Alternative perspective on the Harman loudspeaker preference curve.

[YSC]

Yep.

This is why I've often pointed out that the criticism of a more colored system on the basis "because it may make some recordings sound better, but not others" applies to a neutral system as well, due to variations in recording quality.

can't agree on the latter part.

1) yes, the recording and mixing quality are all over the place, but for most cases, the "intentional" balance is chosen by the producer, and their system is "supposed to be neutral", so in order to enjoy it as what it was intended you still need an honest, neutral system and not a second flavour to all content with the coloured speakers

2) in a neutral system with well controlled directivity, you can easily and consistently EQ in a few presets for e.g. bass shy, bass boosted, treble boosted, treble cut etc. without throwing the balance off due to reflected sound of directivity error. but in a coloured speaker, more often than not the directivity is least of their concern, so EQing the trend always get something weird kicking in.

It's much like going to restaurant, order food, taste a bit and add your own desired extra flavour v.s. blindly put pepper sauce on everything before even tasting

 

[MattHooper]

can't agree on the latter part.

1) yes, the recording and mixing quality are all over the place, but for most cases, the "intentional" balance is chosen by the producer, and their system is "supposed to be neutral", so in order to enjoy it as what it was intended you still need an honest, neutral system and not a second flavour to all content with the coloured speakers

The "circle of confusion" puts that in doubt.

2) in a neutral system with well controlled directivity, you can easily and consistently EQ in a few presets for e.g. bass shy, bass boosted, treble boosted, treble cut etc. without throwing the balance off due to reflected sound of directivity error. but in a coloured speaker, more often than not the directivity is least of their concern, so EQing the trend always get something weird kicking in.

Then each time you mess with EQ to taste, you are no longer playing back with a neutral system: you've colored the sound. This is what I mean by the fact that playing back on an actually "neutral system" doesn't get you out of the problem of "sounds good with some recordings, not good with others."

If someone chooses a colored system that, to him/her sounds great across a wide variety of their music, that can work too.

 

[JoachimStrobel]

I believe the message is: Do not over-engineer your system. Good speaker and a bass and treble control is enough. And bass and treble is like salt in a restaurant’s table: Despised by gourmets but needed by others and as a restaurant should care for all customers so should vendors.

The trouble starts with Multichannel. There is a need to level- and time align the speakers. As to their varying size or their obscure placement, a tonal matching seems appropriate. But then one needs a target curve. And only then are rich discussion on that subject fruitful in my mind.

 

[Sancus]

The "circle of confusion" puts that in doubt.

The industry does seem to be converging though, the adherence of speakers to the flat on-axis, smooth off axis rule is much better now than 10-20 years ago, let alone 40+. The most popular mixing and monitoring tools more-or-less adhere to this too.

And anyway if you start neutral with a well designed speakers and use tone controls or EQ you can account for all possibilities.

Not much reason to pick speakers with weird tonality especially if they come with busted off axis.

 

[YSC]

The "circle of confusion" puts that in doubt.



Then each time you mess with EQ to taste, you are no longer playing back with a neutral system: you've colored the sound. This is what I mean by the fact that playing back on an actually "neutral system" doesn't get you out of the problem of "sounds good with some recordings, not good with others."

If someone chooses a colored system that, to him/her sounds great across a wide variety of their music, that can work too.

It's just like the pepper sauce analogy, a "neutral" system gets you as close to the mixer/artist intended it to be sounding, which is what you "should" appreciate in the raw sense, and for individual recordings, if one decides to add flavour in later, you can always do so with on and off axis sound balance not being off.

Sure when you can enjoy (sounds good) the music you owned, it's perfectly fine for that individual at that moment, nothng stops ppl enjoying ancient radio music right? it's a fine music system for a lot of ppl, it is not a must to chase for better sytem, but if one is asked for a more "hifi" sound or a "better system", then the goal should converge to a more neutral speaker with good directivity and low distortion. Nobody is forcing others to chase indifinately for the next best thing

 

[MattHooper]

I don't don't disagree with most of the points made for the neutral system. Though I don't think it's unreasonable for different approaches.

I was just making a very specific point though, that a truly neutral system does not avoid the problem of "sounds good on certain recordings, doesn't sound as good on others."

I don't disagree that in principle you can't me more fine tuned in how you color it to taste with poorer recordings. But...as I said, sticking with pure neutrality leaves you with essentially the same problem of "sounds good/sounds worse" issue with recordings, and if you are going to abandon neutrality based on recordings, then you end up coloring the sound anyway.

