care about the timbres, others might care about the hall reverberation, some might care about the bass line, etc. So they can arrive at different conclusions about what is most accurate.

 

[baskingshark]

For sure this hobby is very subjective. As you say, different people prioritize different areas. Some want timbre, some want tonality, some want technical excellence.

Even for the same music track, two people may hear very differently . We all have different hearing health (due to age, occupation or leisure exposure), we may have different ear anatomies (which my affect pinna gain), we have different sources/eartips, we have different listening volumes (Fletcher Munson curve), all of which may affect the sound.

 

[old tech]

It depends on the context. Accuracy of a recorded signal, say from the mic through right through to the storage/playback format and device and then to the output cables from the amp is very objective. In this hobbyist context, the only subjectivity is perceptual errors and placebo effects.

 

[magicscreen]

It depends on the context. Accuracy of a recorded signal, say from the mic through right through to the storage/playback format and device and then to the output cables from the amp is very objective. In this hobbyist context, the only subjectivity is perceptual errors and placebo effects.

What is objective about output cables? They are sounding different. Only you cannot measure the difference. And you cannot hear it because you have a bad hearing! You have to accept that fact and do not spread misinformation!

 

[castleofargh]

I argue that accuracy is subjective. First of all we have to be in a context where accuracy is meaningful.

An example is a performance of acoustic music in a hall, then listening in the control room to see how well the sound is captured.

As each listener, say the conductor or the sound engineer, compares the recording to the live sound, they are listening for different things. Some might care about the timbres, others might care about the hall reverberation, some might care about the bass line, etc. So they can arrive at different conclusions about what is most accurate.

You make a case about accuracy by discussing decisions, personal preferences and listeners. Of course you'd end up with subjectivity when the only variables you look at are themselves subjective. If your very notion of being correct depends on who's talking, then we have opinions. One can be of the opinion that something is accurate, but it's only his opinion. It doesn't define how accurate something really is.

If instead you consider accuracy as a technical term, then it's all about objective data and some specific predetermined target. If both are well defined objectively, then you can check for accuracy. Opinions are irrelevant.
Of course if you lack data on the target or on the thing you wish to check for accuracy, then there isn't much we can do. It's like talking of fidelity when we don't have a clue what the original was like. It leads nowhere.
Instead of arguing that accuracy is subjective, I would argue that accuracy probably shouldn't be brought up in a subjective discussion. The feeling of accuracy could be related to accuracy, but ultimately it's only a feeling, not necessarily factual accuracy. It would often be a mistake to assume they're the same thing.

 

[bfreedma]

What is objective about output cables? They are sounding different. Only you cannot measure the difference. And you cannot hear it because you have a bad hearing! You have to accept that fact and do not spread misinformation!

Proof of any of this being audible would be great. Since this is Sound Science, please post it.

If not, I’ll file your claim next to Bigfoot sightings and alien abduction stories. Where evidence free rants belong...

 

[bigshot]

Do a null test and then tell me that accuracy is subjective.

Not only is accuracy objective, it's relatively simple to measure and test for.

 

[johncarm]

There are many reasons to record music. But let's consider the example of hearing an acoustic performance in a concert hall that gives us pleasant feelings. It is the pleasure and the subjective effect of those sounds that motivate us to record them.

Then we step into the control room and listen. Do we experience those same feelings and effects? If so, the recording is accurate. If not, it is inaccurate.

Y'all seem to have a low opinion of opinion. But in these case, the case of judging the accuracy of the recording, opinions are all we have. The only way to judge the success of the recording is that it successfully reproduces the aesthetics of the experience that motivated us to record it in the first place.

I'll grant that A/D units and cables are perfect. We still have the effects of the microphones, microphone placement, and speakers on the accuracy.

 

[bigshot]

You're a little bit off there... The recording studio is a creative place, and creativity is a subjective process. Performers are performing and deciding how to arrange and play the music. Judgements are being made about miking distances, balances, tone and ambience. These are all creative and subjective. When they are happy with the recording, they put their subjective creativity on a CD and sell it to you.

Playback of the sound electronically is not creative. You play the CD. You listen to it on headphones. No decisions are being made, no judgements, except for setting the volume level. Ideally, the process from CD to your ears should be a straight line- totally accurate to the original monitoring chain used in the studio. No modifications, no deviation from the sound the people who made the recording intended. That is fidelity, and fidelity is objective.

Now some people don't want an accurate playback. They buy tube amps with euphonic distortion. They skew EQ into V shaped curves. They apply reverb or compression. All that is fine. But that isn't accurate and it isn't objective. It's a totally subjective way to listen to music. If they want to get even more subjective, they can get up and dance! I'm sure it's a lot of fun. But none of that can be defined as "high fidelity". Fidelity = Accuracy.

Now "experience" is something entirely different. Can I listen to Sgt Pepper and *experience* it the way the Beatles did? Nope. (well not without LSD!) Maybe I don't even like Sgt Pepper. It's just a bunch of noise to me. That is a subjective experience, but it isn't what was intended. I can get fancier playback equipment and spend a lot of money, but it won't help. Equipment and subjective impressions are two completely separate things.

Personally, I think a lot of audiophiles conflate accuracy and subjective impressions in an attempt to make themselves seem smarter than they actually are. They use words to describe in deliberately vague ways and use semantics to obfuscate in an attempt to avoid cold hard objective facts. I think I know where they learned that way of blathering... from reading too much high end audio snake oil sales pitch!

 

[ThEvil0nE]

You can't talk about accuracies/inaccuracies without the means of measurements. Sound preference on the other hand is subjective.

