Room distortions vs SINAD

[CapMan]

Rooms introduce distortion and so does equipment. As we drive towards vanishingly low SINAD in equipment I wondered at what point the room distortion completely masks this improvement.

I only ask because (as per my other posts) fixing my room with treatment has had way more impact than my last amp change.

Interested to hear the view of others on this.

 

[tuga]

Which types of distortion are you referring to?

I suggest that you ignore SINAD, it's a worthless metric...

 

[Vive la dissonance!]

I suggest that you ignore SINAD, it's a worthless metric...

It's certainly not worthless.

 

[HarmonicTHD]

Rooms introduce distortion and so does equipment. As we drive towards vanishingly low SINAD in equipment I wondered at what point the room distortion completely masks this improvement.

I only ask because (as per my other posts) fixing my room with treatment has had way more impact than my last amp change.

Interested to hear the view of others on this.

Agree. Money is always better spent on better speakers, room EQ or even room treatment. Probably even in this order.

 

[amirm]

Room transformation is linear so doesn't create distortion (new frequencies). Of course if you have rattles and such that get activated, then that would be distortion but I am assuming you are not asking about that.

 

[CapMan]

Room transformation is linear so doesn't create distortion (new frequencies). Of course if you have rattles and such that get activated, then that would be distortion but I am assuming you are not asking about that.

Probably a poorly framed question. To try and restate it - the listening room forms part of the end to end system of reproduced sound along with electronics and speakers.

Intuitively the distortion the room causes to the recorded sound (and for simplicity I define this as comb filters, modes etc) surpasses the distortion caused by the electronics at some point.

I don’t known if this rephrasing helps - I wondered when this tipping point occurred. When do room ‘distortions’ dominate what we hear in the end to end system.

Does this help ?

As a thought experiment -
Which would sound better - an amp with class leading SINAD in an untreated room (with poor acoustics) or an amp with mediocre SINAD in a well treated room (with good acoustics) .

 

[HarmonicTHD]

Probably a poorly framed question. To try and restate it - the listening room forms part of the end to end system of reproduced sound along with electronics and speakers.

Intuitively the distortion the room causes to the recorded sound (and for simplicity I define this as comb filters, modes etc) surpasses the distortion caused by the electronics at some point.

I don’t known if this rephrasing helps - I wondered when this tipping point occurred. When do room ‘distortions’ dominate what we hear in the end to end system.

Does this help ?

As a thought experiment -
Which would sound better - an amp with class leading SINAD in an untreated room (with poor acoustics) or an amp with mediocre SINAD in a well treated room (with good acoustics) .

The latter

 

[tuga]

It's certainly not worthless.

It's good for marketing purposes...

 

[Cars-N-Cans]

Probably a poorly framed question. To try and restate it - the listening room forms part of the end to end system of reproduced sound along with electronics and speakers.

Intuitively the distortion the room causes to the recorded sound (and for simplicity I define this as comb filters, modes etc) surpasses the distortion caused by the electronics at some point.

I don’t known if this rephrasing helps - I wondered when this tipping point occurred. When do room ‘distortions’ dominate what we hear in the end to end system.

Rooms do not generate any sort of distortion generally unless there is something weird going on. Distortion is something that results from system nonlinearities, e.g. an amplifier that clips the tops off of a sine wave due to excessive drive. That will generate copious harmonics, and thus distortion. The room merely alters the frequency and time domain responses with its various modes, comb filtering, reverberation, etc.

For what it’s worth, typical quiet room has a background noise of about 30-35 dB on the A-weighted scale. If you are listening to music at 80 dB average, that puts your S/N ratio at about 50 dB. In reality since the audio is not stochastic (random), some details can be heard below the noise a bit, but still I would say once the SINAD of the equipment gets above 65 dB it’s no longer really a factor. With that in mind, though, the speakers will be the dominant source of distortion. If the electronics are good, then they can be regarded as “perfect” since they will be at least an order of magnitude better than most speakers. As such speakers and room treatment/corrections should be your primary focus.

 

[CapMan]

Rooms do not generate any sort of distortion generally unless there is something weird going on. Distortion is something that results from system nonlinearities, e.g. an amplifier that clips the tops off of a sine wave due to excessive drive. That will generate copious harmonics, and thus distortion. The room merely alters the frequency and time domain responses with its various modes, comb filtering, reverberation, etc.

