Flat speaker frequency response in room = bad?

[lovinthehd]

All the tweaking is done at the crossovers not any external Eq.

In your system I assume is what you mean?

 

[TLS Guy]

In your system I assume is what you mean?

Correct, yes.

The only significant change when I moved from our lake home to this new room, was changing the BSC. This room required a significant increase in Baffle Step compensation. This is not surprising as BSC required is highly dependent on boundary effects. As more active speakers enter the market this is a feature that could become commonplace. I regard variable BSC as an important game changing feature of this system. JBL copied it on a high end speaker, but you had to pay the airfare for a technician to fly out to make the adjustment! So that went over like the proverbial lead balloon. However, this would be easily added to any active speakers, and setting it would be well within the capability of most owners.

 

[dlaloum]

Well.... your ears will seperate direct from reflected sound, but the standard continuous signal measurement of frequency response won't.

So what your ears "want" to hear is a relatively flat frequency response.

If you take a speaker that measures "flat" in an anechoic measurement (ie with no reflected sound- just the speaker alone) - and you place it in a typical listening room / living space, your f/r measurement will then be altered by the various reflections reaching the microphone from all around... so what you are measuring is the direct sound + the reflected sound.

Any reflected sound that is sufficiently delayed, will be "discounted" by our brains as part of the primary signal - and we then interpret it as ambiance/soundstage... but it does not factor into imaging/detail.
However the standard F/R method does not take this into account.

In your typical domestic space - much of the higher frequency reflections will be reduced by furnishings, curtains, etc... whereas the bass will typically bounce around more freely (unless the room is specially - and expensively - treated)....

So you will be adding more bass to the flat base signal, than highs - and the end result for a "good" (anechoically flat) speaker, in a real room, will tend towards the tilted bass high treble low profile.

If you have a flat F/R in room, then typically the base direct signal from the speaker is not flat but tilted towards the highs - hence some will characterise it as "harsh".

However there is always the matter of adjusting for taste! - it used to be quite common to see people using the analogue equalisers back in the day, to tune for a "smile" response with boosted bass and highs, and lower midrange - not my cuppa tea, but each to their own...

Key thing in optimising speaker to room setup, is to try to ensure that reflections are sufficiently delayed so that they don't impact on the direct sound and therefore imaging / detail.
That is why many like to place their speakers further from back and side walls... that delays the reflections! - however if the speaker is well damped, and has negligible output to the sides and rear, then this may be a non issue (or in the real world, a lesser issue).

Also it is nice to have an ample "soundstage" - ie: reflections are not a total negative - quite the contrary... reflections (when properly delayed as discussed earlier) provide the fabulous soundstage that really good setups have.

So we then have various speaker designs that vary in terms of directivity - ranging from fully "omnidirectional" (ie speakers with equal output in all directions- including back!) - which require careful placement and/or treatment to get the reflection timing right. (further from the wall, or absorbant / dispersing treatments used)
To highly directional designs... where if you aren't right "in line" with the speaker, you get a substantially lower output.

The worst of the output from speakers is uncontrolled resonances from the cabinet... well designed and damped speakers substantially minimise that...

My personal preference is wide dispersion speakers, and to place them with caution in the room - thereby getting the expansive soundstage from the reflections while still getting the imaging and detail.

End result "In Room" is still the slightly tilted F/R...

So yes, if your measurements are showing a flat F/R "in room" - it is cause for suspicion - either the speaker is anechoically "non flat" - or the room is highly reflective - typically neither of those is optimal.... (but, there is no accounting for tastes... some people like it like that)

 

[highfigh]

Hi Mark and thanks for putting in the effort to reply and provide relevant screenshots of measurements.



This part mainly repeats the same information mentioned in the video and provides similar measurements, but doesn't address my question - Why does it (flat response at the sitting position) sound bad?"


This part describes most rooms and speaker interaction behavior, but again doesn't explain the why.

- Am I supposed to blindly believe this statement?

I'm not trying to sound combative. I really wanted to understand this phenomenon/behavior.

If you can and want to save the current equalization file, do that and then, make the frequency response as flat as possible, manually. Use the smallest increments in the AVR's equalization menu and save that. If the AVR has buttons for personalized setup, use them to compare- that will allow faster toggling between the flat and sloped responses. Flat sounds, well, a bit 'flat'.

