Best Room Response

[QMuse]

I agree that the 400-800Hz range may benefit from correction.

I think where we disagree is in terms of what should be corrected here.

I'm arguing that, although the influence of the room in this region can be significant, it is inevitably (due to the wavelengths involved) highly position-dependent. In other words, the impact of the room will differ significantly with changes in position of as little as 10cm (in the case of 800Hz) and 20cm (in the case of 400Hz).

Therefore, correction in this region, if any, should be speaker correction, not room correction.

If you make a room correction at 400Hz, it can only possibly (physically) be valid within a listening position range of around +/- 20cm. For 800Hz, that range shrinks to around +/- 10cm.

This is too narrow for listeners who prefer not to have their head in a vice (so to speak).

Ok, we stated where we agree and where we disagree. I also made some effort to back up my opinion with measurements, so I suggest you try to do the same and then we'll see how the discussion would went from there.

 

[Willem]

I want to add that a smoother response in the region around the transition frequency can be helped by having a larger room lowering the transition frequency and using multiple subs (if they still have some output there). Good low frequency response in a small room is hard to achieve. My personal preference in small rooms is to forget about deep bass and just have a sealed bookshelf speaker with clean bass.

 

[andreasmaaan]

Ok, we stated where we agree and where we disagree. I also made some effort to back up my opinion with measurements, so I suggest you try to do the same and then we'll see how the discussion would went from there.

Haha ok. Firstly, I'm not somewhere where I have/can take measurements right now.

Secondly, I don't understand why you would insist on seeing measurements to "back up" the physical fact that 800Hz has a wavelength of around 42cm and that therefore room effects at this frequency will necessarily have a 21cm spacing between each peak and dip

 

[QMuse]

Secondly, I don't understand why you would insist on seeing measurements to "back up" the physical fact that 800Hz has a wavelength of around 42cm and that therefore room effects at this frequency will necessarily have a 21cm spacing between each peak and dip

You'll see when you try to measure it. Btw, you'll be measuring room modes in the 400-800Hz range, not wavelength.

 

[andreasmaaan]

You'll see when you try to measure it. Btw, you'll be measuring room modes in the 400-800Hz range, not wavelength.

Yes of course, but the period of the mode is defined by the wavelength.

And I have taken many measurements of this kind before, which is why I agree with you that "mild" room modes are present in this frequency range. The discussion as I understood it is not about whether or not they exist (we both agree they do), but rather about whether or not they should be corrected.

 

[QMuse]

Yes of course, but the period of the mode is defined by the wavelength.

And I have taken many measurements of this kind before, which is why I agree with you that "mild" room modes are present in this frequency range. The discussion as I understood it is not about whether or not they exist (we both agree they do), but rather about whether or not they should be corrected.

Yes, that is indeed our discussion. And how to differentiate them between direct speker response. It is not easy to measure direct sound response in the room in 400-800Hz range because of reflections.

 

[andreasmaaan]

Yes, that is indeed our discussion. And how to differentiate them between direct speker response. It is not easy to measure direct sound response in the room in 400-800Hz range because of reflections.

Yes, and of course all I've said above is based on the assumption that you have accurate anechoic (or quasi-anechoic) data on the speaker.

 

[QMuse]

Yes, and of course all I've said above is based on the assumption that you have accurate anechoic (or quasi-anechoic) data on the speaker.

Well, anechoic you usually don't have and quasi anechoic is problematic to measure in the 400-800Hz region, especially with floorstanders.

 

[andreasmaaan]

Well, anechoic you usually don't have and quasi anechoic is problematic to measure in the 400-800Hz region, especially with floorstanders.

I generally use a version of the groundplane technique below around 1 kHz. Obviously this is not a measurement that can be done in-room.

Also, many people will be using speakers for which they have measurements from either the manufacturer or a third-party source (e.g. ASR).

 

[QMuse]

I generally use a version of the groundplane technique below around 1 kHz. Obviously this is not a measurement that can be done in-room.

Also, many people will be using speakers for which they have measurements from either the manufacturer or a third-party source (e.g. ASR).

Oh c'mon, Spinorama measurements exist only for handfull of speakers and ASR doesn't really measure floorstanders. (pun intended!)

 

[andreasmaaan]

Oh c'mon, Spinorama measurements exist only for handfull of speakers and ASR doesn't really measure floorstanders. (pun intended!)

https://speakerdata2034.blogspot.com/2019/03/all-spinorama-data-index.html

https://www.soundstagenetwork.com/index.php?option=com_content&view=category&id=77&Itemid=153

^^There are reliable measurements of perhaps 200 popular loudspeakers of all sizes and price ranges, not including those done by ASR.

