"Things that cannot be measured"

[SIY]

Possibly/Probably not but that's not practical today. I just downloaded a recording of the 3 major Stravinsky ballet scores although I already have multiple recordings of them in my library with other conductors and orchestras. I am, as always, looking forward to hearing something new and interesting in the interpretation and performance.

There is no possibility that, in my lifetime, there will be a synthetic release of this music that will also incorporate the artistic direction of a particular conductor and/or a particular orchestra, regardless of how convincingly realistic it may seem.

I enjoy pissing off my classical musician friends by referring to their orchestra as a cover band.

 

[MattHooper]

Getting it solved like 100% perfect, that's indeed not going to happen. But you can get close, or at least avoid getting derailed even further (if it interests you).

Sure. As I've said, the fact one can never achieve perfection doesn't mean one can't move in a direction from worse to better, given a goal.
Perfection is the enemy of the good, as the saying goes.

Though I'm not sure really how close one can get to solving the circle of confusion, especially if one has a wide ranging interests that comprise recordings having been made in so many different room/equipment scenarios. Looking at my own collection it would clearly be hopeless to think I'm recreating a significant fraction "as they heard it in their studio on their mixing gear." I could have no idea what they heard, and there's no reason to think that the teeny amount of distortion I might introduce with my tube amps pushes me further from that sound in any significant degree.

I have musician friends who bring music they have recorded and produced to my place for a listen. My system is more revealing than the equipment they use for recording. They hear even more in to their recordings on my system (and like it). What would the solution be there?
If I'm trying to reproduce "the recorded signal" then I may well be doing so "better" than they were hearing when they laid the signal down.
But if the ultimate goal is to "hear what the artist heard" - which is the circle of confusion issue - then would I want to reduce the capabilities of my system to those of their monitors instead?

And allowing people to tune to their liking, already confirmed that on multiple occasions. Even made it my signature.

Yes it's agreed all around. But I think some issues are starting to get a bit mixed up. There is "what is the goal for our system and why?" (Which includes people's feelings on the circle of confusion).

Then there was the subject "But IF we allow seasoning to taste as legitimate...how should we go about that?" Which is where many here have the stance "You shouldn't introduce your seasoning as a set and forget version that adds it to everything. You should have a neutral system and then ADD seasoning selectively, e.g. by EQ, if you wish." And I was trying to bring a perspective to that as well (where it can make sense to "add seasoning to the whole sound system" for some people).

 

[Geert]

Is tube amp distortion audible? (Or is any specific claim of audible tube amp distortion TRUE?)

The best tube amps sound like... solid state amps

Otherwise what type of "constant distorting" gear are people thinking of, that they would reject?

My primary concern isn't gear, it's the process where people often go wrong. Bringing home a new sound of speakers that sounded brilliant in the showroom and being surprised they don't like what they hear, not understanding the speakers are to close to the wall or they're dealing with room modes. So after consulting a forum they go out to buy silver interlinks. Back home the interlinks sound fine when playing vinyl, but now their DAC "sounds digital" so they need a new R2R DAC...

So even if we presume the goal of "reproducing the recorded signal without distortion" what are we supposed to get from that? What do you think we are *losing* if we have a certain level of distortion in a system?

That the distortion of choice often only works in a very specific context, while people on forums generalize it. E.g. vinyl sounds better than digital.

 

[Geert]

Though I'm not sure really how close one can get to solving the circle of confusion, especially if one has a wide ranging interests that comprise recordings having been made in so many different room/equipment scenarios.

How close you can get I can't quantify, but I believe you'll obtain the best result with a well designed neutral system. A system that deviates a lot might do better with a specific music genre, but will probably fail with others.

I could have no idea what they heard,

Why not? These days most of them use near field monitors designed for a flat response. Today people have access to tools and products to replicate that at home (move in a direction from worse to better like you said).

I have musician friends who bring music they have recorded and produced to my place for a listen. My system is more revealing than the equipment they use for recording. They hear even more in to their recordings on my system (and like it). What would the solution be there?

If they like what they hear, there's no problem to be solved. I can't judge your particular situation, but I've never heard a system being more revealing than a top studio system.

Then there was the subject "But IF we allow seasoning to taste as legitimate...how should we go about that?" Which is where many here have the stance "You shouldn't introduce your seasoning as a set and forget version that adds it to everything. You should have a neutral system and then ADD seasoning selectively, e.g. by EQ.

