Why you need room EQ even with the speaker with perfect anechoic measurements

[Krunok]

Another way to accomplish that, though I don't know how different the results would be, is to make a few measurements over the area of interest.

Very true - averaged log sweeps made over the area of interest will produce identical result as MMM RTA method made over the same area. Usually 9+ sweeps are needed for teh same accuracy, but that depends on the size of the area, how violent room modes are etc.

 

[Krunok]

I would advise just to adjust minimum phase bass peaks, below a couple of hundred hertz.
Keith

Once you adjust the phase of passive crossover loudspeaker response becomes minimum phase at pretty much all frequencies. The room will mess with the phase at your LP so some more adjustment will be needed to assure minimum phase at LP. Room EQ systems like Acourate, DRC-FIR etc. do that automatically.

In my experience you can try to adjust the dips as well as along as you are willing to sacrifice some headroom. Btw, it is a common misconception that speakers/amp will suffer from that but the truth is the whole response is pushed down so only the headroom will suffer meaning you will loose the same ammount ammount of dB on top SPL capacity your system will be able to play as is the maximum boost you applied to elevate the dips.

 

[Ivanovich]

I know this is a popular view here but I do not agree.
Firstly I know one "hears through" a room and secondly the idea that a test signal could accurately measure the time dependant effect of exciting resonant modes (room modes) is actually laughable if you have done any work on noise and vibration (I am an old bloke but this used to be my job).
Certainly you can get a before and after plot of the microphone output for the excitation type you used for the measurement, at the place(s) you measured. Mine shows the bass peaks lopped off. That means it doesn't sound right for the room any more (to me).
I have been recording with microphones for 50 years and reducing resonance excitation by moving speakers for almost as long. I am happy with the results and absolutely do not agree with your point of view though accept you and many others here are of like mind.

Anybody who has done any recording knows that what you hear at a particular location and what the output of a microphone at the same location sounds like are quite different. You automatically compensate for room, luckily, if you didn't life would be very confusing as you moved about, even in the same room never mind from place to place.
Try it. You probably won't be quite so confident afterwards in the technique if your experience is like mine.

I don’t know what to make of this. I accept what you say you perceive, but I don’t understand it.

When I listen I am not moving around the room, so there is no “life” confusion for me. What I do hear is what the OP stipulated - peaks and nulls - some bass notes are accentuated and others are reduced in volume.

The brain does do its own smoothing as our bandwidth sensitivity is not X Hz wide but some fraction of an octave. At low frequencies there’s plenty of resolution in our hearing’s perception to distinguish between 1/2-step tones that are only a couple of Hz apart.

After locating speakers and LP as best the room decor allows, I reduce the peaks with PEQ and bass response is night and day better. (I actually use satellite mains, hi-pass cut-off at 60 Hz, and threes subs lo-passed at 100 Hz, 48dB/octave. The overlap fills in some SBIR and nulls issues from ceiling bounce and room modes.)

I’m curious about the OP’s use of Room EQ to deal with SBIR issues. Can you explain how you do that? I’ve been under the impression that you need room treatment and multiple subs to effectively address SBIR issues.

 

[Krunok]

I’m curious about the OP’s use of Room EQ to deal with SBIR issues. Can you explain how you do that? I’ve been under the impression that you need room treatment and multiple subs to effectively address SBIR issues.

Let's assume you have made accourate enough measurements over your LP area as a basis for your room EQ (either manual or automatic). In the response you just measured everything that matters for the sounnd is accounted for: anechoic response of your speakers, room modes, SBIR, reflections from the table in front of your listening sofa and every ohter thing that affected the yound we cannot even think of.

Now let's assume that once you measured the corrected response your room EQ generated it resulted in a reasonably close response to your desired target curve. If that was the case every factor that affected the sound (anechoic response of your speakers, room modes, SBIR, reflections from the table in front of your listening sofa and every ohter thing that affected the yound we cannot even think of) has been accounted for and the total result is your target curve. Including SBIR. Including everything else that affected the sound in your room and we forgot to mention it or didn't even think it would affect the sound. But everything was corrected as the final result what you mesured after correction is the only thing that counts.

 

[JohnPM]

Should probably remove "do" from the thread title.

 

[daftcombo]

If I remember correctly, Dr. Toole never recommended to use EQ to get rid of room modes. He only said that maybe it could help.

 

[Krunok]

Should probably remove "do" from the thread title.

Ok, I did.
Was that purely a grammar error or ..?

 

[daftcombo]

Ok, I did.
Was that purely a grammar error or ..?

With the "do" you asked a question. Without, you answer it.

You can also correct "anechioc".

 

[Krunok]

With the "do" you asked a question. Without, you answer it.

You can also correct "anechioc".

LOL, thanks!

I can also start drinking 2 coffies before starting a thread instead of 1.

 

[Krunok]

If I remember correctly, Dr. Toole never recommended to use EQ to get rid of room modes. He only said that maybe it could help.

I have no idea why people often think so as that is simply not true.

