Accurate and boring or colored and fun

[krabapple]

Would either of you guys see any utility in a speaker whose in-room response can be equalized independent of the first-arrival sound?

Isn't that what physical room treatments do, in effect?

Also this whole conversation is a simplified version of reality. In reality, listeners tend to prefer a 'Harman curve' at the LP. 'flat but tilted clockwise'

 

[Duke]

Not me personally, because first-arrival is all I want and all I get. But I would be fascinated to see it done.

I was thinking it might be useful in some situations.

Isn't that what physical room treatments do, in effect?

I'm not sure that room treatments are typically as targeted (in the frequency domain) as EQ can be...

In reality, listeners tend to prefer a 'Harman curve' at the LP. 'flat but tilted clockwise'

... but obviously the point is moot if what listeners prefer is already being done.

* * * *

I thank you both for your replies.

 

[antcollinet]

Do you not see the incoherence in that position? He feels constrained to choose the flattest-possible on-axis speaker, while fully intending to immediately render it much less flat, in pursuit of improvements at the listening position. If you were to buy a car fully intending to immediately paint it orange, why would you care what color it was originally?

How about an equally stupid tshirt dye analogy? If you start with a black tshirt, you can't dye it sky blue.

Similarly if a speaker FR has a serious dip, at a frequency where the room needs a boost (or vice versa), you'll have more of a problem with the EQ. You want your speakers to be flat so you've got maximum room for whatever equalisation you want to do.

 

[Inner Space]

You want your speakers to be flat so you've got maximum room for whatever equalisation you want to do.

Illogical, captain. With no data beforehand about where you need to finish, you can't say where you need to start. Could be a flat speaker might need way more EQ than a random non-flat speaker, in which case the non-flat speaker offers "maximum room".

I understand the lay intuition toward flatness, but with EQ on the menu, flatness is meaningless.

 

[Spocko]

Great question and I have been thinking about this for quite a while, here are my thoughts:

That exact speakers, those with a straight frequency curve, low distortion are dull to listen to? Thinking of studio monitors in the first place. Why would they be dull? Isn't a colorless sound the best, in the long run, if you do not want to get tired of listening?

This is a myth - until recently, most speaker reviewers did NOT measure their speakers so to correlate straight curve to boring is impossible if the reviewer does not have access to said measurements. Additionally, even if a speaker was flat in an anechoic chamber, in-room response could be completely different if the reviewer sat in a null where sound is sucked out at a critical frequency range. So even if the speaker was flat in an anechoic chamber, it's definitely NOT flat in the reviewers' improperly treated room. And most reviewers do NOT have an acoustic analysis of the room modes so they do not even know the location of these nulls in order to either move the speakers or move their chair. Worst case scenario, a truly flat speaker sounds boring, dull and lifeless because all these speakers will be subject to the same suck-out at all the wrong frequencies, and speakers that have peaks (like bright, non neutral speakers) suddenly sound great because they are actually "flat" depending on which reviewer's room they're in!

Here is what a sound engineer I have had some contact with said:
I have worked in the studio and as a sound engineer all my life

My studio monitors are straight, flat and piss off dull but damn accurate!
At home I want to cuddle with a little voodoo-flum and cuddly sound "
He likes old tube amps and vintage speakers, among other things.

Listener fatigue likely? He wants to STOP hearing everything above 8 Khz because he spends his entire day listening/mixing in this fatiguing range so we should NOT be comparing ourselves to a professional who has listener fatigue at the end of the day. He wants to enjoy the music without the "sizzle" so he would likely prefer a roll-off on top to give his ears a rest, no?

But for the rest of us who are not professional sound technicians...
Isn't it easier to have a solution that does not color the sound and when, if you want a colored sound, you plug in an EQ?
I suspect that a colored sound may at first seem attractive (for example, an elevated "disco" bass) but that you get tired of it in the long run.

Assuming your room allows your speaker to play in a neutral manner consistent with listener preferences as posited by the Harman curve, then after listening to your favorite music with speakers that are considered neutral/accurate, are you dissatisfied? Go to the core of your dissatisfaction and discern whether it's the source (the song was mastered for AM radio in the 70's and not digital playback in 2020), your room or your speakers. Sometimes what we believe is flat (speaker measurements in an anechoic chamber) is actually not flat at all and all you need is DSP like Dirac to make the speakers flat as they were intended.