Human hearing doesn't need flat, it's much more selsitive in the area around 3100 Hz. In order to perceive bass, midrange and treble equally loud, the bass needs to be boosted. A lot. Treble perception depends on the listener- we don't all hear sound equally, mainly because of the loud sounds and noises we're exposed to during our lives, jobs and other experiences. If someone uses firearms, serves/served in the military or law enforcement, worked in loud places, listens to loud movies/music, rides a motorcycle/drives a loud car or with the windows open, their hearing HAS BEEN damaged, irreparably.

As far as the sound reflecting from the room's surfaces, it's because most materials don't reflect all of the sound. This fact is what's used in acoustics and the specification for materials called 'NRC' or Nose Reduction Coefficient shows how materials can be used for treatment, depending on how they attenuate at specific frequencies, and how much. This is shown in the specs for all acoustical materials and many others that are used often for interiior spaces.

The link shows the NRC for many materials and after room analysis, the choices are made, based on how the space will be used, at defined SPL and choices can make a place usable or unusable for music, speech, etc. In a highly reflective room, speech will be unintelligle because the reverberation and echo will continue for too long and in a very absorptive room that's used for speaking with mics & amplifiers, more power is needed but it can be a more pleasant experience. Loud sounds in a reflective room are annoying, unpleasant & very irritating and WILL lead to hearing damage.

Noise Reduction Coefficients (NRC) Ratings for Common Building Materials & Products

 

[shadyJ]

If you want a highly detailed answer to this question, read Floyd Toole's 'Sound Reproduction,' but there are some good answers above in this thread. To put it succinctly, higher frequencies are more readily absorbed than lower frequencies in typical rooms. Higher frequencies do not reflect as well because a lot more stuff absorbs them, whereas it takes some serious acoustic treatments to absorb low frequencies, and even bass traps don't work well for subwoofer-range frequencies. This can vary depending on the size of the room and the in-room materials, but you will almost always have a downward-sloping in-room response. So if you are listening to a sound source at anything but a very close distance, the sound spectra become attenuated as the frequencies rise.

The brain is really good at compensating for this; it apprehends and adapts to different acoustic situations very quickly, so this all sounds natural, even in greatly differing acoustic environments. So if you have something externally that tries to 'compensate' for the downward-sloping in-room response, it will sound unnatural to human hearing, with over-boosted high frequencies.

A couple of points I want to touch on related to all of this: one of the reasons why I am not a fan of auto-EQ systems like Audyssey is because each room has a different profile of absorption, but these auto-EQ systems have a single target curve that it wants to force on every acoustic situation. It can often degrade the sound outside of the bass range. The best solution is to just get a neutral speaker and restrict auto-EQ systems to correction below 500Hz or so, so it just deals with low-frequency room modes.

For the same reason, I am not crazy about the 'estimated in-room response' curves that we sometimes see. It might serve as a good example of what a speaker might do in a certain type of room, but that will not be that speaker's sound profile in a non-generic room, i.e., your room. Different rooms have different curves, and your hearing will be able to detect what sounds natural in each of them.

 

[TLS Guy]

Well.... your ears will seperate direct from reflected sound, but the standard continuous signal measurement of frequency response won't.

So what your ears "want" to hear is a relatively flat frequency response.

If you take a speaker that measures "flat" in an anechoic measurement (ie with no reflected sound- just the speaker alone) - and you place it in a typical listening room / living space, your f/r measurement will then be altered by the various reflections reaching the microphone from all around... so what you are measuring is the direct sound + the reflected sound.

Any reflected sound that is sufficiently delayed, will be "discounted" by our brains as part of the primary signal - and we then interpret it as ambiance/soundstage... but it does not factor into imaging/detail.
However the standard F/R method does not take this into account.

In your typical domestic space - much of the higher frequency reflections will be reduced by furnishings, curtains, etc... whereas the bass will typically bounce around more freely (unless the room is specially - and expensively - treated)....

So you will be adding more bass to the flat base signal, than highs - and the end result for a "good" (anechoically flat) speaker, in a real room, will tend towards the tilted bass high treble low profile.

If you have a flat F/R in room, then typically the base direct signal from the speaker is not flat but tilted towards the highs - hence some will characterise it as "harsh".

However there is always the matter of adjusting for taste! - it used to be quite common to see people using the analogue equalisers back in the day, to tune for a "smile" response with boosted bass and highs, and lower midrange - not my cuppa tea, but each to their own...

Key thing in optimising speaker to room setup, is to try to ensure that reflections are sufficiently delayed so that they don't impact on the direct sound and therefore imaging / detail.
That is why many like to place their speakers further from back and side walls... that delays the reflections! - however if the speaker is well damped, and has negligible output to the sides and rear, then this may be a non issue (or in the real world, a lesser issue).

