Studio Monitors & The Circle of Confusion- What We Know/Don’t Know

[Travis]

Moderator note: this thread was created from existing content from another thread. The following text was copied from the original OP here and is now later in this thread...

I searched high and low for an existing thread on studio monitors and how it relates to COC, came up empty. Saw a few threads on studio monitors under this Topic, Pro-Audio. I like that it’s in Pro-Audio so the focus can be on how monitors are used in mixing and mastering, vs. what was found to be a preference for end user listening.

EDIT: Am adding the following Dr Olive quote to serve as further guidance for the discussion in this thread:

"As Toole points out, the key in breaking the circle of confusion lies in the hands of the professional audio industry where the art is created. A meaningful standard that defined the quality and calibration of the loudspeaker and room would improve the quality and consistency of recordings. The same standard could then be applied to the playback of the recording in the consumer’s home or automobile. Finally, consumers would be able to hear the music as the artist intended."


Start of source content…

It has never been a significant leap in loudspeaker design, regardless of brand, and they all still primarily use the same basic driver technology as the very first loudspeakers made, and other design aspects of loudspeakers don't really make a big difference, or a difference at all, in real-world user scenarios. If Genelec/Neumann speakers were truly that much better than ATC speakers for studio use, where are the results of that? Can anyone point me to any research showing that music productions are generally better using Genelec/Neumann speakers than using ATC speakers?

All talk about this loudspeakers being so much better than that loudspeakers is just bullshit, it has no bearing in real-world use. For some people, a Genelec monitor will be the perfect tool, and for others, that will be a Neumann speaker, while the same is true for people who prefer using ATC speakers. Different tools for different people, that's why you see great audio engineers using all types of different speaker brands, because all of them measure, in the ballpark, good enough. The result is more down to the skills of the audio engineers than what monitor brand they choose to use.

Kisses and hugs.

This is absolutely true, what recording and mixing engineers listen for is completely different than what the end user is listening to. These are called audio attribute and this is a summary from the book described below.

Whatever monitor allows you to hear and distinguish those attributes the best is going to be the one you want to use as a monitoring/mixing engineers.

It’s also very easy for anybody to understand what engineers/mixers listen for in recording and mixing by getting Dr. Corey’s book, Audio Production and Critical Listening: Technical Ear Training which includes audio samples and tests for all the aspects of critical listening for audio engineers. It’s part of the AES Series on Audio Engineering.

It’s easy to prove, one way or another, what difference, if any, monitors make in these attributes. You simply take the tests “blind” with Brand A vs. Brand B.

I have seen Dr. Corey do these tests for audiences, in a couple of different physical situations (you can see him do it on YouTube) and for general examples with larger audiences it didn’t seem to matter what speakers he was using. He may have some preliminary data on how students do with one speaker vs. another, or headphones vs. monitors. He hasn’t published anything on that as of today.


Amazon Description of book and audio tests:
Audio Production and Critical Listening: Technical Ear Training, Second Edition develops your critical and expert listening skills, enabling you to listen to audio like an award-winning engineer. Featuring an accessible writing style, this new edition includes information on objective measurements of sound, technical descriptions of signal processing, and their relationships to subjective impressions of sound. It also includes information on hearing conservation, ear plugs, and listening levels, as well as bias in the listening process.

The interactive web browser-based "ear training" software practice modules provide experience identifying various types of signal processes and manipulations. Working alongside the clear and detailed explanations in the book, this software completes the learning package that will help you train you ears to listen and really "hear" your recordings.

This all-new edition has been updated to include:

• Audio and psychoacoustic theories to inform and expand your critical listening practice.
• Access to integrated software that promotes listening skills development through audio examples found in actual recording and production work, listening exercises, and tests.
• Cutting-edge interactive practice modules created to increase your experience.
• More examples of sound recordings analysis.
• New outline for progressing through the EQ ear training software module with listening exercises and tips.

 

[Travis]

Can anyone point me to any research showing that music productions are generally better using Genelec/Neumann speakers than using ATC speakers?

There isn’t any research, that I can find, one way or the other. Dr. Toole has suggested/hypothesized there is a connection and forms the basis for the Circle of Confusion concept. Most of his critical listening test tracks he used at the NRC were McGill university recordings where he knew who recorded, using monitors he made to try and control for Circle of Confusion variables.

