Alternative perspective on the Harman loudspeaker preference curve.

[Daverz]

I've been using the "Olive-Toole" target from the paper below together with Acourate. This is a slide taken from page 42 of

https://accuratesound.blob.core.windows.net/sotadrc/Understanding_the_SOTA_of_DRC.pdf

It's essentially a tilt of -1 dB/octave. It sounds pretty good to me.

 

[youngho]

@sigbergaudio: you can find a lot more data from this paper (and others) here: https://www.aes-media.org/sections/pnw/pnwrecaps/2014/solive_sep14/

There are a number of points that might warrant further discussion:
1. Slide 14 and 15 seems to indicate that the response of all channels in the HRR is reasonably linear before equalization, though the smoothing hides a lot. I wonder why the side and rear channels don't show any rise in the bass relative to the left and right front channels when they appear to be much closer to the walls.
2. Slide 16 indicates that Target 1 was based on measurements of the JBL LSR (6332, though that was specified in the slide) in the HRR, but then slide 19 would imply a linear response. I don't think that's actually the case because of #4 below. Data for the LSR 6332 can be found at https://jblpro.com/en-US/site_elements/lsr6332-spec-sheet, can compare with slide 24
3. Slide 20: note that JBL Target 2 (basically Target 1 with slightly less bass and treble) was preferred over Target 1. That only makes sense if that Target 1 did not reflect the linear response in slide 15.
4. Slide 24: compare the Before EQ measurements of the Revel 208 with what we saw in slide 15. Here we see what appear to be significant modal and boundary effects below the presumed Schroeder frequency range. I have to assume that the JBL LSR 6332 measurements resulting in Target 1 above also reflected these effects, as well, so that could explain why slightly bass as in Target 2 was preferred over Target 1. In terms of the After EQ curve, I don't quite understand why the 80 Hz peak couldn't be flattened more, since the net result of averaging seems to result in a response that's a little bass-shy below 60 Hz relative to a little bright above 1 kHz.
5. Slide 30: the red arrows seem to be in the wrong position along the x-axis. Reading the discussion by Todd Welti at https://hometheaterhifi.com/technic...n-interview-with-todd-welti-and-kevin-voecks/ but ignoring the diagram, I would have thought that the range from 105 to 2.5 kHz was pretty linear, rather similar to https://tech.ebu.ch/docs/tech/tech3276.pdf. These tilts seem to be pretty smooth without the lumpiness/variability seen in Toole's curve from his paper: https://www.aes.org/e-lib/browse.cfm?elib=17839
6. Slide 33: here is some information on the listeners.
7. Slide 37 and 38: I think there's an error in one of these slides, since the untrained listeners preferred an average of -0.8 treble with speaker and headphone combined, but then they preferred an average of +1.8 on headphones and again on speakers when the data is separated? I'm going to guess that the headphones should have been -2.6, not +1.8. The right part of the graph goes towards the relative difference of treble and bass.
8. Slide 39: Wow, look at the huge variability between listeners! Who were the untrained listeners? Probably 400, 401, and 402. Maybe the three women were 4, 58, and 346.
9. Slide 40: again, a lot of variability in the delta
10. Slide 45: "The preferred bass and treble levels of the target function for loudspeakers and headphones varied among individual listeners (see Fig 14). For loudspeaker playback, the range of preferred bass and treble levels was 17 dB and 11 dB, respectively." That certainly seems like a lot of variability to me!

 

[gnarly]

This is what Dr Toole also said in another post at AVSForum:

https://www.avsforum.com/threads/official-jbl-synthesis-pro-revel-home-theater-thread.2515137/page-42?post_id=57820394&nested_view=1#post-57820394

​

If a "target' curve has been achieved, and the sound quality is not satisfactory, the suggestion is often to go into the menu, find the manual adjustment routine, and play around with the shape of the curve until you or your customer like the sound. This is not a calibration. This is a subjective exercise in manipulating an elaborate tone control. Once set it is fixed, and in it will be reflected timbral features of the music being listened to at the time. In other words, the circle of confusion is now included in the system setup. By all means do it, but do not think that the exercise has been a "calibration". Old fashioned bass & treble tone controls and modern "tilt" controls are the answer and they can be changed at will to compensate for personal taste and excesses or deficiencies in recordings. Sadly, many "high end" products do not have tone controls - dumb. It is assumed that recordings are universally "perfect" - wrong!​

​

Thank you, i love that quote from Dr Toole.

