Some comments from Floyd Toole about room curve targets, room EQ and more

[Fledermaus]

Harman Target doesn't apply to my room, I thought.

Maybe you could also try this : https://www.audiosciencereview.com/...mmm-method-and-website-by-jean-luc-ohl.22407/
MMM works fine IME for in situ fine-tuning, particularly in the bass department - faster and much more convenient than multi-spot averaging.

 

[Tim Link]

Maybe you could also try this : https://www.audiosciencereview.com/...mmm-method-and-website-by-jean-luc-ohl.22407/
MMM works fine IME for in situ fine-tuning, particularly in the bass department - faster and much more convenient than multi-spot averaging.

I'll definitely try that. Thanks!

 

[DavidMcRoy]

Not sure I even understood what you meant.

How exactly have you been making listening window measurements, and how have you been applying a target curve average (also not sure what applying an average means) to that?

I make them by placing the microphone in the listening window. I have then applied EQ via DSP to approximate a “target curve” for evaluation. You may ignore the term “average.”

 

[sigbergaudio]

I make them by placing the microphone in the listening window. I have then applied EQ via DSP to approximate a “target curve” for evaluation. You may ignore the term “average.”

At the listening position? Or do you mean listening position when you say listening window?

Listening window measurements typically refers to an average of a number of anechoic measurements (+/- 30 horisontal and +/-10 vertical). From that definition it's still not perfectly clear what you are doing.

Applying dsp to create a target curve is not really recommended in the first place if this means forcing the speaker into a different in-room curve than it has naturally. It should ideally be used just to even out the response, and if needed you may use it to adjust the tonal balance by tilting the bottom end up or down.

 

[Tim Link]

I'll definitely try that. Thanks!

I tried MMM over the weekend, and it showed consistent peaks at 63 Hz and 400 Hz. Taking that 400 Hz down a few dB seems to be an improvement so far. I thought I was hearing too much emphasis around there but it wasn't clearly obvious with the single position measurement. Funny thing, I already had a PEQ set for 63 Hz but somehow it had gotten turned off.

 

[Tim Link]

Maybe you could also try this : https://www.audiosciencereview.com/...mmm-method-and-website-by-jean-luc-ohl.22407/
MMM works fine IME for in situ fine-tuning, particularly in the bass department - faster and much more convenient than multi-spot averaging.

I've had some time to listen since I tried the mmm method and so far I'm still liking it. No desire to change things back, so that's a good sign.

 

[mps]

Now, if you measure such a curve or something very close to it, and your speakers are conventional forward firing designs, it means that you probably have chosen well.

Accepting Dr. Toole’s statement, what target curves should we aim for with non-forward firing speakers? In particular, my ESL 57s with Sunfire subwoofers. Is the Harman curve still a good target (but then why qualify these statement above) or are there other research-backed targets for dipoles?

 

[nathan]

Accepting Dr. Toole’s statement, what target curves should we aim for with non-forward firing speakers? In particular, my ESL 57s with Sunfire subwoofers. Is the Harman curve still a good target (but then why qualify these statement above) or are there other research-backed targets for dipoles?

Flat anechoic on axis response is likely still the goal but the off axis response will behave differently and produce a different looking curve….and remember it is not a TARGET. It is a RESULT…. it is how a “good speaker” in the way defined by user preference tests measures in a standardized room.

I haven't seen data that shows what the ideal in room response ends up looking like for dipoles. The CEA2034 data for dipole speakers does look interesting. See the measurements in reviews on Erin's Audio Corner in addition to some here.

 

[nathan]

(I'd assume consistent off axis response ALSO matters with a BIPOLE but I am not sure what that look like given how quickly they have phase cancellation off axis. I would love to learn more.)