DSP for psychoacoustics

0 Members and 1 Guest are viewing this topic. Read 3068 times.

abomwell

  • Full Member
  • Posts: 366
DSP for psychoacoustics
« on: 9 Aug 2021, 03:03 pm »
I know conventional thinking is that speakers should only be EQed for bass issues below around 300Hz. But so many classical recordings sound unnaturally bright (on any speaker). I have found that inserting a 3-4 dB dip, centered at 3kHz that extends down to 1 kHz, has made those recordings sound more natural. Has anyone experimented with this? You can try this for yourself if you have parametric EQ capabilities and I'd love to hear about your thoughts on this after
experimenting.  Here is a quote from the highly respected Sigfried Linkwitz regarding that.

"Around 3kHz our hearing is less sensitive to diffuse fields. Recording microphones, though, are usually flat in frequency response even under diffuse field conditions. When such recordings are played back over loudspeakers, there is more energy in the 3kHz region than we would have perceived if present at the recording venue, and a degree of unnaturalness is introduced. This applies primarily to recordings of large orchestral pieces in concert halls where the microphones are much closer to the instruments than any listener [would be]. At most listening positions in the hall the sound field has strong diffuse components.”

In order to achieve this you can use a -3 or -4 dB dip with a 3kHz center frequency and a Q of .75.

Al

DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #1 on: 9 Aug 2021, 04:21 pm »
I know conventional thinking is that speakers should only be EQed for bass issues below around 300Hz. But so many classical recordings sound unnaturally bright (on any speaker). I have found that inserting a 3-4 dB dip, centered at 3kHz that extends down to 1 kHz, has made those recordings sound more natural. Has anyone experimented with this? You can try this for yourself if you have parametric EQ capabilities and I'd love to hear about your thoughts on this after
experimenting.  Here is a quote from the highly respected Sigfried Linkwitz regarding that.

"Around 3kHz our hearing is less sensitive to diffuse fields. Recording microphones, though, are usually flat in frequency response even under diffuse field conditions. When such recordings are played back over loudspeakers, there is more energy in the 3kHz region than we would have perceived if present at the recording venue, and a degree of unnaturalness is introduced. This applies primarily to recordings of large orchestral pieces in concert halls where the microphones are much closer to the instruments than any listener [would be]. At most listening positions in the hall the sound field has strong diffuse components.”

In order to achieve this you can use a -3 or -4 dB dip with a 3kHz center frequency and a Q of .75.

Al

If it works for you and makes your listening experience better then that's awesome.  I disagree with "conventional wisdom" when it comes to using EQ above 300Hz.  Every concert system you've ever heard, every movie theatre, every professional audio system has EQ across the spectrum to account for difficulties in the room.  What speakers are you listening on? I would think that open baffle or electrostatics would also help you achieve what you're looking for. 
I like the quote you put up.  When microphones are close to something, you don't get the same feel as actually listening to the instrument.  That's a constant challenge for studio and live engineers. 

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #2 on: 9 Aug 2021, 05:00 pm »
I'm currently listening to Dutch and Dutch 8C's which have built-in
DSP capabilities. I measure, and hear, a null at the listening position around 50Hz due to speaker interaction. Measuring individual speakers doesn't reveal that null. I'm hoping the dipole's lateral bass cancellation will eliminate that interaction problem. Using subwoofers in opposite room corners also corrects the problem.

Al

DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #3 on: 9 Aug 2021, 09:19 pm »
I'm currently listening to Dutch and Dutch 8C's which have built-in
DSP capabilities. I measure, and hear, a null at the listening position around 50Hz due to speaker interaction. Measuring individual speakers doesn't reveal that null. I'm hoping the dipole's lateral bass cancellation will eliminate that interaction problem. Using subwoofers in opposite room corners also corrects the problem.