This type of tweaking would certainly be an advantage for the individual who is turned off by a slightly colored system. But for another individual who has found some coloration they like that enhances pretty much everything to their ears, and is a set-and-forget no need to fiddle result, then that can make sense too.
I mean, I know some will always want to start with a fully neutral system and "tweak from there," but I've enjoyed certain not-exactly-neutral systems that I don't think would be perfectly replicated just fiddling with EQ.

 

[sigbergaudio]

Having had the privilege of visiting studioes the past year, I think there are some misunderstandings with regards to both the circle of confusion, what neutral really means, and also what "the artist intended" or what "the engineer intended" really means.

Some points / facts, that have somewhat counter intuitive consequences:

1) Wildly inaccurate monitors like the Yamaha NS10 are still present in a vast number of studios. However, this doesn't necessarily mean that the mix is inaccurate, since engineers typically both know their monitors quite intimately, and also check how their mix translate to other systems on a regular basis. Thus, it is not impossible to produce a balanced mix on inaccurate monitors (though unecessarily difficult).

2) Building on the first point, these inaccurate monitors have been around for so long, many engineers think that inaccurate monitors is a good thing, as it helps them understand how the mix translates to "poor" systems. In practice, mixes that sound good on neutral monitors will typically also sound good on lesser systems.

3) It sometimes appears like people think the sound actually goes THROUGH the monitors, but that is of course not the case (and everyone understand that if they think about it). This means the monitors used in the control room, mixing room or mastering room obviously does not directly affect sound quality.

4) "What the artist / engineer intended" - In practice I think this "correct" reproduction of what the engineer or artist heard doesn't really exist except in the broadest terms. This past year I've been sitting next to several engineers listening through their own mixes through our speakers, going "oh." realizing that if they've had this transparency available when creating the mix, they would have done things differently. Also, with the exception of black metal and trash metal bands, most artists don't intentionally want to hurt your ears, so if you think the treble is excessive, or the bass too lean, your favorite artist won't think less of you if you adjust your setup.

5) Too thin or too lean is not accurate. Without exception, all engineers I've collaborated or presented sound systems to have preferred a room curve with 6-8dB rise in the bass (50hz-10khz), and felt that this is the closest to listening live. Personally I've also never been to a live / acoustic concert where the instruments have sounded hard or lacking of bass. Quite the contrary.

6) Building on the previous point: If you pain yourself through a too lean system that you don't really enjoy, but think to yourself "at least it is accurate" or "at least this is the way the artist intended it", you are probably wrong on both accounts.

7) An anechoically flat or close to flat system, will NOT be flat in most rooms with a reasonably listening distance.

8) Most home stereo systems are too lean in the 100-500hz area, even when the speakers are anechoically flat and thus "correct". So completely neutral speakers may not always be the answer for "accurate" in-room sound. This is somewhat counter intuitive, since we want all our other components (amps, sources) to be neutral.

 

[JoachimStrobel]

Having had the privilege of visiting studioes the past year, I think there are some misunderstandings with regards to both the circle of confusion, what neutral really means, and also what "the artist intended" or what "the engineer intended" really means.

Some points / facts, that have somewhat counter intuitive consequences:

1) Wildly inaccurate monitors like the Yamaha NS10 are still present in a vast number of studios. However, this doesn't necessarily mean that the mix is inaccurate, since engineers typically both know their monitors quite intimately, and also check how their mix translate to other systems on a regular basis. Thus, it is not impossible to produce a balanced mix on inaccurate monitors (though unecessarily difficult).

2) Building on the first point, these inaccurate monitors have been around for so long, many engineers think that inaccurate monitors is a good thing, as it helps them understand how the mix translates to "poor" systems. In practice, mixes that sound good on neutral monitors will typically also sound good on lesser systems.

3) It sometimes appears like people think the sound actually goes THROUGH the monitors, but that is of course not the case (and everyone understand that if they think about it). This means the monitors used in the control room, mixing room or mastering room obviously does not directly affect sound quality.

4) "What the artist / engineer intended" - In practice I think this "correct" reproduction of what the engineer or artist heard doesn't really exist except in the broadest terms. This past year I've been sitting next to several engineers listening through their own mixes through our speakers, going "oh." realizing that if they've had this transparency available when creating the mix, they would have done things differently. Also, with the exception of black metal and trash metal bands, most artists don't intentionally want to hurt your ears, so if you think the treble is excessive, or the bass too lean, your favorite artist won't think less of you if you adjust your setup.