 

[Newman]

even if a speaker was flat in an anechoic chamber, in-room response could be completely different if the reviewer sat in a null where sound is sucked out at a critical frequency range. So even if the speaker was flat in an anechoic chamber, it's definitely NOT flat in the reviewers' improperly treated room. And most reviewers do NOT have an acoustic analysis of the room modes so they do not even know the location of these nulls in order to either move the speakers or move their chair. Worst case scenario, all flat speaker will sound boring, dull and lifeless because all these speakers will be subject to the same suck-out at all the wrong frequencies, and speakers that have peaks or roll-off early (non neutral speakers) suddenly sound great because they are actually "flat" depending on which reviewer's room they're in!

IMO there is a big caveat to what you are claiming. Namely, that if the in-room direct sound (i.e. anechoic sound) is flat, then we are able to detect and prefer that, regardless of the room contribution. So, a speaker that was flat in an anechoic chamber, DEFINITELY IS delivering flat direct sound in the reviewer’s room and that is appreciable. Except for the bass region, where ‘direct sound’ is not a relevant concept and EQ (preferably after room treatment) is essential.

Which is not to deny that reflected sound contribution is also important, but you can’t just sum the two and describe the result as ‘flat’ or not.

 

[Spocko]

IMO there is a big caveat to what you are claiming. Namely, that if the in-room direct sound (i.e. anechoic sound) is flat, then we are able to detect and prefer that, regardless of the room contribution. So, a speaker that was flat in an anechoic chamber, DEFINITELY IS delivering flat direct sound in the reviewer’s room and that is appreciable. Except for the bass region, where ‘direct sound’ is not a relevant concept and EQ (preferably after room treatment) is essential.

Which is not to deny that reflected sound contribution is also important, but you can’t just sum the two and describe the result as ‘flat’ or not.

So if the speaker was flat anechoically, it would be consistently flat above 500Hz you think?

 

[Duke]

Which is not to deny that reflected sound contribution is also important, but you can’t just sum the two and describe the result as ‘flat’ or not.

Agreed.

Imo a speaker should get two things right: The direct sound, and the reflected sound.

Obviously the room plays a role in the latter, but it's easier to preserve spectrally-correct reflections than it is to fix spectrally-incorrect ones via room treatment, so by implication what the speaker is doing off-axis matters in most applications. Not that this is "news" anymore around here.

 

[DanielT]

Great question and I have been thinking about this for quite a while, here are my thoughts:


This is a myth - until recently, most speaker reviewers did NOT measure their speakers so to correlate straight curve to boring is impossible if the reviewer does not have access to said measurements. Additionally, even if a speaker was flat in an anechoic chamber, in-room response could be completely different if the reviewer sat in a null where sound is sucked out at a critical frequency range. So even if the speaker was flat in an anechoic chamber, it's definitely NOT flat in the reviewers' improperly treated room. And most reviewers do NOT have an acoustic analysis of the room modes so they do not even know the location of these nulls in order to either move the speakers or move their chair. Worst case scenario, a truly flat speaker sounds boring, dull and lifeless because all these speakers will be subject to the same suck-out at all the wrong frequencies, and speakers that have peaks (like bright, non neutral speakers) suddenly sound great because they are actually "flat" depending on which reviewer's room they're in!


Listener fatigue likely? He wants to STOP hearing everything above 8 Khz because he spends his entire day listening/mixing in this fatiguing range so we should NOT be comparing ourselves to a professional who has listener fatigue at the end of the day. He wants to enjoy the music without the "sizzle" so he would likely prefer a roll-off on top to give his ears a rest, no?


Assuming your room allows your speaker to play in a neutral manner consistent with listener preferences as posited by the Harman curve, then after listening to your favorite music with speakers that are considered neutral/accurate, are you dissatisfied? Go to the core of your dissatisfaction and discern whether it's the source (the song was mastered for AM radio in the 70's and not digital playback in 2020), your room or your speakers. Sometimes what we believe is flat (speaker measurements in an anechoic chamber) is actually not flat at all and all you need is DSP like Dirac to make the speakers flat as they were intended.

I took the quote from the studio technician mostly because I suspected that it would trigger a discussion here.
People are complex and this with sound has a lot to do with psychology. That was what he said at the time. He's no acquaintance with me, so I do not know if he said that in order to provoke me. So maybe he only felt when he told me.