For what it’s worth, typical quiet room has a background noise of about 30-35 dB on the A-weighted scale. If you are listening to music at 80 dB average, that puts your S/N ratio at about 50 dB. In reality since the audio is not stochastic (random), some details can be heard below the noise a bit, but still I would say once the SINAD of the equipment gets above 65 dB it’s no longer really a factor. With that in mind, though, the speakers will be the dominant source of distortion. If the electronics are good, then they can be regarded as “perfect” since they will be at least an order of magnitude better than most speakers. As such speakers and room treatment/corrections should be your primary focus.

That’s really informative - many thanks. Sorry I took a few liberties in describing room acoustic issues like comb filters as ‘distortions’ .

 

[abdo123]

Around 1990 when solid state dominated the market.

 

[tuga]

Intuitively the distortion the room causes to the recorded sound (and for simplicity I define this as comb filters, modes etc) surpasses the distortion caused by the electronics at some point.

To the best of my knowledge, some types of room-generated acoustic interference may or may not mask some types of distortion produced at the transduction stage and upstream.
Low fidelity speakers will also allow one to identify some types of distortion upstream, I would expect it to be the same with the room.

My view is that you should address all types of audible distortion and all types of audible room-generated acoustic interference.
As such, I find your view somewhat wishful thinking. Made worse by using SINAD and the Speaker Preference Rating as guidance.

 

[Sokel]

What always made me think is that even if the FR in 1m acoustical measurement seems nice the distortions seems a little less over the spectrum in the normal 2.8m of my listening position.I don't know why,maybe it's the size and distance from the drivers that makes it seem smoother.
(Totally amateur observation,take it with a sea of salt )

 

[Cars-N-Cans]

My view is that you should address all types of audible distortion and all types of audible room-generated acoustic interference.

I would agree that in electronics we shouldn’t just slum it and just settle for whatever is “good enough” to sound good. It’s good practice and good engineering hygiene to address them. But it’s equally important to not loose sight of the forest as well from spending too much time gazing at the analog trees. An amp that has 0.005% distortion won’t sound any different than one with 0.05% distortion. Obviously if it’s a tube or oddball minimum discrete transistor “purist” design with 5% distortion that is a different matter entirely. I say this not to contradict you, but rather to mention that those not as well versed in the parlance of audio equipment design may not have a feel for the relative impacts of all the forms of distortion from the various elements in the system. We see this all the time with reviewers gushing over one amp or another. But if the listening was done blind they would have no preference, as we know by now.

Just something to keep in mind so people can put their money/effort where it matters most. As far as I’m concerned, with electronics just ensure you buy equipment from competent manufacturers that meets your desires for functionality/looks and speaker power requirements and you should be good to go on that front. Any distortion they have will likely be far below that which is detectable.

 

[Cars-N-Cans]

What always made me think is that even if the FR in 1m acoustical measurement seems nice the distortions seems a little less over the spectrum in the normal 2.8m of my listening position.I don't know why,maybe it's the size and distance from the drivers that makes it seem smoother.
(Totally amateur observation,take it with a sea of salt )

There are a lot of moving parts to the audibility of distortion, for what it’s worth. I would use the measurements more as a metric of the speakers capabilities at higher SPLs and ability to not color the sound or fall apart at high SPLs. If you like listening to Russian hardbass at absurd volumes a speaker that has a tiny woofer and close to 100% distortion at low frequencies would not be a good choice It will probably just make lots of weird noises (edit: One could also argue that is already weird noises to begin with, even if I do kind of like it). Conversely if a tweeter has a nasty resonance with lots of THD at all SPL levels in the measurement at that one location it will add odd coloration and potentially unpleasant brightness or have a piercing quality to it when those said resonances are activated by the audio content.

 

[Sokel]

There are a lot of moving parts to the audibility of distortion, for what it’s worth. I would use the measurements more as a metric of the speakers capabilities at higher SPLs and ability to not color the sound or fall apart at high SPLs. If you like listening to Russian hardbass at absurd volumes a speaker that has a tiny woofer and close to 100% distortion at low frequencies would not be a good choice It will probably just make lots of weird noises. Conversely if a tweeter has a nasty resonance with lots of THD at all SPL levels in the measurement at that one location it will add odd coloration and potentially unpleasant brightness or have a piercing quality to it when those said resonances are activated by the audio content.