And what's to stop people taking groundplane measurements?

Anyway, I agree that, if you are not able to take or obtain accurate measurements of your speakers, playing with the in-room response between 400Hz and 800Hz may have a positive impact. But it's not the optimal way to go about it, is my point.

 

[QMuse]

https://speakerdata2034.blogspot.com/2019/03/all-spinorama-data-index.html

https://www.soundstagenetwork.com/index.php?option=com_content&view=category&id=77&Itemid=153

^^There are reliable measurements of perhaps 200 popular loudspeakers of all sizes and price ranges, not including those done by ASR.

And what's to stop people taking groundplane measurements?

Anyway, I agree that, if you are not able to take or obtain accurate measurements of your speakers, playing with the in-room response between 400Hz and 800Hz may have a positive impact. But it's not the optimal way to go about it, is my point.

Sure, I'm aware of those sources, but it's really a minor number compared to the number of different speaker's in use and on the market.

And what's to stop people taking groundplane measurements?

Length of the cables? Street noise? ..?

Anyway, I agree that, if you are not able to take or obtain accurate measurements of your speakers, playing with the in-room response between 400Hz and 800Hz may have a positive impact. But it's not the optimal way to go about it, is my point.

How about measuring midwoofer's nearfield response? To me it seems safe to assume port and tweeter doesn't influence that range in any meanigfull way.

 

[andreasmaaan]

How about measuring midwoofer's nearfield response? To me it seems safe to assume port and tweeter doesn't influence that range in any meanigfull way.

Also valid. The main issue is that nearfield measurements won't take into account the effects of the baffle (can be modelled/estimated/compensated for of course though).

 

[QMuse]

Also valid. The main issue is that nearfield measurements won't take into account the effects of the baffle (can be modelled/estimated/compensated for of course though).

True. Luckilly for me I believe my design is of "infinite baffle".

 

[andreasmaaan]

True. Luckilly for me I believe my design is of "infinite baffle".

Tell me more?

 

[QMuse]

Tell me more?

Twin pipe quarterwave transmission line, 2 midwoofers, 1 firing forward the other upwards.

 

[QMuse]

Also valid. The main issue is that nearfield measurements won't take into account the effects of the baffle (can be modelled/estimated/compensated for of course though).

Let's think of a simple experiment. My speakers, as most modern speakers, have pretty linear response in the 400-800Hz region also verified by 3rd party measurements. Now, if I was to introduce some mild resonance in their response via DSP, say +2dB with Q=4 centered at 550Hz, you're saying I wouldn't be able to measure the difference between original and modified response unless I use groundplane tecnique or nierfield midwoofer measurement?

Surely with single sweep I might have difficulties to catch something like that, but what about detailed MMM over my sitting position?
Or you think I would be swiping through peaks and dips of that resonance so it would average to 0?

 

[andreasmaaan]

Let's think of a simple experiment. My speakers, as most modern speakers, have pretty linear response in the 400-800Hz region also verified by 3rd party measurements. Now, if I was to introduce some mild resonance in their response via DSP, say +2dB with Q=4 centered at 550Hz, you're saying I wouldn't be able to measure the difference between original and modified response unless I use groundplane tecnique or nierfield midwoofer measurement?

Surely with single sweep I might have difficulties to catch something like that, but what about detailed MMM over my sitting position?
Or you think I would be swiping through peaks and dips of that resonance so it would average to 0?

Sure, you might be able to measure it that way. You'd likely be able to hear it too.

But (assuming the speakers had linear anechoic response before EQ) would it, after adding the resonance, sound better?

 

[QMuse]

Sure, you might be able to measure it that way. You'd likely be able to hear it too.

But (assuming the speakers had linear anechoic response before EQ) would it, after adding the resonance, sound better?

Oh, no - my idea was to simulate that resonance exists just to see if i would be able to measure it in a precise enough manner so I would be able to correct it if a resonance like that would really be there.

 

[andreasmaaan]

Oh, no - my idea was to simulate that resonance exists just to see if i would be able to measure it in a precise enough manner so I would be able to correct it if a resonance like that would really be there.

Ok, so it seems that we actually agree that, in the 400-800Hz range, although the room may have a significant influence on the steady-state response, it is only the speaker's (anechoic) response that should be corrected.

I understood you to be arguing earlier that room correction should be implemented in the 400-800Hz range. As I understand it now, though, what you're actually saying is that only the speaker should be corrected in this range (with which I agree) - but that one way to indirectly measure the speaker is by averaging out a large number of in-room measurements.

Did I understand correctly?