If Dr. Corey is correct about the attributes that audio engineers need to be able to identify (and correct) I would think that he would have some preliminary data on how much the monitors play into that, if at all (headphones) will do.

So far I haven’t seen any research, by anyone, that monitors that measure a particular way improve/degrade critical listening test scores for engineers.

 

[goat76]

Which reference mixes may also have done with coloured equipment and then the circle of confusion goes on and on.

It doesn't matter what type of equipment the reference tracks were made on. What matters is that YOU know they are good and well-balanced sounding tracks, tracks which you have most likely heard on many great audio systems over the years, and they may be tracks that you know other people also regard as great sounding audio productions. I'm quite sure you could name them pretty fast without thinking that long, and I'm also sure you already have a bunch of reference tracks for every genre of music you are interested in. Those are YOUR reference tracks.

If you are a sound engineer, all those reference tracks should sound exactly as good as you've heard them before, on the studio monitors of your choice. If not, then there's something wrong with the sound system, and not with the reference tracks.

When you mix music and put all your focus on small details in the mix, your hearing will most likely start to acclimate itself to the overall tonal balance of the not yet finalized audio mix, even if/when the overall tonal balance is fairly skewed. At this point, it doesn't help even if you have the most accurate monitor system in the world; your hearing has still acclimated itself to the tonal balance of the mix without you even noticing if it's skewed or not.
The tonal balance of the mix has become the norm, so what the hearing regularly needs is a "recalibration" to a well-known reference, and that is the reference tracks of your choice. When your audio production starts to "mimic" your reference tracks in the overall tonality, you can also be pretty sure it will translate well to other systems in a similar way as the reference tracks.

So, in short. The most accurate-sounding monitor system will not be the solution to "the circle of confusion" on its own. The solution is well-known reference tracks, but it will also be helpful as a starting point that those tracks sound as you expect them to sound on the studio monitors, whatever those studio monitors may be.

 

[Joe852]

The obsession with frequency response is silly when we all know anyone who cares can eq them ruler flat if that's important to them.

 

[lowkeyoperations]

Again, it’s simple logic that the chances are higher that someone will produce a mix which sounds well/neutral on neutral loudspeakers when using neutral loudspeakers.

Unless you have a study, or any other form of evidence (which ASR is based on) then this is simply a combination of conjecture and blind faith.

Is there any study that supports your thesis?

 

[lowkeyoperations]

Provided by Genelec:
View attachment 460534
Assume your system is introducing +5 db bass coloration.

Step	Equation	Value
True bass level	x	0 dB
What you hear while mixing	x+ 5	5 dB (too much)
You compensate	y=x−5	-5 dB (too little)
Final mix bass	y	-5 dB

Primary school mathematics. You guys talk smug while not being able to figure out the most basic transfer function equations.

You’re so set on your primary school dogma that you haven’t realised that real life contains more variables.

For example. You know your speaker has 5dB extra bass, because you have listened to music on them before. You’ve also listened to other speakers in your lifetime. So the +5db response isn’t a surprise.

So when you mix, you don’t reduce the bass by 5dB. You mix bass heavy, knowing the speakers aren’t flat.

True bass level 0dB
What you hear when mixing +5dB
You compensate 0dB
Final Mix Bass 0dB

Even primary school maths adds up :lol:

This is why NS10s can still be useful tools.
You don’t try to mix bass on them to sound like ATC-200s. You mix bass on them to sound like NS10s.

Your conjecture might hold true for a primary school age kid who has never heard the speakers before, who didn’t bring a collection of reference tracks to audition and who didn’t realise that not all speakers and rooms are perfectly flat, and who never bothered checking their mix on a second system such as headphones, an iPhone, a car stereo or anything else that makes sound.

Yours is a drastic example at +5dB. In real life it’s probably less, and that’s why someone earlier said something along the lines of “close enough to be useful”. If your monitor has a 2dB boost from 2-4kHz you can accommodate that in real life without too much drama. You don’t do it by cutting 2dB in that area though. Unless you have no idea what you are doing.

 

[Blockader]

You’re so set on your primary school dogma that you haven’t realised that real life contains more variables.

