Importance of Tonality (?)

[Easternlethal]

@MattHooper Thanks for articulating the question. This isn't an area I've had many discussions with other people about so I have a hard time explaining my thoughts. Maybe I provided too much background and irrelevant information.

@Inner Space Thanks! You anticipated my next post.

Let's assume that we have a system that is totally transparent okay. And that was the goal when we selected our system - it measures great and we have no distortion etc. So no arguments or questions there.

Then we accept that due to our personal preferences, hearing sensitivity changes etc. we want to introduce coloration to the system. So this here is a secondary goal (I hope that's allowed ). Then the next question is a) 'how do I know what coloration I want' and b) 'how do I achieve that'?

So a) Is it possible to use a precision analyser reading of a piece of equipment I like, and then figuring out from that that is the coloration I want (e.g more harmonics / frequency roll-off etc.)?

b) Is it possible to replicate those colorations using equalisers, software, DSPs etc.? I know there are many post-processing tools out there but they all seem to be imperfect (at least the ones we use for music production seem to be) and I do not know what is really possible with them and where to start.

I know the answer depends on what coloration is exactly and is subjective to a degree, but let's assume it's something non-trivial, like say a valve amp or one of those products that measured poorly. And if the answer is "No there isn't any post-processing tool that gives you so much control over tone that you can replicate such equipment, much less the sound of the instrument in the space you remembered / imagined", then maybe there really shouldn't be another goal after all.

Sorry maybe this question is really outside the scope of this forum.. but I hope not.

 

[oursmagenta]

To me there is no mystery here.
a) the only piece of information that you have access to is the final master (arguing about the fact that we have access, in classical music for instance to the score is pointless in the sense that the gap between earing a symphony from the score and earing it from the Berlin philharmonic under karajan is just unbridgeable, makes the argument purely theoretical)
b) the reproduction system should reproduce anything, so you don't want to optimize for a particular genre, a particular composer, a particular track, or particular moment in a particular track. Especially when the target for the optimization is not known (the intent of the parties making the recording).

The only possible move from here is to get a system that when fed with a given content (say from pink noise, to actual recordings) will reproduce what an average mastering room will produce if fed with said content.

Then you hope for the best that
a) you nailed what is an average mastering room
b) that the whole process involving the composer, performer, recording engineer, mastering engineer etc... Had a say on the final master eared in the mastering room, and that the intent for this recording is coming from a weighted (that can sometimes unfortunately be zero for one or some of the parties) consensus among all this parties listening to the same mastered content in the same mastering room.

The mastering room is the single point of convergence where things actually get real and tractable. Indeed, getting an audio system that reproduce what would reproduce an average mastering room is a goal that you can tend to reach because the target is documented, and we have the measurements tools to know when we are close enough to the target. All the rest is past the horizon point like for a black hole, there is no measurements associated with the intent of any of the parties forming a recording, and even less for the intent coming from a weighted consensus.

 

[Easternlethal]

To me there is no mystery here.
a) the only piece of information that you have access to is the final master (arguing about the fact that we have access, in classical music for instance to the score is pointless in the sense that the gap between earing a symphony from the score and earing it from the Berlin philharmonic under karajan is just unbridgeable, makes the argument purely theoretical)

Yes I understand this argument and maybe I'm belabouring the point now, so I'll stop posting after this for a little while. But let's take karajan and berlin phil for example. Its generally thought that a lot of the deutsche grammophon recordings of his are really sub-standard compared to say, something that decca would produce. In some cases it really does sound like they just used one stereo mic above the orchestra and post-processed everything else. And many a time I am sitting there listening and just moaning at the poor recording instead of focusing on the music and wondering whether there's a piece of equipment out there or some process that can make it sound 'better'.

Doesn't this happen to anybody here? Or am I succumbing to the dreaded bias and wasting my time?

 

[oursmagenta]

Yes I understand this argument and maybe I'm belabouring the point now, so I'll stop posting after this for a little while. But let's take karajan and berlin phil for example. Its generally thought that a lot of the deutsche grammophon recordings of his are really sub-standard compared to say, something that decca would produce. In some cases it really does sound like they just used one stereo mic above the orchestra and post-processed everything else. And many a time I am sitting there listening and just moaning at the poor recording instead of focusing on the music and wondering whether there's a piece of equipment out there or some process that can make it sound 'better'.