5) Too thin or too lean is not accurate. Without exception, all engineers I've collaborated or presented sound systems to have preferred a room curve with 6-8dB rise in the bass (50hz-10khz), and felt that this is the closest to listening live. Personally I've also never been to a live / acoustic concert where the instruments have sounded hard or lacking of bass. Quite the contrary.

6) Building on the previous point: If you pain yourself through a too lean system that you don't really enjoy, but think to yourself "at least it is accurate" or "at least this is the way the artist intended it", you are probably wrong on both accounts.

7) An anechoically flat or close to flat system, will NOT be flat in most rooms with a reasonably listening distance.

8) Most home stereo systems are too lean in the 100-500hz area, even when the speakers are anechoically flat and thus "correct". So completely neutral speakers may not always be the answer for "accurate" in-room sound. This is somewhat counter intuitive, since we want all our other components (amps, sources) to be neutral.

I confess that I do not understand why one can not match a loudspeaker room response to a concert sound: Record an non-amplified Jazz sextet (incl. horn and reed) in a small concert hall und “fit” the house curve to that envelope. That will need more than two speaker, but will 11 speaker suffice? What is wrong with that concept?

 

[sigbergaudio]

I confess that I do not understand why one can not match a loudspeaker room response to a concert sound: Record an non-amplified Jazz sextet (incl. horn and reed) in a small concert hall und “fit” the house curve to that envelope. That will need more than two speaker, but will 11 speaker suffice? What is wrong with that concept?

I am not sure I understand exactly what you are trying to achieve here, but I assume it is to make the sound at home as close as possible to what you hear in the concert hall? There are a few of challenges with reproducing what you experience a concert venue.

The first one is dynamic range. The difference between the softest and loudest passages in a non-amplified concert hall like you are describing is very large. This means during the production process the dynamic range is compressed, to make it possible to listen and enjoy at moderate levels (few listen at concert levels in their own home). How agressive this process is varies from record to record.

The second is that the acoustics in such a large space is very different from a regular living room, and the acoustics and reflections determine the natural house curve. So the natural house curve in a small concert hall is not equal to that of your room. As you imply you can certainly go some way to mimic this with a surround system, and even a well recorded live event combined with a well setup stereo system works surprisingly well.

The third is the fact that you are listening to a recording of an event, which also (depending on how it was recorded) contain the acoustic effects and spatial cues from the original venue. So when you replay that in your living room, you get a "double room" effect. The acoustic effect of the original venue, and the acoustic effect of the space you are listening in. So typically you want to limit the latter, by dampening the room and choosing speakers that are not too affected by it. Luckily you don't typically get a "double room curve", because the producer of the record also listened to it in a room when mixing/mastering it, so he/she has already balanced the mix so it sounds right when replayed in a room from that perspective.

Not sure if this adressed your question, please clarify if not

 

[JoachimStrobel]

Many thanks. I am thinking of a small jazz venue for 60 people with a size of around 200 Sqm. So it is not too far out from a 60 sqm living room.
And while dynamics could be overcome, the comment about double-room is great.
And that seems to be the problem.
And yes, sometimes an over enthusiastic music lover mics the instruments up and squeezes the sound through loudspeakers. Therefore creating a horrible mix of “room responses”
But then again: If the instruments would be recorded in cubicles with close mics and then played in this concert hall, then I should be able to retrieve a transfer function. If I then were able to buy music with instruments recorded without a room response then I could mimic that concert hall. However, the music would be horrible as the musicians would not be able to interact. Future AI codes might be able to extract a room-free instrument response from any recording and add a room-transfer function. I guess that leads to synthetic music and may not be something to aim for.

 

[sigbergaudio]

Many thanks. I am thinking of a small jazz venue for 60 people with a size of around 200 Sqm. So it is not too far out from a 60 sqm living room.
And while dynamics could be overcome, the comment about double-room is great.
And that seems to be the problem.
And yes, sometimes an over enthusiastic music lover mics the instruments up and squeezes the sound through loudspeakers. Therefore creating a horrible mix of “room responses”
But then again: If the instruments would be recorded in cubicles with close mics and then played in this concert hall, then I should be able to retrieve a transfer function. If I then were able to buy music with instruments recorded without a room response then I could mimic that concert hall. However, the music would be horrible as the musicians would not be able to interact. Future AI codes might be able to extract a room-free instrument response from any recording and add a room-transfer function. I guess that leads to synthetic music and may not be something to aim for.