For example. You know your speaker has 5dB extra bass, because you have listened to music on them before. You’ve also listened to other speakers in your lifetime. So the +5db response isn’t a surprise.

So when you mix, you don’t reduce the bass by 5dB. You mix bass heavy, knowing the speakers aren’t flat.

True bass level 0dB
What you hear when mixing +5dB
You compensate 0dB
Final Mix Bass 0dB

Even primary school maths adds up :lol:

This is why NS10s can still be useful tools.
You don’t try to mix bass on them to sound like ATC-200s. You mix bass on them to sound like NS10s.

Your conjecture might hold true for a primary school age kid who has never heard the speakers before, who didn’t bring a collection of reference tracks to audition and who didn’t realise that not all speakers and rooms are perfectly flat, and who never bothered checking their mix on a second system such as headphones, an iPhone, a car stereo or anything else that makes sound.

Yours is a drastic example at +5dB. In real life it’s probably less, and that’s why someone earlier said something along the lines of “close enough to be useful”. If your monitor has a 2dB boost from 2-4kHz you can accommodate that in real life without too much drama. You don’t do it by cutting 2dB in that area though. Unless you have no idea what you are doing.

how do you know that the music you listen can be regarded as a reference? How do you know that the music you enjoy listening has correct level of bass, mid and treble? How did you even conclude that it passes as a reference? Is it an opinion or is it something you can be sure of? Even if you have one perfectly recorded music, how do you know how your room modifies the recording compared to a real reference like anechoic room? You need to substract your rooms influence from the ideal perfect reference to have a real reference.

thewas and I are parroting Toole and Sean Olive's concept called circle of confusion. It is part of the Toole and Olive's research on sound reproduction.

I will write longer later about why one can never have a true reference.

 

[Rettica]

how do you know that the music you listen can be regarded as a reference?

if it sounds good on many different systems in many different rooms/spaces/situations

when talking about 'true reference' we approach philosophical territory rather than a real life practics

 

[dfuller]

how do you know that the music you listen can be regarded as a reference? How do you know that the music you enjoy listening has correct level of bass, mid and treble? How did you even conclude that it passes as a reference? Is it an opinion or is it something you can be sure of? Even if you have one perfectly recorded music, how do you know how your room modifies the recording compared to a real reference like anechoic room? You need to substract your rooms influence from the ideal perfect reference to have a real reference.

thewas and I are parroting Toole and Sean Olive's concept called circle of confusion. It is part of the Toole and Olive's research on sound reproduction.

I will write longer later about why one can never have a true reference.

So in my experience doing this professionally - you aren't looking for something to be tonally perfect, that doesn't exist.

You use specific characteristics in each song as "tells". I have songs on my reference playlist that objectively do not sound good, but I use characteristics of them as "does this speaker represent this range/this characteristic correctly". For example - I use a song with really piercing bright synth stabs to check high end extension. They should sound bad if a speaker is accurate in the high end. If it sounds good, I know the top end is kind of rolled off. If it's so bright I can't listen to it, the top end is boosted.

 

[Purité Audio]

Wouldn’t the speaker’s anechoic and in room measurements tell you whether the treble is ‘boosted’?
Keith

 

[dfuller]

Wouldn’t the speaker’s anechoic and in room measurements tell you whether the treble is ‘boosted’?
Keith

You don't always have those on hand.

Besides that, having the graphs doesn't always exactly tell you what things sound like. Frankly most data is at best incomplete, if it's even available.

Plus there are things we have no way to objectively measure e.g. perceived stereo width.

 

[Purité Audio]

Thus I would have thought the need for neutral/transparent monitors in the creation and ideally in its reproduction.
I can understand the usefulness of a ‘known qualtity’ but surely it would be preferable if that reference were transparent.
Keith

 

[dfuller]

Thus I would have thought the need for neutral/transparent monitors in the creation and ideally in its reproduction.
I can understand the usefulness of a ‘known qualtity’ but surely it would be preferable if that reference were transparent.
Keith

Even "neutral" varies. Genelecs don't sound like Neumanns don't sound like D&Ds don't sound like Kiis.

 

[Purité Audio]

Kii/D&D/Genelec sound pretty similar, and can be adjusted to sound very similar.
Keith

 

[thewas]

Um..... data? This is art.