Doesn't this happen to anybody here? Or am I succumbing to the dreaded bias and wasting my time?

That is off the point (at least to me), if the recording is bad, it is bad that's it, there is no point in trying to enhance it (at least on our end) for two reasons:
a) we don't have access to the raw premastered material
b) if we had, we would improvising ourselves mastering engineers, which is not a small feat, and do we really want to do that?

To me that would defeat what is listening to music, which is listening to something that someone passed you over, without changing it. You don't ask a performer to change its interpretation because you don't like it. Now apply that to the whole audio chain down to the mastering room (composer, performer, recording engineer, mastering engineer). Off course they can betray each other (performer not being faithful to the composer, mastering engineer not being faithful to... etc..) but that's not our business.

 

[Roland]

Yes I understand this argument and maybe I'm belabouring the point now, so I'll stop posting after this for a little while. But let's take karajan and berlin phil for example. Its generally thought that a lot of the deutsche grammophon recordings of his are really sub-standard compared to say, something that decca would produce. In some cases it really does sound like they just used one stereo mic above the orchestra and post-processed everything else. And many a time I am sitting there listening and just moaning at the poor recording instead of focusing on the music and wondering whether there's a piece of equipment out there or some process that can make it sound 'better'.

Doesn't this happen to anybody here? Or am I succumbing to the dreaded bias and wasting my time?

Yes. You’re not wasting your time, but you are on the wrong forum because you are in an echo chamber for a dogma eloquently described by Matthooper above. Therefore your question is heretical.

 

[oursmagenta]

Yes. You’re not wasting your time, but you are on the wrong forum because you are in an echo chamber for a dogma eloquently described by Matthooper above. Therefore your question is heretical.

Why twisting the thing into a dogma/heretics opposition? Do we have the right to make a choice on what rational perspective we want to adopt, without going into this dogma/heretics rabbit hole?

The moment we ear each other thoughts and reasonings, the debate can still go on in a rational fashion.

The question raised by @Easternlethal is valid, the debate that raises from it also. We may not have the same rational perspective on it, that's all...

 

[Frank Dernie]

I am long time lurker of this forum and a big big fan of objective measurements. And as a part-time music producer, composer, performer and hi-fi listener I would like to discuss something which is has been in the fringes of many conversation - namely, tonality.

But before that I just want to say that this is not about the validity or benefits of using measurements. Rather, this is to kick off discussions about whether there should be any interrelationship between measurements and tonality, and if so what.

To start with, I would like to share my own experiences from someone who spends more time at the production end than at the listening end - where there are as many disagreements between composers, conductors and performers as there are between hi-fi listeners and manufacturers! If you look at music as a series of processes, where a signal is being created between to links in the chain, ie. dac to preamp. At the risk of offending music performer, I and many composers think of music as a signal between the composer and the audience, with the performers being the processors. So we start with something which we want to communicate. And at this stage, even the process of writing music is already a lossy process because I am confined by the notes on the page and the instruments which are available. Then comes the performer, who has the job of reading the notes he has been presented. 'Ah!' he thinks. 'I can make this music sound good'. So he picks up his instrument and displays his performance techniques, making something a little louder here, a little softer there etc. Maybe he even adds a few overtones because he has a very expensive instrument and good enough technique to do that. Then the production company gets involved and thinks it would sound even better if the piece was recorded in a cathedral. So then comes the recording engineer who has to record the whole thing and capture the environment. So he mikes everything near and far. But it doesn't sound as good so he makes adjustments - adding a carpet here or there, angling the mic or maybe even using EQ (yes, some recording engineers even do that!). At the end of the day everybody packs up and goes home having experienced the performance from completely different perspectives. Up until this point, fidelity is very low in the list of priorities because the performer is more concerned with playing, the recordist is trying to capture or reproduce what to his ears, is unique about the environment and the composer is just happy that people are going to hear what he wrote. In each case, either they are focusing on tonality or the content of the music itself.

Then in the evening, the files get sent to the mastering engineer who fires up his workstation. Note that he is called the mastering engineer for a reason, which is to get paid by the production company for making everything sound good. So he listens to each track again. Playing the raw tracks on his yamaha, JBL and genelac monitors he thinks the instrument is a bit too bright. So he runs the signal through his vintage teletronix compressor to give him that warmth he is so accustomed to hearing from his other recordings. Then he mixes it with the ambient noise track recorded by the recordsist's omnidirectional condensor (which picks up a very faint rumble from a nearby passing train).

