DSP: EQ’ing Full Range or Not – Testing the Ears

[AJ Soundfield]

I'll say "corrected", because Dirac does much more than EQ.

I'll wait for Sonnie's speaker measurements.

If what you are suggesting about hearing acuity were true, we'd see an unnatural rise above 10kHz at the LP

No. At/off the speaker.
The presumption here is that you all weren't sitting 4-6' from the speakers. Been a while since I been there, but I recall sitting further back than that. It's impossible for HF to be the same at 4' and 9' +

 

[AJ Soundfield]

The point being... whatever Dirac did, it "improved" the sound at least slightly enough for preference, but it didn't kick the other option to the curb either.

Yep, you confirmed (b) better than (a). At least in your room/setup. You're using good speakers, so indeed the effect should be subtle.

We certainly aren't talking huge waves here, but preference does matter, even if it's minor. There are those that like to seek out every last drop of improvement.

Yep, the (c) folks.
Still looking forward to you "speaker" measurements and what was occurring there. Forgot to ask where you place mic vertically. For the Revels, centered on tweeter or between tweet/mid?

cheers

 

[Sonnie Parker]

Still looking forward to you "speaker" measurements and what was occurring there. Forgot to ask where you place mic vertically. For the Revels, centered on tweeter or between tweet/mid?

I've been keeping the mic at the same height as it is at the primary listening position when I do the 1 and 2 meter measurements, but I can place it anywhere you want me to for this measurement.

We are only just a little over 2 meters since I moved my chair a bit forward. The Revels were farther away with the final placement that we measured, maybe 2.5 meters.

 

[FargateOne]

If I may. This discussion is very interesting but, it must be said that your FR are quite perfect to begin with, so EQing almost perfect is obviously a question of nuances. I have not the chance to get that starting point. You will easily recognise the"signature" 2,5Hz 3 dB valley of B&W's

 

[Sonnie Parker]

If I may. This discussion is very interesting but, it must be said that your FR are quite perfect to begin with, so EQing almost perfect is obviously a question of nuances. I have not the chance to get that starting point. You will easily recognise the"signature" 2,5Hz 3 dB valley of B&W's

Very true! These are all relatively excellent speakers to start with... and then setup in what seems to be a fairly respectable dedicated room that fairs well for most speakers. The low-end needs some help, as most do, but the mid to top-end would be satisfactory for most listeners I would think. My problem is...(whether I can personally hear it or not is irrelevant)... I am now biased by others that I trust have excellent hearing and who believe it sounds slightly improved with full EQ, thus it causes me to prefer it as well.

 

[AJ Soundfield]

We are only just a little over 2 meters since I moved my chair a bit forward. The Revels were farther away with the final placement that we measured, maybe 2.5 meters.

Whoa, that's almost nearfield. NRC typically measures at 2m for multi-driver integration and full baffle effects in their anechoic chamber.
That certainly changes the perspective. You might be closer to (c) than (b) there. I would posit that to be a somewhat atypical setup, with a predominant direct/reflected field. You made it a bit easy on the auto EQ system. Didn't realize you sat that close, memory was saying biggish room. Knew you had done surfaces revisions based on some idiots advice, but not seating per se.

I've been keeping the mic at the same height as it is at the primary listening position when I do the 1 and 2 meter measurements, but I can place it anywhere you want me to for this measurement.

Consistency would be good, so keep whatever height that was. Since 2m is practically where the speaker was corrected from, somewhere between1-1.5m would be ok. Was trying to see what the EQ did to "speaker" response, given that anechoic data for this model exists.

cheers

 

[Sonnie Parker]

Yeah... it's always been right about 2.5 meters from the ears to the speakers because I pull my speakers on out into the room. A few months ago I found moving my chair forward helped improve imaging and soundstage even more, so now ears are prolly 2.1 or 2.2 meters. I'll check it and make sure. I removed the recliners to the left and right of me on the front row... two small tables on each side. New recliner has about 4-5 layers of blanket on it to help with reflections. Not a lot going on, but some. If I lay the recliner all the way back, you can see the difference, but again, it's minor. Previously when it was leather it was very reflective. I do believe we are getting much more direct sound in my room than most probably will in their rooms.

 

[ddude003]

Could it come down to the difference between a dipole for highs and mids (forward and back radiating) + forward and back firing lows of the ML15a verses mostly forward radiating design of the Revels??? It would be interesting to compare Kipple Spinoramas for all of the speakers tested...