Also it is nice to have an ample "soundstage" - ie: reflections are not a total negative - quite the contrary... reflections (when properly delayed as discussed earlier) provide the fabulous soundstage that really good setups have.

So we then have various speaker designs that vary in terms of directivity - ranging from fully "omnidirectional" (ie speakers with equal output in all directions- including back!) - which require careful placement and/or treatment to get the reflection timing right. (further from the wall, or absorbant / dispersing treatments used)
To highly directional designs... where if you aren't right "in line" with the speaker, you get a substantially lower output.

The worst of the output from speakers is uncontrolled resonances from the cabinet... well designed and damped speakers substantially minimise that...

My personal preference is wide dispersion speakers, and to place them with caution in the room - thereby getting the expansive soundstage from the reflections while still getting the imaging and detail.

End result "In Room" is still the slightly tilted F/R...

So yes, if your measurements are showing a flat F/R "in room" - it is cause for suspicion - either the speaker is anechoically "non flat" - or the room is highly reflective - typically neither of those is optimal.... (but, there is no accounting for tastes... some people like it like that)

No I don't use any "funny wall coverings". If the off axis response of a speaker closely matches the axis response, then sound will actually be good even in a room with a definite echo. The bass issue is very much related to baffle step compensation and having that factor adjustable is a major speaker advance in technology of speaker design.

The other issue is that my speakers are aperiodic transmission lines that have a much higher quality of bass than other designs.

 

[highfigh]

1)- Well.... your ears will seperate direct from reflected sound, but the standard continuous signal measurement of frequency response won't.

2)- So what your ears "want" to hear is a relatively flat frequency response.

3)- If you take a speaker that measures "flat" in an anechoic measurement (ie with no reflected sound- just the speaker alone) - and you place it in a typical listening room / living space, your f/r measurement will then be altered by the various reflections reaching the microphone from all around... so what you are measuring is the direct sound + the reflected sound.

4)- Any reflected sound that is sufficiently delayed, will be "discounted" by our brains as part of the primary signal - and we then interpret it as ambiance/soundstage... but it does not factor into imaging/detail.
However the standard F/R method does not take this into account.

So you will be adding more bass to the flat base signal, than highs - and the end result for a "good" (anechoically flat) speaker, in a real room, will tend towards the tilted bass high treble low profile.

5)- However there is always the matter of adjusting for taste! - it used to be quite common to see people using the analogue equalisers back in the day, to tune for a "smile" response with boosted bass and highs, and lower midrange - not my cuppa tea, but each to their own...

Also it is nice to have an ample "soundstage" - ie: reflections are not a total negative - quite the contrary... reflections (when properly delayed as discussed earlier) provide the fabulous soundstage that really good setups have.

6)- To highly directional designs... where if you aren't right "in line" with the speaker, you get a substantially lower output.

My personal preference is wide dispersion speakers, and to place them with caution in the room - thereby getting the expansive soundstage from the reflections while still getting the imaging and detail.

Point 1- our ears will distinguish direct vs reflected unless the delay is long- speech is very difficult to understand if the delay is longer than about .6 seconds in a small room, longer in larger rooms. The spec RT60 is one of the measurements used to ensure this in room designs. This is important in classrooms, lecture halls and other places where the message needs to reach the listeners. In homes, for music, it's still important but not often mentioned in most discussions among people who don't really need to know it.

Points 2 and 5- Our ears hear what they can but what they 'want' is up to that pair and the ability of that brain to reconcile what's coming in and how it's processed. If you search for 'Equal Loudness Contours', you'll see typical sensitivity to various frequencies- we're sensitive in the middle, not at the extremes and this explains the 'happy face' EQ curve.



Points 3 and 6- Flat signal going in resulting in flat response coming out in an anechoic space means the speaker is faithful to the input and it's more for comparing apples to apples, but dispersion has a lot to do with the listening experience because wide dispersion in a narrow, reflective room causes problems and isn't very tolerable but in a wider room, the problems aren't as apparent and may not actually be a problem at all. In a hard room, narrow dispersion can be used to reduce the needed SPL at the listening positions and avoid a direct line to the side walls.

Point 4- our hearing does what it can, but it can't discount all delayed sounds. The sounds that reach us first determine directivity, later sounds won't and this is called the Haas Effect, or precedence effect- two almost simultaneous sounds from different locations with short delay are perceived as a single sonic event, but the sound from the first source will cause us to perceive that as the location.