 

[Davesrose]

So in short, it depends on your term of "accuracy". I'm sure many audiophiles just think of "accuracy" being a term about detail or "transparency" of a recording that's revealing. Especially when considering perception of audio from the transducer, that is more subjective. But throughout audio, there are many areas that are clearly objective. Let's just look at the recording process: there are quantitative measurements of say noise floor, volume level, balances, etc.

 

[johncarm]

Bigshot: the process you describe is only one way to carry out recording. If this is a process you think is best, it's your opinion that it's best. But other musicians and recording engineers (I've talked to some at Sheffield Lab and read about some BBC experiments) start from the point of view that music is a phenomenon created by musicians who aim for certain expressive effects. You would like to claim that these effects are meaningless to accuracy because they are subjective or indefinable. But these musicians are conscious of the patterns that go into creating these effects and as part of their work pay attention to the conditions in which musical expression is audible and when it's not. So there is an obvious meaning to accuracy which is that the music effects can be perceived in the recording.

Even in the process you describe, the musicians have in mind some musical expression or effect that they are aiming to create. If they can hear that effect when they play live, and they find that effect more valuable than what you would get from a so-so recording, then their attempts to improve the recording amount to recreating that effect accurately.

There are some musicians who just don't care that much about the effect they create and don't care whether it is replicated accurately, or perhaps they listen with very uncritical ears to the recording. I've met musicians like that. Maybe they are the majority. However, there is this other distinct perspective, which shows that your idea of the recording process is an opinion that not all share. So I could state that "how recording should go" just like "accuracy" are opinions, and that opinions are all we have. We have no objective criteria. Saying that amps and A/Ds and D/As are perfect doesn't help because you still have to deal with microphones and speakers and headphones.

 

[bigshot]

First of all, I've supervised recording sessions and mixes. Secondly, musicians aiming for expressive effects is exactly what I described. Recording and mixing are creative, expressive, subjective. Playback in the home is objective and all about fidelity. Subjectivity is the way creative artists work. Fidelity is how home audio works. I've never worked with an artist who didn't care. If I did, I probably wouldn't work with them again. Microphones, speakers and headphones aren't perfect. But they can still have a high degree of fidelity, which is quantifiable and measurable. Or they can not be accurate.

 

[Redcarmoose]

I argue that accuracy is subjective. First of all we have to be in a context where accuracy is meaningful.

An example is a performance of acoustic music in a hall, then listening in the control room to see how well the sound is captured.

As each listener, say the conductor or the sound engineer, compares the recording to the live sound, they are listening for different things. Some might care about the timbres, others might care about the hall reverberation, some might care about the bass line, etc. So they can arrive at different conclusions about what is most accurate.

The actual items of the replay of a actual musical instrument being truth is a needle in a haystack.

The actual event which has been recorded is now lost in time and not available for reference. All we have is a artistic take on the event. At times multi-tracks are made into a whole new artistic statement. If you read the books and read the papers it starts to become very very clear, the most educated and inspired people in the recording and music playback fields also agree on this. People that are delusional about there being an actual factual form of playback fidelity are just that (self wrongly taught) they simply haven’t read enough books by the leading experienced writers in the field.

Though keep in mind the idea of tone balance is getting better all the time. Prior to the papers written in 2010-2012; it was a guess as to what neutral was or is. That is neutral perception of music in IEMs or full-size.

Since there is no standard in the recording industry, there is absolutely no form or path for studio recordings to ever be standardized. It’s probably safe to say (while creative) that new recordings are beautiful and arrive at value due to creativity in creating an emotional response. The very smartest people in both the recording and playback community have expressed it’s only emotional responses that we can ever hope to achieve in the end, nothing else.

But beyond that the variables........personal perception of sound vary from person to person. In the speaker playback, room response, personal ear pinna gain amplification characteristics, torso reflections adding changes are just the start of a multitude of variables. So even if the studios were standardized and something like the 2017 Harman Response Curve was replayed by headphones there would still be variations.

The ear coupler (standard headphone) has literally 100s and 100s of variations to reproduce even a critically recorded and played back sound file.

Head shape, pinna gain reflections, ear canal resonance modes, pad air-tight fit, distance from ear drum to driver, ear size...etc, etc.

So due to the multitude of critical variables in the chain there is absolutely no other way to tackle this issue other than try. So really it’s way way way beyond production subjectivity in the end!

Cheers!

http://seanolive.blogspot.com/2013/04/the-relationship-between-perception-and.html?m=1

 

[johncarm]

First of all, I've supervised recording sessions and mixes. Secondly, musicians aiming for expressive effects is exactly what I described. Recording and mixing are creative, expressive, subjective. Playback in the home is objective and all about fidelity. Subjectivity is the way creative artists work. Fidelity is how home audio works. I've never worked with an artist who didn't care. If I did, I probably wouldn't work with them again. Microphones, speakers and headphones aren't perfect. But they can still have a high degree of fidelity, which is quantifiable and measurable. Or they can not be accurate.

@bigshot Musicians spend many years practicing their instruments creating live acoustic effects. They listen to each other. They have feedback from a teacher. They conduct.

They listen to themselves live when they practice. They don't wear earplugs while practicing. They don't judge their performance through a speaker.

I'm also talking about a perspective in which these live acoustic effects are much more desirable than any playback through a speaker, where there is always something key lost. So in this perspective accuracy means to get as close to the live effect as possible without losing things that are essential to the desired effect. Apparently that's not the way you work, and that's how you like it. But that's not the only way to work.

To evaluate accuracy we must consider the whole chain including microphones and speakers. Although we tend to divide the audio business into the recording and say that is one thing, and playback and say that's another, the only meaning of accuracy must include how they function together. Your perspective is perhaps practical, but also artificial and I would argue counterproductive. Depending on how it affects your recordings, I might find it compromises the quality compared to something better to me like the Sheffield Labs orchestral recordings.