If I may requote the last two sentences below, and add MY bold emphasis.. (not Dr Toole's)

The fact that recordings do vary, dominates the idea that one perfect house curve can be used, with no further tonal adjustments ever needed.
My default house curve, after listening and using tone controls on many recordings when deemed needed, ends up being whatever curve required the fewest recording-by-recording tonal corrections.
Seems like a way to avoid getting too picky over a statistical/theoretical curve, and find what is pragmatic and personal, given the speakers and room.

"Old fashioned bass & treble tone controls and modern "tilt" controls are the answer and they can be changed at will to compensate for personal taste and excesses or deficiencies in recordings. Sadly, many "high end" products do not have tone controls - dumb. It is assumed that recordings are universally "perfect" - wrong!"​

​

​

 

[anotherhobby]

Personally I would be reluctant to alter the natural response of the speakers above the Schroeder frequency. If the speaker is decent, it should have a good natural response. But of course if the overall tonality is too bright or too dark, one could try to tilt this. What I definitely wouldn't do is to try to smoothen out bumps in the response to force it to the target.

I completely agree. Both of my systems degrade in sound quality to me when correction is allowed to go above about 500 Hz. One is Audyssey and one is Dirac.

 

[waynel]

This is what Dr Toole also said in another post at AVSForum:

https://www.avsforum.com/threads/official-jbl-synthesis-pro-revel-home-theater-thread.2515137/page-42?post_id=57820394&nested_view=1#post-57820394

​

If a "target' curve has been achieved, and the sound quality is not satisfactory, the suggestion is often to go into the menu, find the manual adjustment routine, and play around with the shape of the curve until you or your customer like the sound. This is not a calibration. This is a subjective exercise in manipulating an elaborate tone control. Once set it is fixed, and in it will be reflected timbral features of the music being listened to at the time. In other words, the circle of confusion is now included in the system setup. By all means do it, but do not think that the exercise has been a "calibration". Old fashioned bass & treble tone controls and modern "tilt" controls are the answer and they can be changed at will to compensate for personal taste and excesses or deficiencies in recordings. Sadly, many "high end" products do not have tone controls - dumb. It is assumed that recordings are universally "perfect" - wrong!​

​

I believe that this is the relevant link:

https://www.avsforum.com/threads/official-jbl-synthesis-pro-revel-home-theater-thread.2515137/page-168#post-57820394

 

[YSC]

Personally I would be reluctant to alter the natural response of the speakers above the Schroeder frequency. If the speaker is decent, it should have a good natural response. But of course if the overall tonality is too bright or too dark, one could try to tilt this. What I definitely wouldn't do is to try to smoothen out bumps in the response to force it to the target.

same here, I tried a few EQ options, including a trial on full spectrum EQ (which instantly sounded weird and abandoned), just tune to the untrained and trained listener bass boost on the sub part and ended up liking the bass response of a 1db lower than untrained listener curve to balance out pop music, movies and games, but tilting down the general trend up the HF at nearfield sounding really akward also to me so ended up >200hz is untouched

 

[anotherhobby]

This is what Dr Toole also said in another post at AVSForum:

https://www.avsforum.com/threads/official-jbl-synthesis-pro-revel-home-theater-thread.2515137/page-42?post_id=57820394&nested_view=1#post-57820394

​

If a "target' curve has been achieved, and the sound quality is not satisfactory, the suggestion is often to go into the menu, find the manual adjustment routine, and play around with the shape of the curve until you or your customer like the sound. This is not a calibration. This is a subjective exercise in manipulating an elaborate tone control. Once set it is fixed, and in it will be reflected timbral features of the music being listened to at the time. In other words, the circle of confusion is now included in the system setup. By all means do it, but do not think that the exercise has been a "calibration". Old fashioned bass & treble tone controls and modern "tilt" controls are the answer and they can be changed at will to compensate for personal taste and excesses or deficiencies in recordings. Sadly, many "high end" products do not have tone controls - dumb. It is assumed that recordings are universally "perfect" - wrong!​

​

The link offers much better context. Thank you.

 

[test1223]

There are two very important side conditions which influence a good in room response curve which I like to point out.

1. The cuves of equal loudness perception which are sound pressure level dependent. So the level difference between spl target of the master recording and the spl which you use at your listening position have to be taken into account.

2. The reverberation time especially in the bass. I have tried some different bass concepts like bass reflex, closed box and dipole bass and you need to adapt the relative bass spl with lower reverberation time in the bass you need more bass spl.

There are some other important points which influence a good target curve.

So the best target curve isn't independent from your specific set up.