Al

Those are interesting speakers. I've never heard them but they seem great.  A null around 50hz is common and if the subs correct it then that's great. It's likely just the size of the room causing it and 50Hz is cancelled at the listening spot. 
When I read about those speakers you're using, they utilize a phase cancelling mid range system that reduces the reflections of the speaker.  I'm wondering if this almost works too well for classical music which depends on so many reflections off of the walls in a concert hall or wherever the instrument is being played.  Maybe your ear prefers to actually hear all of the mid range reflections from the speaker in your room and that makes it sound more natural to you.  This is just a thought since their description talks about making the mid range more directive.  In reducing the 3kHz you are increasing a lot of the lower mid range (body of the instruments) and therefore bringing back the "room" sound. 
Do these speakers have a room correcting DSP program?

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #4 on: 9 Aug 2021, 09:43 pm »
Good thoughts, Danny. The D&D speakers work with REW and have extensive DSP filters that REW automatically populates upon its room measurement process. The midrange cancellation you mention works like a dipole to eliminate midrange colorations that occur when speakers are placed close to the front wall boundary. The speaker, thus, not only allows close front wall placement but benefits from it by boundary coupling of the rear firing subwoofers.

The idea of the design is to have a very even polar response so reflected sound off walls are full spectrum copies of the direct sound. When that occurs the wall reflections are perceptually ignored. Omnis and dipoles work this way. In a conversation with Clayton he feels this is vitally important and an important goal in his designs.

Al

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #5 on: 9 Aug 2021, 09:54 pm »
A more in depth article by Robert E Greene, of The Absolute Sound, about this psycho-acoustical phenomena is here:
https://www.theabsolutesound.com/articles/in-memoriam-siegfried-linkwitz-19352018

Al

DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #6 on: 10 Aug 2021, 07:09 pm »
A more in depth article by Robert E Greene, of The Absolute Sound, about this psycho-acoustical phenomena is here:
https://www.theabsolutesound.com/articles/in-memoriam-siegfried-linkwitz-19352018

Al

Cool stuff.  At the very least, I love how that guy thought so in depth about the listening process. 
For live shows, one of the more popular PA system companies uses a similar technology with their line arrays.  They have side firing 10" drivers on every box that has cancel behind them.  This has several advantages; 1) the band on stage is not bombarded by what the back of the PA sounds like (it's literally 25db quieter behind an arena PA system than in front of it.  It's almost unreal.  2) This minimizes the reflections off of the terrible acoustics of arenas and other large venues so that you get more of the direct sound.  The line array is:
https://www.dbaudio.com/global/en/products/series/sl-series/gsl12/

It's definitely not consumer stuff but when you get down to it, the physics are the same. 

I really love when Linkwitz is talking about the ear canal shape.  Since we know that we all have different ears, we are all experiencing it differently.  We normally have a "lead ear" that we listen more with.  Much of the sound we hear is actually a reflection off of the rest of our ears to locate it in the physical space.  How do we account for all of this in our home stereo system?  What sounds good to you might be slightly off to me.  For me, that's the fun of it. 

DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #7 on: 10 Aug 2021, 07:10 pm »
Good thoughts, Danny. The D&D speakers work with REW and have extensive DSP filters that REW automatically populates upon its room measurement process. The midrange cancellation you mention works like a dipole to eliminate midrange colorations that occur when speakers are placed close to the front wall boundary. The speaker, thus, not only allows close front wall placement but benefits from it by boundary coupling of the rear firing subwoofers.

The idea of the design is to have a very even polar response so reflected sound off walls are full spectrum copies of the direct sound. When that occurs the wall reflections are perceptually ignored. Omnis and dipoles work this way. In a conversation with Clayton he feels this is vitally important and an important goal in his designs.

Al
Have you tried different listening spots and re-doing the REW several times to see what you like?  I have mixed feelings about the automatic DSP.  I think it misses sometimes and it's good to recalibrate.
[/list]

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #8 on: 10 Aug 2021, 07:39 pm »
That 50Hz null has been a problem with all the speakers I've had in my current room, Danny. To your point, if I move so as to sit within inches of the rear wall the null is mostly eliminated, but other comb filtering issues appear. And I've tried various speaker and listening locations to no avail. As I mentioned, subs in opposite corners eliminates the problem but I don't want to use subs with the X5's if I can avoid it. I'm hopeful that dipole cancellations will do the trick.