If hard to ask from something to perform above it's capabillities,you're absolutely right.

Here is an example I have already measured.

Mid

woofer

Both 2.8 meters.

 

[Cars-N-Cans]

If hard to ask from something to perform above it's capabillities,you're absolutely right.

Here is an example I have already measured.

Its worth mentioning that the radiation pattern of the distortion products will be impacted by the same things that normally impact the content of the speaker with respect to frequency. If the tweeter beams at high frequencies, the higher frequency distortion products will be more directional, too. And I don't know if there is rule of thumb to quantify how the distortion of the woofer propagates if its cone breakup or something, but some of those products are likely to be directional as well in some form. There is also things like masking by the audio content, etc. Lots of things that will impact your perception of it, not to mention the measurement method itself could also potentially have an impact on the results. I have not seen it yet, Amir just did a video on this same topic. Might be interesting to see if you have not already done so. In my humble opinion the 1m is a good thing to stick with since we are more interested in the results in a comparative and not absolute sense.

And its worth noting your microphone can also contribute to the distortion results:

 

[Cote Dazur]

The room merely alters the frequency and time domain responses with its various modes, comb filtering, reverberation, etc.

Room transformation is linear so doesn't create distortion

I am not sure to follow those 2 comments:

Merely alters? The room and speaker placement/seat positioning is the biggest factor of sound change when listening to music, in my experience, there is nothing subtle about it, sound from different device, merely affect anything in comparison. Speakers having more difference within the device.

Room does not create distortion? I must not understand distortion, how is something profoundly affecting the sound we hear, not distorting, ie changing, the sound?

 

[Cars-N-Cans]

I am not sure to follow those 2 comments:

Merely alters? The room and speaker placement/seat positioning is the biggest factor of sound change when listening to music, in my experience, there is nothing subtle about it, sound from different device, merely affect anything in comparison. Speakers having more difference within the device.

Room does not create distortion? I must not understand distortion, how is something profoundly affecting the sound we hear, not distorting, ie changing, the sound?

Distortion as most people here define it is the creation of new frequency components via the actions of nonlinearities in a system. Since air acts as an ideal gas at the SPL levels we are concerned with and the surfaces in your room basically either absorb or reflect the sound, no new frequencies are created. A pure 1 kHz tone stays a pure 1 kHz tone, but the modal distribution in the room will cause its intensity to vary as you move about in the room. Same with low frequencies. The room mode will, more or less, dictate the actual SPL that result with respect to frequency. The speaker is basically just there to excite it. Now one can validly argue that altering the frequency response or phase response is also "distortion", but that is overly broad and I think will just confuse people if they conflate them with distortion as would be defined in terms of the electronics, like THD, noise, etc. It makes more sense to segregate these since the mechanisms are so different. But, I would not consider room modes to be distortion since a 120 Hz tone is still 120 Hz on an RTA, just the dB or dBFS varies with the mic position. However if you pass 50 Hz at 10V through a tiny speaker, it will make 50Hz, 100 Hz, 150 Hz, and a bunch of other harmonics and noise as well from the driver grossly running out of travel. That is distortion since we added new frequencies to what we were trying to reproduce, and it can also produce IMD as well since it will modulate any other frequencies you play through it at the same time.

 

[Cars-N-Cans]

I am not sure to follow those 2 comments:

Merely alters? The room and speaker placement/seat positioning is the biggest factor of sound change when listening to music, in my experience, there is nothing subtle about it, sound from different device, merely affect anything in comparison. Speakers having more difference within the device.

Room does not create distortion? I must not understand distortion, how is something profoundly affecting the sound we hear, not distorting, ie changing, the sound?

Additionally, we cannot remove the distortion produced by nonlinearities, but the "distortion" from the room effects can be ameliorated by speaker placement, PEQ and room treatments. However if speaker is producing 10-15% THD, the only recourse is to turn the volume down. There is no corrective action we can do at that output level to correct it other than getting a new speaker. In other words, it has altered the fundamental spectral content with respect to frequency components. New stuff is there that wasn't before. The room wont do that.