I may be far off now ( I may be completely wrong) so take it with a pinch of salt. Some different ideas and hypotheses:

He may be mixing and playing music that he does not like to listen to privately, but then it may be about Pavlovian conditioning, that is, sound signature X = music he does not like. If he hears music he likes with sound signature X, it is the speakers that are bad. Which leads to the fact that sound signature Y is what he prefers (even if it is colored) because he associates it with music he likes. Maybe a slightly far-fetched hypothesis, but you do not know.

He may be a Marxist: "Those are my principles, and if you do not like them ... well I have others."

His ears are tired and he wants a sound without details and sharpness when he listens at home.

At work, he sits in the perfect listening position, sweetspoot. At home, he listens in the apartment with kids running around yelling and playing, does not listen in sweetspoot but moves in the apartment when he listens. Then maybe four speakers, which color the sound a bit, placed in his room is preferable to two more colorless that are best suited in his listening room at home when the kids are not screaming and he can sit and enjoy music in the royal chair ,the designated listening chair (which does not exist in his home). Another hypothesis. I have no idea if that is the case. I'm just speculating now.

How do a couple of small studio monitors, with exceptional data measured on paper, such as Gelenec, do at his home IF he only listens to music at very low volume? This compared to a pair of speakers that have an extra boost, an extra increase in the bass range (on paper measured deviation from the ideal frequency curve, ie colored sound). A permanent loudness for those speakers, so to speak. Who knows, the ones in his listening room, with his furniture, at low volume might sound better (which also has to do with taste) than a pair of Gelenec.

Or is it some other simple explanation. He loves bass pumping music and has a couple of big a ... s subwofers at home. He is completely sold at high volume, home speakers with extreme SPL (loud is better).

....and so on and so forth....

 

[Newman]

So if the speaker was flat anechoically, it would be consistently flat above 500Hz you think?

Yes, if gated to prevent the measurement microphone from picking up reflections.

If not gated, it will be “anything goes”, and if you don’t smooth it out, it will look like an intense comb filter. If you do smooth it, it will usually be a fairly lumpy ride that slowly drops in level to the higher frequencies, for a typical domestic room.

Please note that I generally agree with your main points, that anechoically flat speakers are not boring etc, and that the room makes a difference. I just wanted to add that the room can’t compensate for an anechoically not-flat speaker. Like Duke says above, the best sounding speakers get both the direct and reflected components of sound right. The direct component being a flat anechoic response, and the reflected component being the product of the speaker‘s off-axis behaviour and the room.

 

[Frgirard]

I took the quote from the studio technician mostly because I suspected that it would trigger a discussion here.
People are complex and this with sound has a lot to do with psychology. That was what he said at the time. He's no acquaintance with me, so I do not know if he said that in order to provoke me. So maybe he only felt when he told me.

I may be far off now ( I may be completely wrong) so take it with a pinch of salt. Some different ideas and hypotheses:

He may be mixing and playing music that he does not like to listen to privately, but then it may be about Pavlovian conditioning, that is, sound signature X = music he does not like. If he hears music he likes with sound signature X, it is the speakers that are bad. Which leads to the fact that sound signature Y is what he prefers (even if it is colored) because he associates it with music he likes. Maybe a slightly far-fetched hypothesis, but you do not know.

He may be a Marxist: "Those are my principles, and if you do not like them ... well I have others."

His ears are tired and he wants a sound without details and sharpness when he listens at home.

At work, he sits in the perfect listening position, sweetspoot. At home, he listens in the apartment with kids running around yelling and playing, does not listen in sweetspoot but moves in the apartment when he listens. Then maybe four speakers, which color the sound a bit, placed in his room is preferable to two more colorless that are best suited in his listening room at home when the kids are not screaming and he can sit and enjoy music in the royal chair ,the designated listening chair (which does not exist in his home). Another hypothesis. I have no idea if that is the case. I'm just speculating now.

How do a couple of small studio monitors, with exceptional data measured on paper, such as Gelenec, do at his home IF he only listens to music at very low volume? This compared to a pair of speakers that have an extra boost, an extra increase in the bass range (on paper measured deviation from the ideal frequency curve, ie colored sound). A permanent loudness for those speakers, so to speak. Who knows, the ones in his listening room, with his furniture, at low volume might sound better (which also has to do with taste) than a pair of Gelenec.

Or is it some other simple explanation. He loves bass pumping music and has a couple of big a ... s subwofers at home. He is completely sold at high volume, home speakers with extreme SPL (loud is better).

....and so on and so forth....