Everyone's free to choose their approach, but I believe EQ-ing is the most efficient and flexible one. For some people switching gear is part of the hobby of course.

 

[MattHooper]

My primary concern isn't gear, it's the process where people often go wrong. Bringing home a new sound of speakers that sounded brilliant in the showroom and being surprised they don't like what they hear, not understanding the speakers are to close to the wall or they're dealing with room modes. So after consulting a forum they go out to buy silver interlinks. Back home the interlinks sound fine when playing vinyl, but now their DAC "sounds digital" so they need a new R2R DAC...

Ok, agreed but that's an entirely different subject than the one I've been addressing.

(And again I'm not trying to address the audible difference of any particular tube amp: I"m just using the concept of tube amplifier distortion as some example to discuss the concept of adding non-neutrality to a system, and why someone would or wouldn't do that).

That the distortion of choice often only works in a very specific context, while people on forums generalize it. E.g. vinyl sounds better than digital.

Except all the examples where people have liked the distortion "across the board" as in my own example for my tube amps (and in the experience of many other people).

And in terms of the goal of neutrality, once you are saying "it's ok to season SOME recordings to taste" it seems kind of arbitrary to point to someone else and say: "but you are doing it too much!"

This is what I'm getting at. I understand the view "go for neutrality and then you can fiddle with changing the sound to taste with EQ in certain circumstances and then go back to neutrality at will." It's just the assumption that this is "the best way to go about it"...which may be true for some people and their own goal, but not for others. Some may indeed find a "set and forget" type of coloration, even if slight, is preferable.
They actually do like the coloration on pretty much everything, and it relieves them of even bothering fiddling with EQ per track. They just sit back and enjoy.

 

[MattHooper]

Why not? These days most of them use near field monitors designed for a flat response. Today people have access to tools and products to replicate that at home (move in a direction from worse to better like you said).

This is why I said earlier: even if every studio switched to the same, neutral monitoring gear today, we would still be stuck with the vast library of recorded music that was monitored in countless different rooms and equipment. Think of how many records were monitored via the NS10!
Let alone all the different speaker designs and monitoring rooms.

If they like what they hear, there's no problem to be solved. I can't judge your particular situation, but I've never heard a system being more revealing than a top studio system.

These musicians weren't using top studio systems. Tons of music today is being produced in essentially home studios like the ones my brother and friend use. They cobble together whatever speakers they can afford. So I can't simultaneously accurately reproduce what they heard AND what an engineer using a high grade neutral system heard at the same time, and that goes for plenty of examples in between.
And especially given the amount of different music many of us listen to, we generally can't hope to know whether any particular instance of playback in our room is "accurate to what the artists monitored." Which means IMO that flavouring the sound doesn't necessarily entail a full disavowal even of "accuracy" but can simply be part of a "practical, realistic" understanding of the compromises we all make in our systems.

Everyone's free to choose their approach, but I believe EQ-ing is the most efficient and flexible one. For some people switching gear is part of the hobby of course.

I agree with that general approach. EQ is the most flexible way to depart from neutrality when you want that.

 

[Frgirard]

The flat response does not exist. The day it happens, the speakers industry will be sick

 

[Geert]

Except all the examples where people have liked the distortion "across the board" as in my own example for my tube amps (and in the experience of many other people).

I don't know anyone who actively listens to all styles of music, from classic chamber music to orchestral, to jazz over folk and rock to pop, rap, house and techno. (And certainly not someone using a tube amplifier to do so). So I'm not that sure about the experience of 'many other people'. What I do see is many people constantly switching gear.

This is what I'm getting at. I understand the view "go for neutrality and then you can fiddle with changing the sound to taste with EQ in certain circumstances and then go back to neutrality at will." It's just the assumption that this is "the best way to go about it"...which may be true for some people and their own goal, but not for others. Some may indeed find a "set and forget" type of coloration, even if slight, is preferable.

I don't have a problem with applying permanent low and high end eq to taste. As already mentioned before, even if you go the more objective route choosing a certain target house curve already involves preference.

But this discussion took the direction of adding distortion. Adding permanent distortion is something I don't consider to be effective in all situations. As a recording engineer you also don't apply tape saturation to all input channels and recordings, although it can be very effective in specific use cases. As a customer it's also very difficult to add the right type and amount of distortion by selecting different types of gear. But if it works for you, fine by me.