 

[Daverz]

Rather than just use a pre-existing target, Denis Sbragion's DRC-FIR uses a pscyhoacoustic method to derive a target from the in-room response. It's described here:

DRC: Digital Room Correction

Here's the correction this gives in my room. Red is the uncorrected in-room response, green is the corrected response, and blue is the psychoacoustic target.

The correction has a slight tilt down, which tends to sound too bright to me. So perhaps best combined with something like a tilt control (as in the new version of CamillaDSP).

 

[Geert]

So the "untrained" listeners - on average - did not EQ any bumps into the FR but preferred a continuous trend for more brightness - compared to "trained".

They bumped up the bass, made a dip in low mid to avoid bloated bass, and then bumped up the highs to balance them with the bass. While the trained listeners were happy with the natural in room response of an anachonic neutral speaker.

 

[olieb]

While the trained listeners were happy with the natural in room response of an anachonic neutral speaker.

Actually they were not, at least that is what it seems.
Assumed the response of the "neutral speaker" was anything near to the predicted in room response for a highly rated loudspeaker (black) then red would be the EQ needed to arrive at the curve for "trained".
The "trained listeners" kind of tuned down the highs, upped the bass and pronounced the mids somewhat (or dipped upper bass if you prefer to see it that way). Certainly it is not flat.

 

[Geert]

The "trained listeners" kind of tuned down the highs, upped the bass and pronounced the mids somewhat (or dipped upper bass if you prefer to see it that way). Certainly it is not flat.

What I meant to say is they we're happy with something closer to neutral. The conditions these tests happen in aren't text book perfect, so there's no point in arguing over details. It's the trend that matters.

 

[sigbergaudio]

Rather than just use a pre-existing target, Denis Sbragion's DRC-FIR uses a pscyhoacoustic method to derive a target from the in-room response. It's described here:

DRC: Digital Room Correction

Here's the correction this gives in my room. Red is the uncorrected in-room response, green is the corrected response, and blue is the psychoacoustic target.

View attachment 336036
The correction has a slight tilt down, which tends to sound too bright to me. So perhaps best combined with something like a tilt control (as in the new version of CamillaDSP).

But it corrects rather actively way up in frequency then. Will probably be a bit hit and miss.

 

[sigbergaudio]

Actually they were not, at least that is what it seems.
Assumed the response of the "neutral speaker" was anything near to the predicted in room response for a highly rated loudspeaker (black) then red would be the EQ needed to arrive at the curve for "trained".
The "trained listeners" kind of tuned down the highs, upped the bass and pronounced the mids somewhat (or dipped upper bass if you prefer to see it that way). Certainly it is not flat.
View attachment 336041

That's not exactly what the data says. The black dotted line (black line in your graph) is predicted curve, and shaded area (allowing from ~0-6dB rise in the bass) is possible results depending on the room and presumably listening distance. So the trained listener curve is within that shaded area except in the upper end. It would of course be useful if the study included information about what was the actual natural response in the room, so in what way did untrained and trained listeners deviate from that.

The original graph from the study also looks a bit strange to me. I rarely see speakers with less than 5dB natural rise in a room with natural "living room" listening distance. I also basically never see a natural response in the room that do not fall quite a lot at least after 10khz and sometimes sooner. While this graph indicates a natural response that goes flat to 20khz. That's simply wrong.

 

[youngho]

That's not exactly what the data says. The black dotted line (black line in your graph) is predicted curve, and shaded area (allowing from ~0-6dB rise in the bass) is possible results depending on the room and presumably listening distance. So the trained listener curve is within that shaded area except in the upper end. It would of course be useful if the study included information about what was the actual natural response in the room, so in what way did untrained and trained listeners deviate from that.

The original graph from the study also looks a bit strange to me. I rarely see speakers with less than 5dB natural rise in a room with natural "living room" listening distance. I also basically never see a natural response in the room that do not fall quite a lot at least after 10khz and sometimes sooner. While this graph indicates a natural response that goes flat to 20khz. That's simply wrong.

Hi, I sent you a copy of the 2014 AES PNW presentation by Sean Olive, which I believe addresses a number of questions you had. See also my reply above once you look at the slides: https://www.audiosciencereview.com/...er-preference-curve.50602/page-3#post-1819700

 

[Sancus]

This is not a calibration. This is a subjective exercise in manipulating an elaborate tone control. Once set it is fixed, and in it will be reflected timbral features of the music being listened to at the time. In other words, the circle of confusion is now included in the system setup.