I noticed that even @AJ Soundfield is experimenting with forward and back radiating designs... The OBT1 is an open baffle and the CTA1 has a rearward firing horn into a diffusor... Interesting... Would love to see what Kipple thinks of these...

Does @Sonnie have first reflection point absorbers or diffusers in his room???

 

[JStewart]

It would be interesting to compare Kipple Spinoramas for all of the speakers tested...

Ha, no spins available for the MLs. Erin offered to run the scans but Sonnie thought they would be too much trouble to move. They're only 140lbs each. Whatever...

 

[JStewart]

forward and back radiating designs

Cardoid pattern IIRC and they sounded pretty good to me when I heard 'em at the Florida Audio Expo Jan 2020.

 

[AJ Soundfield]

Would love to see what Kipple thinks of these...

The Klippel would be fine, Amir would be completely confused. The rear radiation is delayed, decorrelated and diffuse. Would be highly amusing though.

 

[AJ Soundfield]

the Florida Audio Expo Jan 2020.

Just signed the 2022 contract. We're back. See if we can lure Sonnie down...

 

[JStewart]

Just signed the 2022 contract. We're back. See if we can lure Sonnie down...

 

[Sonnie Parker]

Does @Sonnie have first reflection point absorbers or diffusers in his room???

No ... and I don't think there is much radiating to the sides of the 15A's ... however, with the F328's, they are much wider dispersion speakers and there may be some reflections. Toole did the testing where the majority preferred no reflections, but IIRC it was only one speaker, which wouldn't work for me, as I wouldn't be able to determine proper soundstage and imaging with one speaker. Not sure how preferring no side reflections translates to good SS&I.

 

[ddude003]

Are you talking about The Practical Effects of Lateral Energy in Critical Listening Environments by Toole? For early reflections 8 people were for diffusion, 7 were for absorption and 11 were for early reflections... I wonder how that would work out for di-poles vs speakers in a box... No matter how hard one works towards everything being equal in these kinds of testing for preferences, nothing is equal... And thats OK...

 

[ddude003]

The Klippel would be fine, Amir would be completely confused. The rear radiation is delayed, decorrelated and diffuse. Would be highly amusing though.

What does Amir have to do with any of this and who cares what Amir thinks?

 

[ddude003]

Ha, no spins available for the MLs. Erin offered to run the scans but Sonnie thought they would be too much trouble to move. They're only 140lbs each. Whatever...

You know those old guys... First thing to go is the ears and then the back...

 

[Sonnie Parker]

Are you talking about The Practical Effects of Lateral Energy in Critical Listening Environments by Toole? For early reflections 8 people were for diffusion, 7 were for absorption and 11 were for early reflections... I wonder how that would work out for di-poles vs speakers in a box... No matter how hard one works towards everything being equal in these kinds of testing for preferences, nothing is equal... And thats OK...

Yep... and I'm not sure it's relevant since there was not a real clear winner. There are 8 vs 18, 7 vs 19 or 11 vs 15... so the majority sides with the two groups that don't prefer one of the three. The as you mention... types of speakers ... then we have different rooms, probably of which very few are similar to where they tested... yep... nothing is equal.

In 2014, Tervq et al. tested the preferences of 15 sound engineers who listened in nine different environments. They found that the preference depended on the task of the engineer, mixing or mastering, and on the specific song. In general it was found that mixing engineers preferred the clarity provided by a lack of reflections, while mas tering engineers preferred more reverberant environments. The latter is probably good in that it is closer to how customers listen. Again, they found evidence of significant adaptability.

So, if we are looking to professional sound engineers for guidance in loudspeakers and acoustical treatment, we find that they are not all in agreement—except, perhaps, about what they prefer to listen to when they relax.

Numerous arguments exist that delayed sounds degrade sound quality, imaging, soundstage, clarity, speech intelligibility and so on. For some people this is true. But, for others it is not.

There is evidence that some professionals are able to mix in a variety of different acoustical circumstances, indicating that adaptation is possible.

There is also the adaption to what we are hearing... we get use to it.

What does Amir have to do with any of this and who cares what Amir thinks?

I think he was making a joke at Amir... as he's been mentioned a few times in reference to testing Harman speakers.

You know those old guys... First thing to go is the ears and then the back...

I might resemble that remark.