I don't use the REW capabilities with my D&D speakers as I have a DSPeaker Anti-mode X4 preamp/processor that offers automatic room correction as well as extensive manual parametric EQ functions. And it's easier to use than REW.

Al

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #9 on: 10 Aug 2021, 08:14 pm »
Cool stuff.  At the very least, I love how that guy thought so in depth about the listening process. 
For live shows, one of the more popular PA system companies uses a similar technology with their line arrays.  They have side firing 10" drivers on every box that has cancel behind them.  This has several advantages; 1) the band on stage is not bombarded by what the back of the PA sounds like (it's literally 25db quieter behind an arena PA system than in front of it.  It's almost unreal.  2) This minimizes the reflections off of the terrible acoustics of arenas and other large venues so that you get more of the direct sound.  The line array is:
https://www.dbaudio.com/global/en/products/series/sl-series/gsl12/

It's definitely not consumer stuff but when you get down to it, the physics are the same. 

I really love when Linkwitz is talking about the ear canal shape.  Since we know that we all have different ears, we are all experiencing it differently.  We normally have a "lead ear" that we listen more with.  Much of the sound we hear is actually a reflection off of the rest of our ears to locate it in the physical space.  How do we account for all of this in our home stereo system?  What sounds good to you might be slightly off to me.  For me, that's the fun of it.

I definitely agree that people hear, and listen, differently.  BTW, that psycho-acoustical dip is not new. It was used on several BBC monitor speakers as well many years ago.

Al

DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #10 on: 10 Aug 2021, 08:52 pm »
I definitely agree that people hear, and listen, differently.  BTW, that psycho-acoustical dip is not new. It was used on several BBC monitor speakers as well many years ago.

Al

Would you say that your speakers are mostly flat across the frequency spectrum before you implement the dip around 3KHz?

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #11 on: 10 Aug 2021, 09:37 pm »
Would you say that your speakers are mostly flat across the frequency spectrum before you implement the dip around 3KHz?

Yes, at 1 meter, they measured either flat, as in the Morrison model 29, the Janszen ESL's, or slightly downsloping as with the Dutch and Dutch 8C's. Keep in mind the dip is mainly for large scale orchestral or choral recordings. My DSPeaker X4 preamp/processor has several presets. I use one preset for such recordings and others, without the dip, for jazz, and chamber music.

Al

Al

DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #12 on: 10 Aug 2021, 10:23 pm »
Yes, at 1 meter, they measured either flat, as in the Morrison model 29, the Janszen ESL's, or slightly downsloping as with the Dutch and Dutch 8C's. Keep in mind the dip is mainly for large scale orchestral or choral recordings. My DSPeaker X4 preamp/processor has several presets. I use one preset for such recordings and others, without the dip, for jazz, and chamber music.

Al

Al

Very cool.

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #13 on: 10 Aug 2021, 10:40 pm »
The really cool thing is that anyone can try this easily for themselves at no cost if one has access to some form of DSP. Nothing to lose!

Al

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #14 on: 15 Aug 2021, 01:23 pm »
FWIW, I have a good friend who played principal trombone in the Los Angeles Philharmonic for over 30 years. When he comes over to listen he always prefers the sound when using the psycho-acoustical dip, on orchestral recordings, vs. no EQ. Trombone players, to my mind, can be excellent judges of accurate orchestral sound because they spend half the time playing in the orchestra, and, since there are so few compositions written before the early 19th century with trombone parts,  half the time listening in the hall during rehearsals.