I visited such a jazz venue not too long ago, and there's something about how the sound spreads from actual instruments in a room that I think is pretty difficult to mimic.

But the overall tonality could of course be replicated, and my impression that the tonality in such a situation is pretty warm and full. So I think what is actually neutral, in the sense of the same tonality as such a situation, is what most people would consider warm / somewhat bass heavy. It seems like many of the neutral purists actually have a system that is too lean / have too little bass.

 

[ZolaIII]

Things in reality are much more complicated. On white noise - 20 dB calibrated system to 86 dB SPL stereo i prefer even little less bass but still the same shape and about average - 1 dB per octave in highs starting from 1 KHz. I use normalisation on the sources that I can in the EBU R128 form and equal loudness normalisation (ISO 226 2003) to the desired SPL. However there are objective factors that don't meet the eye in the FR reading but sure pop in others. For this purpose I will stay on waterfall plots, RT60 decay times and clarity. You can have great FR but pore decay times lowering the clarity (basically ringing) and even single energy peek in waterfal plots (usually room mood) can smare large portion of the response or completely depending where it is and how pronounced so dealing with those is priority. Recently I started doing my sub's slope responses as close to mathematical ideal slopes as possible crossed high with my mains (120 Hz) and did all possible to annul their (mains) response under the crossover point (using PEQ, low self and high pass filters) as even with port plugs they where way off. This resulted in better RT60 decay times 4 to 6 dB in under 120 Hz area and improved overly clarity. This is not an easy thing to accomplish in the small room like mine.
There are numerous subjective factors that shouldn't be neglected. Like current state of hearing health and it's adoption. Most people's brain is not adopted to get a full range bass response nor happy when suddenly put in such position. It even complains and you feel it as discomfort. Man are in favourable position hire as their vocal range starts lower (and that's the most acceptable range for the brain as it's natural communication range) then female one. All of this is of course high speculative. Old folks with hearing loss mostly in highs and possible Tinitus will avoid brightness in the ringing range and prefer lower SPL on average (all to keep it under control) and with it they will also want lower bass levels. For me it's remarkable how fast actual auditory system (brain in the first place) adopts to the changes (discarding additional details that are not crucial). It needs 30 to 60 min and you need to increase SPL level up for about 1.5 dB in that tame scale in order that it keeps perceived details from the beginning of listening session. My warm recommendation is for that among other reasons you try to start listening on moderate listening levels so that you can end the season listening at loud levels and that way you don't lose any pleasure in listening (with ISO 226 2003 of course) and protect your hearing. Of course neighbors and hause hold members will probably also appreciate it.
Of course speakers shouldn't be pushed much and absolutely not to do what their drivers possibly can not. Having a potent main to high bass woffer is very important for dynamic property's of system so cut your mains properly and high enough to get them there.
Best regards and have a nice time.

 

[HiMu]

I visited such a jazz venue not too long ago, and there's something about how the sound spreads from actual instruments in a room that I think is pretty difficult to mimic.

Sound engineer Dave Rat built a speaker that tries to mimic how sound is radiating from instruments, he also talks about the differences of recording instruments for that kind of playback system.

About recording a guitar with 4 mics

 

[tifune]

3) It sometimes appears like people think the sound actually goes THROUGH the monitors, but that is of course not the case (and everyone understand that if they think about it). This means the monitors used in the control room, mixing room or mastering room obviously does not directly affect sound quality.

Can you elaborate on this? I'm not entirely clear what you mean

Most home stereo systems are too lean in the 100-500hz area, even when the speakers are anechoically flat and thus "correct". So completely neutral speakers may not always be the answer for "accurate" in-room sound. This is somewhat counter intuitive, since we want all our other components (amps, sources) to be neutral.

Also, any way you could illustrate this would be appreciated. Spins, room dimensions, whatever it may be

 

[Justdafactsmaam]

Sound engineer Dave Rat built a speaker that tries to mimic how sound is radiating from instruments, he also talks about the differences of recording instruments for that kind of playback system.

About recording a guitar with 4 mics

Seems like a bad idea. Real instruments aren’t two channel playback. Real instruments all have unique radiation patterns. The recorded sound and the sound we hear is a result of those radiation patterns and how they interact with the performance venue. It’s already done.