 

[MediumRare]

Yeah... it's always been right about 2.5 meters from the ears to the speakers because I pull my speakers on out into the room. A few months ago I found moving my chair forward helped improve imaging and soundstage even more, so now ears are prolly 2.1 or 2.2 meters. I'll check it and make sure. I removed the recliners to the left and right of me on the front row... two small tables on each side. New recliner has about 4-5 layers of blanket on it to help with reflections. Not a lot going on, but some. If I lay the recliner all the way back, you can see the difference, but again, it's minor. Previously when it was leather it was very reflective. I do believe we are getting much more direct sound in my room than most probably will in their rooms.

Your test LP was just right, IMO. 2.5 meters is 8.2 feet; that is absolutely not "nearfield". Nearfield is, like, 2.5', e.g. sitting at a mixing desk. Genelec has an excellent chart describing distance and the appropriate sized speaker in their line for each. My room is not small, 17' x 27', with the speakers along the long wall, and my listening position is exactly 2.5 meters.

By the way, the LP for perfect soundstage will vary according to smoothness of the dispersion at different mid-high frequencies, combined with the toe-in. The toe-in, as you well know, changes not only how you perceive the FR rolloff but, critically, the diffraction nulls and peaks at the preferred listening position. The solution, IME, is to do measurements at different toe-ins using the moving microphone process and chose the toe-in with the best curve. Then you can leave your LP at the ideal distance for your room layout and (likely) screen viewing.

 

[MediumRare]

The Klippel would be fine, Amir would be completely confused. The rear radiation is delayed, decorrelated and diffuse. Would be highly amusing though.

Aside from your personal attack on Amir, which is completely unjustified and does not reflect well on you, the whole benefit of the Klippel system is to account for reflections, including from dipoles. A possible issue is in-wall applications where a separate Klippel module is required for to account for the infinite baffle (not this situation at all). Otherwise, Klippel (and Amir) would be able to predict what to expect in a typical listening room. That is not the same as a room-specific EQ, which is what this post is about.

 

[Sonnie Parker]

Your test LP was just right, IMO. 2.5 meters is 8.2 feet; that is absolutely not "nearfield". Nearfield is, like, 2.5', e.g. sitting at a mixing desk. Genelec has an excellent chart describing distance and the appropriate sized speaker in their line for each. My room is not small, 17' x 27', with the speakers along the long wall, and my listening position is exactly 2.5 meters.

By the way, the LP for perfect soundstage will vary according to smoothness of the dispersion at different mid-high frequencies, combined with the toe-in. The toe-in, as you well know, changes not only how you perceive the FR rolloff but, critically, the diffraction nulls and peaks at the preferred listening position. The solution, IME, is to do measurements at different toe-ins using the moving microphone process and chose the toe-in with the best curve. Then you can leave your LP at the ideal distance for your room layout and (likely) screen viewing.

Right now I am actually at 2.2 meters, after moving my chair forward a bit.

What you mentioned is what I've done. I've measured the response with varying toe-ins... and we did this trial once when Wayne and Dennis were here for good measure. Since I've moved my chair forward, I also adjusted the toe-in slightly, but I may do another set of measurements with varying toe-in again. Measurements are already very close, but might be curious to see if the toe-in angles changed anything with the chair a little forward.

 

[MediumRare]

Measurements are already very close, but might be curious to see if the toe-in angles changed anything with the chair a little forward.

Exactly. My measurements (at 2.5 meters) show very little difference below about 8k for toe-in, but that's really dependent on the specific speaker's dispersion pattern.

 

[dsnyder0cnn]

Very cool experiment. Thanks for sharing your results.

I'm using Acourate from AudioVero for digital room correction. A while back, out of similar curiosity, I built a pair of filters that were identical except that one did no correction above 320 Hz. My hypothesis was that I'd easily be able to hear differences and that I'd have a strong preference for the full-range correction. I was wrong. The two corrections sounded way more similar than I expected. Since I was not testing blind, I managed to convince myself that I had a very slight preference for full-range correction, and I've not looked back.

I agree with you that it's difficult to appreciate the differences (or lack thereof) until you've actually tried this experiment. Like you, I'm using room treatments. At the time, I was testing with Legacy Audio FOCUS SE loudspeakers. It's possible that different speakers in an untreated room would produce different results.

 

[Flak]

I'll say "corrected", because Dirac does much more than EQ.

Let me quote this
magnitude response is important but impulse response is relevant as well... as is phase coherence between speaker pairs for better imaging.

 

[dsnyder0cnn]

Let me quote this
magnitude response is important but impulse response is as well... as is phase coherence between speaker pairs for better imaging.

I have Dirac Live in my Emotiva RMC-1L HT preamp, but I've not actually measured the response after correction using REW. I should do that. Curious to see what the step response looks like before/after Dirac corrections.