Now here we are listening to the recording and discussing our equipment...

You can't do anything about the recording, unless you make it yourself.
All your domestic hifi can be expected (hoped) to do is reproduce the recording you have bought without adding or subtracting anything.
I know I bang on about it but the fact is there is far more variability in recording quality than in the various pieces of hifi equipment people may listen to it with.
Add to that most people are not listening on what we on a forum like this would consider hifi anyway and this will always be a moot point.

FWIW the recordings I have felt sounded the most natural to me are those, usually old ones, made with few microphones and the balance achieved by microphone and musician positioning and the recording not being f*cked about with afterwards.
There are not many, and almost none nowadays, not too bad if you like folk, classic rock and classical music, there are plenty of vintage recordings that sound natural (plenty don't of course) but a lot of new stuff is, to my ears, really awful.

Often something recorded on a home studio and put on Bandcamp sounds way more natural than "professionally made" recordings to me.

 

[Jimbob54]

So anyway one of the follow up questions to this long train of thought, is, when the audio precision analyzer shows that something is not transparent, can we use the results to tell whether it is intentional because the manufacturer wants to introduce tonality (and how) or is it always the case that the product is just poorly designed?

No, you can't. The analyser is as dumb as the equipment it measures. Take harmonic distortion, especially in a sold state amp. How do I know that profile was chosen / modelled as opposed to just the result of the bargain bin components used or the lack of proper design of the circuit? Surely as a consumer , all I care about is that it has that distortion profile and I can either choose to buy/ audition that , or not, because I prefer a different sonic profile. Now, whether that then makes me subjectively like that profile compared to transparent is entirely a personal preference.

 

[Blumlein 88]

To me there is no mystery here.
a) the only piece of information that you have access to is the final master (arguing about the fact that we have access, in classical music for instance to the score is pointless in the sense that the gap between earing a symphony from the score and earing it from the Berlin philharmonic under karajan is just unbridgeable, makes the argument purely theoretical)
b) the reproduction system should reproduce anything, so you don't want to optimize for a particular genre, a particular composer, a particular track, or particular moment in a particular track. Especially when the target for the optimization is not known (the intent of the parties making the recording).

The only possible move from here is to get a system that when fed with a given content (say from pink noise, to actual recordings) will reproduce what an average mastering room will produce if fed with said content.

Then you hope for the best that
a) you nailed what is an average mastering room
b) that the whole process involving the composer, performer, recording engineer, mastering engineer etc... Had a say on the final master eared in the mastering room, and that the intent for this recording is coming from a weighted (that can sometimes unfortunately be zero for one or some of the parties) consensus among all this parties listening to the same mastered content in the same mastering room.

The mastering room is the single point of convergence where things actually get real and tractable. Indeed, getting an audio system that reproduce what would reproduce an average mastering room is a goal that you can tend to reach because the target is documented, and we have the measurements tools to know when we are close enough to the target. All the rest is past the horizon point like for a black hole, there is no measurements associated with the intent of any of the parties forming a recording, and even less for the intent coming from a weighted consensus.

There is no average mastering room. So everything following the false premise is off the mark.

 

[Blumlein 88]

That is off the point (at least to me), if the recording is bad, it is bad that's it, there is no point in trying to enhance it (at least on our end) for two reasons:
a) we don't have access to the raw premastered material
b) if we had, we would improvising ourselves mastering engineers, which is not a small feat, and do we really want to do that?

To me that would defeat what is listening to music, which is listening to something that someone passed you over, without changing it. You don't ask a performer to change its interpretation because you don't like it. Now apply that to the whole audio chain down to the mastering room (composer, performer, recording engineer, mastering engineer). Off course they can betray each other (performer not being faithful to the composer, mastering engineer not being faithful to... etc..) but that's not our business.

Not being faithful is what some would consider artistic creation.

 

[Purité Audio]

John Atkinson’s interview on measurement at the Stereophile site touches on this very point, were the designers errors intentional or just incompetent.
Keith

 

[Frank Dernie]

I know the answer depends on what coloration is exactly and is subjective to a degree, but let's assume it's something non-trivial, like say a valve amp or one of those products that measured poorly. And if the answer is "No there isn't any post-processing tool that gives you so much control over tone that you can replicate such equipment, much less the sound of the instrument in the space you remembered / imagined", then maybe there really shouldn't be another goal after all.