Loudspeakers and Rooms for Sound Reproduction
(Page 472)

9.2.2 Below the Transition Frequency
• The modal misbehavior of rooms can be treated by passive
or active acoustical methods. It is a problem over
which we have considerable control.
• This is very good news, since about 30% of our subjective
assessment of overall sound quality is associated
with bass performance.
• Optimum room dimensional ratios exist, but only if the
loudspeaker and listener locations are known in advance.
Generic “good” listening room ratios are a myth.
• Multiple subwoofers, with or without active signal processing,
provide options for achieving more uniformly
good bass at several listening locations in small rooms.
The need for equalization is reduced.
• Equalization is the final touch, and, properly done, it
works because low-frequency room resonances behave
as minimum-phase systems.

Additionaly, Toole is in favor of applying EQ to resonances above the TF as well:

You can apply EQ to non-resonant peaks and gain some benefits, but you have to do it carefully. Here you can find my opinion on that.

 

[spacevector]

Sure, it is well explained and demonstrated here.

This is quite nice. Makes so much easier/quicker to get the data for bass EQ.

Are there any disadvantages of this method compared to making multiple log sweep measurements and averaging them?

Also, can you please help me configure REW for this? I have the following specific questions:

• Is Pink noise to be used for this measurement? The video says so but my understanding is that the spectrum is inversely proportional to frequency - low frequencies will have more energy in the pink noise compared to high frequencies. Is white noise more suitable for this measurement?
• Settings for the RTA window itself. Any guidance for what these should be set to prior to starting the measurement?

 

[raistlin65]

Well, subwoofer enthusiasts will say you are only half right. You need EQ and well-placed dual subs to tame the bass below 80hz

 

[levimax]

Sure, it is well explained and demonstrated here.

In my limited experience with REW I consistently get almost exactly the same results using "multiple positions averaged together" as I do with a single measurement and ERB smoothing. Are there situations where multiple measurements are going to provide meaningfully better information or is ERB smoothing good enough?

 

[vavan]

Is Pink noise to be used for this measurement?

pink periodic noise

Settings for the RTA window itself. Any guidance for what these should be set to prior to starting the measurement?

I believe one doesn't need "forever" in "averages" parameter, I tend to use 32 or even less. "Adjust RTA levels" probably might be useful

 

[spacevector]

pink periodic noise
I believe one doesn't need "forever" in "averages" parameter, I tend to use 32 or even less. "Adjust RTA levels" probably might be useful

Thanks!

I was able to generate this graph using the moving mic technique. Going to attempt comparison with log sweeps and then attempt some EQ. Looks like my small room has bunch of issues to 1kHz.

And here I show a simple comparison between moving mic measurement vs a single point log sweep at MLP:

 

[Pio2001]

I was able to generate this graph using the moving mic technique. Going to attempt comparison with log sweeps and then attempt some EQ. Looks like my small room has bunch of issues to 1kHz.

Nice peaks ! It would be interesting to see if REW can predict them with the room simulation module. It would tell you if the peaks come from horizontal or vertical resonances, and what would happen if you move the speakers or the listening position.

 

[Kvalsvoll]

No EQ can fix this. The loudspeaker needs to change, and then you need EQ - in the bass range, to get the best possible result.

If you start working with EQ on this, you can fix the 100hz peak quite well, then take down the lower peaks a little. Then you are left with too little bass, and the resonances are still present - if you did not EQ away all the bass, that is.

If the loudspeaker is changed into something that has a response without severe dips, smoother, more flat, the remaining resonances can be fixed using EQ.

This is a solved problem now. And the old-school full-range speaker is not part of that solution.

 

[Soniclife]

pink periodic

Thanks, tried this the other week and used white noise, very obviously wrong.
Is there an idiots guide to why to use pink?

In the mode setting what is the best octave setting to use for measuring?

 

[vavan]

Is there an idiots guide to why to use pink?

Periodic Noise (PN) sequences are ideally suited for use with spectrum and real time analysers (RTA's). They contain every frequency the analyser can resolve in a sequence length that matches the length of the analyser's FFT. Their great benefit is that they produce the desired spectrum shape without requiring any averaging or windowing, so the analyser display reacts much more rapidly to changes in the system than it would if testing with pink or white random noise, making them ideal for live adjustment of EQ filters. The PN sequences REW generates are optimised to have a crest factor (ratio of peak level to rms level) that does not exceed 6 dB for full range sequences, narrower sequences should have crest factors less than 6.5 dB. The CTA-2034 filtered noise has a crest factor of 12 dB, per the specification. Use Pink PN when measuring with an RTA or White PN with a Spectrum analyser

 

[FrantzM]

Hi

It is indeed possible to EQ the speakers for a hypothetical Listening position that shouldn't vary by a few inches or cm. It remains however, that the best solution is multiple subs followed by EQ'uing the subs. What those graphs show, is the room response: Any speaker with similar low frequency output shall present the exact same response in the low end give or take a few tenths of dB. To linearize the bass in this region, multiple subs are required, then EQ to smooth out the response or/and house curve.
I don't see this as a case for EQ rather more proof that multi subs should be used in any system regardless of the intrinsic low bass output of the mains. At this point, multi-sub is the best solution to the problem of good bass response in a given room.