These musicians weren't using top studio systems. Tons of music today is being produced in essentially home studios like the ones my brother and friend use. They cobble together whatever speakers they can afford.

That's why in my argumentation I referred to 'well respected' engineers and studio's. Crap in is crap out. It's obvious that if we want to break the circle of confusion an effort is needed at both ends of the recording process. The process needs to be well executed. People making music in home studio's is a given, you can only hope they collaborate with a competent mastering engineer. He takes care of providing a consistent sound, and he typically doesn't use NS10's to do so. If that fails and sound character is all over the place, then I don't see how adding permanent distortion would solve this (unless it's meant to hide imperfections). It's even more reason for custom eq (and replay gain).

 

[rwortman]

That's why mixing is followed by a mastering phase, using a reference system operated by a very experienced sound engineer. That engineer will compare the final product to lots of other well sounding records, and try to match or even top their quality (if no one is forcing them to just make it loud as hell). It's not perfect of course, but works reasonably well. Especially since studio and HiFi systems can be calibrated to a certain extent. Well respected mastering engineers are pretty consistent in the quality they deliver. It's their area of expertise.


That's true of course, but if someone finds most music sounds crap at home I would advise him to review his system before blaming the recordings.

The same circle of confusion applies to the mastering system. You don’t have its duplicate at home, so you don’t get that sound.

A recording doesn’t have to “sound like crap” for there to be room for improvement. Some recordings sound fantastic, most are pretty good, some I wander what they were thinking and I wish I could have the project files to remix it myself. I don’t know, maybe spending hundreds of hours behind a mixing console and messing around in digital audio workstations made me somewhat less inclined to regard a commercial recording as a sacrosanct thing.

 

[Geert]

The same circle of confusion applies to the mastering system. You don’t have its duplicate at home, so you don’t get that sound.

It's a mastering engineers job to make the recording translate to other systems so you don't need a exact duplicate system at home. And no, the process is not perfect. Never claimed that.

I don’t know, maybe spending hundreds of hours behind a mixing console and messing around in digital audio workstations made me somewhat less inclined to regard a commercial recording as a sacrosanct thing.

With the average commercial recording you've choosen the worst case scenario.

 

[Blumlein 88]

You certainly cannot unlock the circle of confusion if you refuse to unconfuse yourself when it is possible.

 

[ahofer]

...and we've collapsed into an analogy singularity again...

No analogy is perfect, hence the analogy ”circle of confusion”.

 

[rwortman]

It's a mastering engineers job to make the recording translate to other systems so you don't need a exact duplicate system at home. And no, the process is not perfect. Never claimed that.


With the average commercial recording you've choosen the worst case scenario.

How is average worst case? By definition it’s the average case.

 

[Axo1989]

Yes it does. You can break the circle by using a calibrated well designed system, and you can avoid enforcing the circle by not adding distortion or series of components that change the frequency response. That was my message.

In the cases you elaborate here, you discuss production workflow from the source, that goes beyond "all you have is the recording". As you say "it's obvious that if we want to break the circle of confusion an effort is needed at both ends of the recording process" which is my original point, in different words.

How is average worst case? By definition it’s the average case.

Yes, I'm getting some semantic breakdown reading/responding to @Geert

 

[pablolie]

The "average recording" these days is woeful. There is autotune and it's mixed for a crowd that couldn't care less. It'll sound "good" on a carefully put together audiophile shrine, but never ever think it's a reference for anything that requires linearity, accurate timbre, revealing staging ... etc etc.

The truth is that we ultimately set up our systems to be as close as ideal as we think we can for certain stuff. Let's admit it: we don't use random songs to evaluate or set up our stuff. We revert to familiar stuff that we also love to listen to, and which we have listened to a hundred times before, probably. And that's awesome.

But we also all know that sometimes we listen to a good recording but all of a sudden, I don't know... the piano seems 30ft wide or the solo guitar sounds borderline like the speakers are out of phase. Sometimes I can be addressed by moving the speakers but come on, once I have my speakers set up for my fav recordings, I will not move them. Same with room correction settings that could help some recordings. Nope, my stuff is static.

With this I am not all all implying that measurements are therefore less relevant - they are very important to me. But I think we all tend to optimize to our private favorites rather than that elusive ideal industry standard. All I am saying is that a perfectly linear, "ideally measured" system may not give you the enjoyment you thought if it simply doesn't work with your favorite (and possibly flawed) recordings.