For me the difficulty here is "do you want to add the same colouration to all recordings"?
I wouldn't personally, not least because some of my recordings sound fabulous without dicking with them (IMO).
I do agree with the awful sound of so many old DG classical recordings though but I can't see what one can do about it.
I have personally felt that the performance is more important than the recording though.
Old Artur Schnabel Schubert recordings are of variable but historic recording quality with no mistakes taken out of the one-shot recording process of the time but they are still more enjoyable and listenable than some other recordings I have heard, so whilst it is certainly better to have a good recording my musical enjoyment is not ruined by a poor one, but this is a personal view.

 

[FrantzM]

Hi

There exist a class of machine whose purpose is to be complicated, They're called Rupe Goldberg machines. I believe that sometimes we do the same with our thinking.
Most artistic oeuvre are collaborative efforts. The artist/s have to use something to reach an audience, the least that "something", imposes of its own, the better it is.
In our case, music, there is an intention, a collaboration that of the artist/s and the production team , let's call it "P". We would like to hear "P". We want it with the least deterioration ... At this point in time, 2021, "P" must to use electronics means to reach a transducer. We know the transducers to be the weak points. The electronics are the least problematic.
So we have to make sure that what goes in electronics, come out the same as it came in. That is where we can do the least harm to the production, to "P".

And these days, what is in bold, can be achieved at very low cost.

The Rupe Goldberg" way of separating people from their money continues however. As an example

This at $9.00

is about as good in objective terms as ...

that ... which has to cost more than $15,000 ( way more than 15,000 US dollars, way more,) here someone is pointing at that device !! )

Peace

 

[dualazmak]

...
We just cannot bypass the final mastering from the engineer. That's about it.
...

Essentially, I agree with you...
And we just cannot bypass the final fine tuning of our DAC, amplifier, speaker etc. by the audio designer and/or engineer...

I recently had similar kind of discussion in terms of so called "musicality" in amplifier design and tuning at here;

 

[Koeitje]

Sometimes it is due to intentional tweaking, or "voicing".

Yes, but then I don't consider it hifi equipement anymore. So are we talking about audio or hifi?

 

[Vince2]

I just wanted to bring attention to the methodology difficulties of establishing accuracy if you don't have a reference point that is grounded in the performance. Even then, the acoustics of the space and seating position of the listener can cause significant variation in the listening experience. Front row vs 15th row?

 

[Jimbob54]

I just wanted to bring attention to the methodology difficulties of establishing accuracy if you don't have a reference point that is grounded in the performance. Even then, the acoustics of the space and seating position of the listener can cause significant variation in the listening experience. Front row vs 15th row?

Forget the performance . There is only what is in the source file/ disc. And all bets are off when it hits the transducers and your room/ ears. All you can do then is get the best you can afford and maximise the in room performance.

 

[Chromatischism]

I just wanted to bring attention to the methodology difficulties of establishing accuracy if you don't have a reference point that is grounded in the performance. Even then, the acoustics of the space and seating position of the listener can cause significant variation in the listening experience. Front row vs 15th row?

To properly experience 15th row at home we would need a recording from the 15th row in a multi-channel format to play back at home on an Atmos or Auro setup. If you live in a cathedral you could try to play a stereo recording and sit far back, but otherwise forget about it.

 

[Katji]

[...] when the audio precision analyzer shows that something is not transparent, can we use the results to tell whether it is intentional because the manufacturer wants to introduce tonality (and how) or is it always the case that the product is just poorly designed?

No. It is what it is. ...Not "subtext" or mental speculation.

 

[Katji]

I just wanted to bring attention to the methodology difficulties of establishing accuracy if you don't have a reference point that is grounded in the performance. Even then, the acoustics of the space and seating position of the listener can cause significant variation in the listening experience. Front row vs 15th row?

No, that is misconception. Hi-fi = high-fidelity, hi-fi system. The purpose of it is to reproduce as accurately as possible what is on the physical media, the product.

the acoustics of the space and seating position of the listener can cause significant variation in the listening experience. Front row vs 15th row?

Of course it does. But it has nothing to do with the purpose of hi-fi, or audio equipment for reproducing.
You concern yourself with front row or 15th row in your home or in the hall or wherever the music is played. (As in, club, bar lounge, school hall disco whatever.)