The circle of confusion is really the heart of the problem with thinking of one specific curve or whatever. Even if every single person preferred an identical amount of treble and bass, you'd still have variations in settings needed to compensate for music recordings which vary wildly.

 

[sigbergaudio]

Hi, I sent you a copy of the 2014 AES PNW presentation by Sean Olive, which I believe addresses a number of questions you had. See also my reply above once you look at the slides: https://www.audiosciencereview.com/...er-preference-curve.50602/page-3#post-1819700

Not sure I understood how they got the conclusion "Preferred in-room loudspeaker response is not flat but has a bass boost
about 6.6 dB @ 105 Hz and treble cut of -2.4 dB above 2.5 kHz" - is it simply an average of all the participants? In that case it's not really accurate for many of the listeners since they seemed to be all over the place

 

[youngho]

Not sure I understood how they got the conclusion "Preferred in-room loudspeaker response is not flat but has a bass boost
about 6.6 dB @ 105 Hz and treble cut of -2.4 dB above 2.5 kHz" - is it simply an average of all the participants? In that case it's not really accurate for many of the listeners since they seemed to be all over the place

Yes, I think that's the average of trained plus untrained. Although Welti commented "There was a fair amount of variation from subject to subject, but if you averaged it all out, you get a bass boost of around 6 dB and a gently rolling off high end, around -2.5 dB at 10 kHz. (Figure 1)," and I already quoted the slides above, I feel like the sheer variance hasn't been highlighted at all here on ASR. It's almost as though we were discussing Target Height as 5' 6.5" because that's about the average height in the US (men about 5'9", women 5'4").

 

[Sined]

This is what Dr Toole also said in another post at AVSForum:

https://www.avsforum.com/threads/official-jbl-synthesis-pro-revel-home-theater-thread.2515137/page-42?post_id=57820394&nested_view=1#post-57820394

​

If a "target' curve has been achieved, and the sound quality is not satisfactory, the suggestion is often to go into the menu, find the manual adjustment routine, and play around with the shape of the curve until you or your customer like the sound. This is not a calibration. This is a subjective exercise in manipulating an elaborate tone control. Once set it is fixed, and in it will be reflected timbral features of the music being listened to at the time. In other words, the circle of confusion is now included in the system setup. By all means do it, but do not think that the exercise has been a "calibration". Old fashioned bass & treble tone controls and modern "tilt" controls are the answer and they can be changed at will to compensate for personal taste and excesses or deficiencies in recordings. Sadly, many "high end" products do not have tone controls - dumb. It is assumed that recordings are universally "perfect" - wrong!​

​

Yep. The "Circle of confusion", again and always...: our worst enemy! Even if you would have calibrated your speakers perfectly to a "well recognized" target curve, it will works great for some recordings, but for some others, you will not get the "pleasing results" expected; it can be even worse, depending on the bass content, if it is not all there or, if it is all there, but too much of it somewhere or all over the place...

Amazing to hear such huge differences between various recordings. I defined 4 "bass profiles" filters in my DSP crossover/EQ with different lower/upper bass ratio that I can easily choose from, hoping that one of them can do the appropriate correction job. I certainly agree with Dr. Toole, a target curve is not a "one size fits all" speaker/room calibration and I don't mind not being considered by the audio fools as a "high-end "wire with gain" type", I do enjoy having a way to actively "tune" my speakers to compensate for the deficiencies in some recordings: that helps me to enjoy them better, even if the SQ is not top notch!

 

[MattHooper]

Yep. The "Circle of confusion", again and always...: our worst enemy! Even if you would have calibrated your speakers perfectly to a "well recognized" target curve, it will works great for some recordings, but for some others, you will not get the "pleasing results" expected; it can be even worse, depending on the bass content, if it is not all there or, if it is all there, but too much of it somewhere or all over the place...

Amazing to hear such huge differences between various recordings. I defined 4 "bass profiles" filters in my DSP crossover/EQ with different lower/upper bass ratio that I can easily choose from, hoping that one of them can do the appropriate correction job. I certainly agree with Dr. Toole, a target curve is not a "one size fits all" speaker/room calibration and I don't mind not being considered by the audio fools as a "high-end "wire with gain" type", I do enjoy having a way to actively "tune" my speakers to compensate for the deficiencies in some recordings: that helps me to enjoy them better, even if the SQ is not top notch!

Yep.

This is why I've often pointed out that the criticism of a more colored system on the basis "because it may make some recordings sound better, but not others" applies to a neutral system as well, due to variations in recording quality.