No one want a screen doesn't measure well. A screen measure well has a good image.
A speaker measures well, sound well. In audio, a lot of people deny the importance of the quality of measurements.

The psychology has nothing to do with the engineering but the marketing.

 

[antcollinet]

Illogical, captain. With no data beforehand about where you need to finish, you can't say where you need to start. Could be a flat speaker might need way more EQ than a random non-flat speaker, in which case the non-flat speaker offers "maximum room".

I understand the lay intuition toward flatness, but with EQ on the menu, flatness is meaningless.

You do seem particularly dense. Perhaps obtuse is the word.

Point is until you get the speakers in the room and measure in the room, in the position, you dont know where you will be with the room response. So the only thing you can do in advance is choose flat. I'm out now, I suspect you are arguing for the sake of it, rather than actually for a point you really believe in.

 

[DanielT]

No one want a screen doesn't measure well. A screen measure well has a good image.
A speaker measures well, sound well. In audio, a lot of people deny the importance of the quality of measurements.

The psychology has nothing to do with the engineering but the marketing.

Well it depends on what you mean by measures well.

Take two different systems of speakers. Both have equally low distortion.

Both do not measure equally in terms of the frequency curve. Is one more colored than the other then? These two different systems are tested in a normal living room, listening room. Which sounds best? It probably depends on what others have mentioned in this thread, room acoustics
,setting of house curves for example. In the end preferences in terms of sound.

If you like to really turn up the volume. If you love SPL. Let it be said that at work you sit and mix and listen with a couple of small exact studio monitors, but at home in your living room you have two to see. see picture ... well ... Those together with some top PA speakers for example would probably not measure as well as a pair of exact studio monitors, but are you high SPL dependent so ...?

Or instead a pair of high SPL PA stack into your listening room together with bass modules, see picture.
It will probably sound loud ass hell at the same time there is the risk of a slightly (understated) colored sound.

https://mh-speaker.en.made-in-china...ss-Speaker-Front-18-Rear-15-Audio-System.html

Edit:

Also, not everything is black or white. Here is Amir's review of the JBL SRX835P "PA / DJ" powered speaker

https://audiosciencereview.com/forum/index.php?threads/jbl-srx835p-reviewed-powered-monitor.25300/

"Measurements show a clearly imperfect speaker here. ...

I think what is going on here is that smaller speakers must be increasing in distortion rapidly, causing one to not turn them up. And if you do not turn them up, then you do not enjoy the full dynamics and details in your music ....

It was an interesting sensation to have frequencies above bass cause physical vibrations in you body! Yes, listening at these levels for too long is not good for your hearing. But a few minutes of it put a smile on my face that I can not yet wipe.

You have not lived until you experience a speaker with such high dynamics. "

When I think about it . It would be fun to have two different systems. One based on something surgically accurate
like Genelec or Neumann. The second a SPL monster. Switch between them. To one's liking and taste. Depending on what you want to listen to.
Right now I have neither the space nor the money for it. In addition, I live in an apartment. The neighbors would go insane. Wasted money ...... maybe wasted ears too.

 

[Frgirard]

The expression good measures is binary. It is or it is not.
The way to measure and Good measures are define by an industrial consensus.
Anechoic flat response (+ or - 2 dB). Low distortion and homogeneous directivity.

All speakers are colored, it's the brand signature.
Good measurements?you can see Genelec, Neumann...

 

[DanielT]

The expression good measures is binary. It is or it is not.
The way to measure and Good measures are define by an industrial consensus.
Anechoic flat response (+ or - 2 dB). Low distortion and homogeneous directivity.

All speakers are colored, it's the brand signature.
Good measurements?you can see Genelec, Neumann...

At what effect? It is a bit of a difference to test a boat in a quiet harbor compared with out at sea when it is windy.
The same metaphor also applies to amplifiers as well as speakers.

Speaking of homogeneous directivity. A pair of omni-directional Duevel loudspeakers in the video.

Maybe not the most ideal compared to a pair of "traditional" speakers with homogeneous directivity if you have the opportunity to place yourself in sweetspoot but if you move a lot in the listening room?

I've never heard a pair of omni-directional loudspeakers, I should add.

 

[Frgirard]

At what effect? It is a bit of a difference to test a boat in a quiet harbor compared with out at sea when it is windy.
The same metaphor also applies to amplifiers as well as speakers.

Speaking of homogeneous directivity. A pair of omni-directional Duevel loudspeakers in the video.