One of the absolute most worthless concepts ever conceived with regards to audio - spare me

Music is art, having the chance to hear something how it was intended to be with a consequent defined reproduction chain is not.

 

[Blockader]

What you guys basically saying is that, there is a chinese text and people translated it to English, Danish, French and German. Because people from all these nationalities enjoyed, things should be fine.

if it sounds good on many different systems in many different rooms/spaces/situations

when talking about 'true reference' we approach philosophical territory rather than a real life practics

We would believe you if this wasn't the result:

These are calibrated, expensive, and supposedly neutral monitors used in professional rooms... yet the variation is massive, especially below 1 khz.

Max/Min Range (magenta): Swings up to ±15 dB across frequencies.
90% Variation (black): Still ±5 to ±10 dB below 200 Hz.
Even within 50% of the rooms, you still get ±3–5 dB deviations.

This graph above proves that human hearing is not an absolute measuring device: it's adaptive, biased, and incredibly sensitive to context.

you guys want us to put trust in this cycle you guys stuck in:
Skewed Room → Skewed Mix → Skewed Reference Track → Skewed Playback Tuning → Skewed Perception → Repeat.

There’s no absolute reference in music reproduction because there was never one in music production. Unless all rooms, all monitors, and all listeners agree on a neutral standard (which they never do), we can never take your "trust me bro, I know how reference should sound like" serious. If you read Toole's book on sound reproduction, you'll see why wee are so skeptic about the "bro-science" attitude that audio engineers have.

Despite common belief, the data consistently shows that audio engineers often overestimate the accuracy of their own hearing. Many rely on subjective impressions and personal experience rather than controlled evidence: essentially a form of 'audio bro science.' For decades, we’ve been told to trust the ears of professionals who, in practice, operate within highly inconsistent monitoring environments. Until the industry can demonstrate objective consistency in perception and decision making, skepticism toward these claims remains not only reasonable, but necessary.

 

[thewas]

It doesn't matter what type of equipment the reference tracks were made on. What matters is that YOU know they are good and well-balanced sounding tracks, tracks which you have most likely heard on many great audio systems over the years, and they may be tracks that you know other people also regard as great sounding audio productions. I'm quite sure you could name them pretty fast without thinking that long, and I'm also sure you already have a bunch of reference tracks for every genre of music you are interested in. Those are YOUR reference tracks.

If you are a sound engineer, all those reference tracks should sound exactly as good as you've heard them before, on the studio monitors of your choice. If not, then there's something wrong with the sound system, and not with the reference tracks.

When you mix music and put all your focus on small details in the mix, your hearing will most likely start to acclimate itself to the overall tonal balance of the not yet finalized audio mix, even if/when the overall tonal balance is fairly skewed. At this point, it doesn't help even if you have the most accurate monitor system in the world; your hearing has still acclimated itself to the tonal balance of the mix without you even noticing if it's skewed or not.
The tonal balance of the mix has become the norm, so what the hearing regularly needs is a "recalibration" to a well-known reference, and that is the reference tracks of your choice. When your audio production starts to "mimic" your reference tracks in the overall tonality, you can also be pretty sure it will translate well to other systems in a similar way as the reference tracks.

So, in short. The most accurate-sounding monitor system will not be the solution to "the circle of confusion" on its own. The solution is well-known reference tracks, but it will also be helpful as a starting point that those tracks sound as you expect them to sound on the studio monitors, whatever those studio monitors may be.

That is not correct, as said the reference tracks can have been done with coloured equipment, it is much easier and expedient to work with non coloured equipment and the only chance to finally end this circle of confusion.

 

[thewas]

The obsession with frequency response is silly when we all know anyone who cares can eq them ruler flat if that's important to them.

As its well known EQ cannot correct directivity issues.

 

[Count Arthur]

Music is art, having the chance to hear something how it was intended to be with a consequent defined reproduction chain is not.

Music creation is art.

Music recording and reproduction is science and engineering.

 

[thewas]

Unless you have a study, or any other form of evidence (which ASR is based on) then this is simply a combination of conjecture and blind faith.

Is there any study that supports your thesis?

It simple logic that with a neutral chain you can better, easier and in a more targeted way get to your result. Do you have a study that can show the opposite?