 

[rwortman]

If one starts from the file on the CD/Stream/LP, and reproduce that with transparent electronics and nominally flat loudspeakers, then there's no Circle of Confusion. What you get at home is the best rendition of what's on the CD. What came before that is of no concern, cannot be of concern as we have no control over it, only what our equipment does.

Choosing equipment that we like the sound of, not because it's accurate, will create the Circle of Confusion, as what we like can change depending on what else we hear, we might hear something we like more. Ignore 'what you like' and just buy accurate, and play the CD. No circle of confusion.

S.

The problem with this reasoning is that what is on the CD is their because of what someone heard in on a playback system in a room, so accurate to what? In an interview Floyd Toole asked the question, ”are we looking for accuracy or realism”?, because sometimes the recording isn’t very realistic sounding and needs a little help.

I really can’t imagine why anyone would spend significant amounts of money to procure a system and spend hours listening to said system when the goal was anything but to produce a sound they liked. As long as we aren’t telling or selling lies about it, I don’t see the problem.

 

[Geert]

How is average worst case? By definition it’s the average case.

Better?

With the average commercial recording you've choosen the worst case scenario.

 

[Purité Audio]

The problem with this reasoning is that what is on the CD is their because of what someone heard in on a playback system in a room, so accurate to what? In an interview Floyd Toole asked the question, ”are we looking for accuracy or realism”?, because sometimes the recording isn’t very realistic sounding and needs a little help.

I really can’t imagine why anyone would spend significant amounts of money to procure a system and spend hours listening to said system when the goal was anything but to produce a sound they liked. As long as we aren’t telling or selling lies about it, I don’t see the problem.

I have never found that adding distortion in any way improves sound quality.
Keith

 

[gnarly]

I believe there is one major sound quality component induced by the circle of confusion that can be fixed.....overall tonality.

There is zero question in my mind how much tonality varies across recordings.
Tonality often varies even across tracks on the same album/CD, as if either different studios were used, or tracks were mastered by different engineers.

By tonality, i mean at the simplest level, the overall slope of the frequency response curve away from flat.
I like to picture it as a see-saw, with the fulcrum placed at the center of the 10 octaves, so somewhere around 640Hz of so.

If i hear a track that is too bright on my system set to flat response, i think that means the track was mastered with in bass-heavy environment, where the recorded signal was brought down to sound flatter in that situation.
It's the opposite when bass sounds too heavy.....there wasn't enough when recorded, and the electrical signal juice was turned up.

Highs can have the same effect....it's deviation at the ends of the see-saw, that create the most tonal havoc. I think a lack of studio high end response can make for overly bright recordings too.

That's at the simplest level, when any deviation from flat response, any tilt in total tonality is all ascribed to excess of deficient bass, or highs,
and can be corrected with simple high and low shelving EQs.

Occasionally, i hear tracks (on a flat system) that have tonal imbalances in the more middle octaves.

My solution is a different form of playback EQ. I have a 5-way system that is tuned to flat mag and phase response. There is an overall volume control, as well as a volume control adjustment for each of the 5-ways. So almost like a 5 band parametric EQ, other than it's actual driver sections being controlled with no EQ spillover into adjacent bands.
(Astute readers / speaker builders are probably throwing their hands up now, saying 'but that changes acoustic xovers and messes up phase etc'. Lemme just say, not so for my system.)

Anyway, I literally remaster tracks on the go, able to adjust tonality at will. It's awesome to dial a track in just right at whatever SPL sounds best...really makes it come alive.
Most of the time, the sub level is all that needs adjusting, but often enough, the HF and VHF does too. The middle two-octave band almost never needs adjusting, and stays at 0 dB.

I can look at the adjustments made, up or down by band, and envision they represent the imbalance from flat in the mastering environment.
For example, if i need to boost the subs a lot, it means they were listening to a ton of subbage when mastering.

Should add, i listen to studio stuff, and the whole idea of high-fidelity doesn't make sense then, i think.
For me, voices, .....their naturalness, intelligibility, and how easily separable from each other, are my test for fidelity.

I guess my real point in this long winded post, is that imho, tonal imbalances in studio recordings is non debatable, and fixable
(and by means simpler than i go to )

 

[rwortman]

Better?

Who here is listening to private recordings?