Maybe not the most ideal compared to a pair of "traditional" speakers with homogeneous directivity if you have the opportunity to place yourself in sweetspoot but if you move a lot in the listening room?

I've never heard a pair of omni-directional loudspeakers, I should add.

You mixe room and speakers. Refer to Asr's tests. Did you know?

For boats, planes or speakers , in the industry we use simulation software. It works fine and you can control all the parameters and test them.

 

[DanielT]

Rooms and speakers are connected. Correct.

Measurements are interesting to take part in. How they are made. For example, is it measured with a bandwidth up to 70 kHz when testing class d amplifiers?

Is the distortion measured at when the effect increases?
Attached pictures test of a class d amplifier, Aiyima TPA 3255 with power supply 36 V 5.56 A

 

[Frgirard]

Rooms and speakers are connected. Correct.

Measurements are interesting to take part in. How they are made. For example, is it measured with a bandwidth up to 70 kHz when testing class d amplifiers?

Is the distortion measured at when the effect increases?
Attached pictures test of a class d amplifier, Aiyima TPA 3255 with power supply 36 V 5.56 A

The speaker engineering does not take care of the room.
The speaker engineering know one room: the anechoic room.

Why you introduce electronic at 70 kHz? And why not the temperature to make good chips...

 

[DanielT]

Regarding measurements. From another forum where measurements around this are discussed:

https://www.stereophile.com/content/nad-c-298-power-amplifier-measurements

Written in Swedish, ran through quotes below via google translate, maybe some details and nuances have been lost:

"Note that the measurements on this Class-D device are performed with the low-pass filter (for = 25 kHz) Audio Precision AUX-0025 connected between the output of the output stage and the input of the analyzer, so all THD measurements are irrelevant above about 6 kHz.
It surprises me that Atkinson does not limit the graphs to just this.

THD is a measure of the nonlinearities of the step - quadratic, cubic and higher order terms in the transfer characteristics. The same terms lead to intermodulation distortion. If you want a correct picture of distortion, the bandwidth when measuring THD up to 20 kHz must be at least 80 kHz, which means that harmonics up to and including order four are included.

Nad C298 shows high THD at treble frequencies above 10 kHz despite the numbers being difficult to beautify due to the low-pass filter used in the THD measurement.
The reason for the measurement procedure is that the analyzer from Audio Precision that Atkinson uses does not handle the copious amount of high-frequency debris typical Class-D throws out. What I object to is that the THD data is still reported up to 20 kHz, even though the data is not relevant above 6-7 kHz. It is also obvious that very many readers of tests have not reflected on this, which of course is fully understandable."

....

If you want clarification about it then unfortunately I do not have enough technical skills. But maybe you can bring it up with the person who wrote it? If you are curious or want clarification that is.


https://www.faktiskt.io/phpBB3/viewtopic.php?f=9&t=72403

Or here. He manufactures amplifiers:
https://www.facebook.com/sybariteaudio.se/

Edit:

What was quoted above should probably be given an indication in the graph that Sterophile publishes, see picture. Those with technical knowledge are welcome to clarify this. I'm curious about how to interpret that.

Fig.2 NAD C 298, stereo mode, small-signal, 10kHz squarewave into 8 ohms.

 

[A Surfer]

I'm fine with either style of system, although I start with the recording. I can no longer even listen to old, poor, thin sounding recordings. I couldn't care less how good the music is, it will sound atrocious to my ears. Most of the Beatles and much of Led Zeppelin's catalogue falls into that camp (but I can tolerate some Zeppelin albums such as In Through The Out Doors).

You can't unhear a more modern well done recording where there is actually meat on the bones so to speak. While I do not prefer highly compressed recordings, I also feel that there still exists the potential that if done well, they can still sound very good. There is skill involved in such matters after all. We now have recording and mixing/mastering engineers who have been working with compression for a long time, as well as new tools so compression has also advanced.

I find it smacks of elitism when people imply the only correct way to enjoy the music is with as neutral a playback chain as possible. Hogwash I say. There is no right or wrong, or more or less correct when it comes to the subjective enjoyment of the experience. I agree that the most neutral system will allow the original recording intent to be the most clear, but who is so full of themselves that they actually think they have the right to dictate how another person should enjoy their personal music experience? If somebody likes the added harmonic distortion of tubes, or the effects a horn based speaker imparts on the music great for them. I don't have to build my system that way.