 

[HiMu]

Can you elaborate on this? I'm not entirely clear what you mean

If you think about how doctors use stethoscope to listen to your heart, your blood doesn't physically flow through the instrument.

Seems like a bad idea. Real instruments aren’t two channel playback. Real instruments all have unique radiation patterns. The recorded sound and the sound we hear is a result of those radiation patterns and how they interact with the performance venue. It’s already done.

The concept is literally trying to tackle the problems that you wrote? Please watch the part(s) where he explains the differences and requirements of recording the unique radiation patterns and how the instruments couple with the room acoustically, compared to stereo or multichannel reproduction.

 

[MattHooper]

Sound engineer Dave Rat built a speaker that tries to mimic how sound is radiating from instruments, he also talks about the differences of recording instruments for that kind of playback system.

About recording a guitar with 4 mics

As someone intrigued by live vs reproduced sound, I enjoyed Rat's experiment! It was interesting that, even mixed down to stereo and played back in my home, his experimental speaker seemed to capture more of a sense that a real instrument was being recorded. Relative to all the other speakers, the recording of Rat's speaker seemed to produce the most complete and rich sound.

And that gets to what I've mentioned before about playing with surround sound in my system: that there seems to be something about just throwing more speakers in to the mix that can, at least to me, remind me a bit more of real sound. Most stereo playback sounds homogenized to me. But when I have tracks upmixed across my L/C/R speakers, I perceive a richer sound, like a greater breadth of timbre and tonality is being portrayed. And this to some degree actually is reduced if I use some of the room correction modes for "flat" sound matching the speakers tonality.

Rat's thesis is based on the idea that real instruments radiate different aspects of the sound at different angles in to the room, hence he captured the sound from various angles and played back through 4 drivers radiating at those same angles. So you are getting slightly different perspective of the sound pumped in to the room, each with a slightly different timbre, but played together recreates more of the full, rich timbre of the real instrument. He seems to report this is fairly successful.

I speculate that when I leave my L/C/R speakers without the room correction, each is putting out a slightly timbrally different version of the sound, which therefore sounds a bit more rich and complex. In this respect it isn't accurately "re-creating" the exact sound of the recorded instruments, but it is "mimicking" a bit more of the timbral complexity I hear in real sound.

 

[youngho]

Yes, I think that's the average of trained plus untrained. Although Welti commented "There was a fair amount of variation from subject to subject, but if you averaged it all out, you get a bass boost of around 6 dB and a gently rolling off high end, around -2.5 dB at 10 kHz. (Figure 1)," and I already quoted the slides above, I feel like the sheer variance hasn't been highlighted at all here on ASR. It's almost as though we were discussing Target Height as 5' 6.5" because that's about the average height in the US (men about 5'9", women 5'4").

From https://www.psaudio.com/blogs/coppe...r-sean-olive-of-harman-international-part-one:

“Our research identified three segments of listeners based on their preferred headphone sound profiles, as well as the underlying demographic factors (age, gender, and listening experience) associated with each segment. The largest segment (64 percent) prefers the Harman Target Curve and includes all age groups, [levels of] listening experience, and genders. The second largest segment (21 percent) preferred the Harman target with 2 – 3 dB less bass and includes a disproportionate percentage of females and older listeners. The smallest segment (15 percent) preferred the 4 – 6 dB more bass than the target and consisted of mostly younger males."

 

[Justdafactsmaam]

From https://www.psaudio.com/blogs/coppe...r-sean-olive-of-harman-international-part-one:

“Our research identified three segments of listeners based on their preferred headphone sound profiles, as well as the underlying demographic factors (age, gender, and listening experience) associated with each segment. The largest segment (64 percent) prefers the Harman Target Curve and includes all age groups, [levels of] listening experience, and genders. The second largest segment (21 percent) preferred the Harman target with 2 – 3 dB less bass and includes a disproportionate percentage of females and older listeners. The smallest segment (15 percent) preferred the 4 – 6 dB more bass than the target and consisted of mostly younger males."

Which of these three curves is considered to be the most objectively accurate?

 

[Deleted member 21114]

...there seems to be something about just throwing more speakers in to the mix that can, at least to me, remind me a bit more of real sound.

Maybe there is something analogous to the Cheerleader Effect going on. The multitude of speakers means the brain has a harder time discerning flaws (resonances, distortions etc) in each speaker. The more the speaker "gets out of the way" the more natural the sound.