Al


DannyBadorine

  • Full Member
  • Posts: 384
Re: DSP for psychoacoustics
« Reply #15 on: 15 Aug 2021, 03:41 pm »
FWIW, I have a good friend who played principal trombone in the Los Angeles Philharmonic for over 30 years. When he comes over to listen he always prefers the sound when using the psycho-acoustical dip, on orchestral recordings, vs. no EQ. Trombone players, to my mind, can be excellent judges of accurate orchestral sound because they spend half the time playing in the orchestra, and, since there are so few compositions written before the early 19th century with trombone parts,  half the time listening in the hall during rehearsals.

Al

It really makes sense to me.  I've been working in live sound for over a decade and when horns or strings get harsh sounding, I hate it.  Many condenser microphones hype the 2kHz-6KHz frequency range so I prefer Condensers on them but that isn't always practical in a live setting.  The dip that you're making is something I often do on a mixing console to get them sounding better.

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #16 on: 15 Aug 2021, 04:30 pm »
Interesting to know you do that, Danny. The dip is very similar to to the shape of the Fletcher-Munson curve above 1kHz. I would think that's not a coincidence.


Desertpilot

  • Full Member
  • Posts: 530
  • Retired
Re: DSP for psychoacoustics
« Reply #17 on: 15 Aug 2021, 05:35 pm »
The really cool thing is that anyone can try this easily for themselves at no cost if one has access to some form of DSP. Nothing to lose!

Al

Nice thread Al.  I'm super busy at home so I have very little time to post here (which some members may cheer - hahaha).

I'm convinced DSP is important to correct room interactions.  I intend on full range correction because I agree, the 2K range may be problematic.  However, having spent so much money on speakers and DAC, I am not going to try and do this myself.  Plus, I am so dang tired of watching YouTube videos and reading forum posts about how to do measurements, DSP and acoustic treatments.  I contacted Mitch Barnett at Accurate Sound and he recommended Audiolense XO as I have a multichannel surround setup (5.1).  I got it and need to get to work with it.  I will also hire Mitch to help me put all this together.  I have a lot of skills but measurements and DSP is not one of them.

Mitch's interview:  https://www.youtube.com/watch?v=H1n8o7Wv4LQ&t=2716s

My only consternation is that it sounds great right now.  Why bother?  But, I know I cannot "hear" problems, (i.e. comb filtering, SBIR, dips and peaks, reflections, etc.).  So, let's do this and see what happens.

Marcus

abomwell

  • Full Member
  • Posts: 366
Re: DSP for psychoacoustics
« Reply #18 on: 15 Aug 2021, 05:53 pm »
Marcus, keep in mind I've yet to hear the X5's so perhaps nothing needs to be changed at all. If you're happy with the sound maybe leave well enough alone? In the words of the late great Sigfried Linkwitz "Unfortunately, marketing departments and dealers think that bass and high frequencies need to be emphasized for products to sell".  I'm sure Clayton is not one of them.

Al


nickd

Re: DSP for psychoacoustics
« Reply #19 on: 15 Aug 2021, 07:19 pm »
Nice discussion gents,
I had dinner with the late James Bongiorno at CES many years ago. He was aware of the issues being discussed here and spent quite some time thinking about how to keep a system “purest” and still compensate for these recording differences.

Not an easy task even for an brilliant designer. His unrealized solution was to mount loudspeakers on a robotic sled with programmed positioning for each album in your collection. The sled would adjust loudspeakers toe and listening distance to compensate for this type of perceived balance.

We both laughed and enjoyed the company and the theory discussion that evening knowing it would likely never come to fruition.

Many years later I adopted DSP room correction (Lyngdorf) and placed my speakers close to the back wall. My system like all others is not perfect in room, but adjustable on the fly and the sound on any reasonable recording is everything I could ask for.

There is not one perfect solution for every system and room, but glad to see you guys out there in the bold DSP frontier working on solutions for your systems and rooms. There is no point standing still, acting like the problems in the sound reproduction chain don’t exist.

In this regard, the pro sound guys are way ahead of the average audiophile. Keep the discussion going.  :popcorn: