Announcement

Collapse
No announcement yet.

Theory and possible debate: Human Hearing Perception (x-over for your dB preference)

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Theory and possible debate: Human Hearing Perception (x-over for your dB preference)

    I'm not too familiar with the measuring devices you x-over designers use to create a crossover for a specific speaker with a flat response, however a topic came to mind that was discussed years ago on another forum that I was a part of about what we as humans hear as flat and how loud each frequency has to be to make us think it is the same level across the frequency spectrum.

    Info Here:

    http://en.wikipedia.org/wiki/Equal-loudness_contour




    Question 1: does the software or measuring equipment take this into consideration at all? Obviously a microphone will measure perfectly but our human ears do not hear the same way a mic does.

    Question 2: I assume the measuring equipment measures test tones and not pre-mixed music that may have already been mixed to take this into account?

    Question 3: When a music album is mixed, does the sound engineer take this into account and if so, at what SPL does he mix at? Listening to anything other then his same "mixing" SPL will not be heard as the sound engineer intended and will be "off".

    Question 4: While everyone looks for a speaker to be ultimately flat. If I listen to music at lower volume levels then the average sound engineer do I want to custom tailor a x-over to my specific dB listening preference to match the perceved human hearing response?


    My Denon receiver somewhat takes this theory into account with its Audyssey Dynamic EQ. It custom EQ's the frequency response according to the dB level the volume is at, boosting the bass and trebble more as you turn the volume down and cutting the bass and trebble as you turn the volume up. Im my opinion it does a terrific job and I much prefer to listen to my music with the dynamic volume on since I usually listen to music at slightly lower volumes.

    http://www.audyssey.com/technology/dynamiceq.html

    I know Im opening a whole new can of worms with this thread but Im just looking for discussion on the topic and to get some opinions of what you think about this...

  • #2
    Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

    It is a can of worms. Its all about Psychoacoustics, and they are heavily ignored in speaker design because somewhere along the way a test was conducted that some believe that a flat speaker is pinnacle.

    Equal loudness should be considered during design, and a target listening level-range for a design should be established.

    I feel loudspeaker design transcends curves and figures, and HOW we perceive sound is of ultimate importance because at the end of the day its not the room, the response curve, or the -60dB of distortion at regular listening levels we perceive as good sound. HRTF's, equal loudness, and how we perceive everything about the reproduced sound is what counts. This includes many, many factors.

    That said, fq responses, power response, phase, impedance, etc are all tools to achieve good perceived sound. "Honest" source reproduction is a subjective device as strong as "a veil has been lifted" since all source material is recorded and engineered. We are striving to reproduce what some audio engineer thinks is real. Listen to a live, unamplified jazz trio in a room. You may actually be disappointed how low fidelity it is!

    When used properly, the above mentioned tools can start to paint a picture of how certain squiggles sound, and when combined, what a given designer believes is real sound. Flat line or not. It is something not achieved over one design, but many over time that a good designer learns from as to what works for them and what does not.

    There are no absolutes. There are great, established and generations old guidelines and scientific rules that must be followed, however it seems that loudspeaker design is still somewhat subjective in approach. Develop your own methods for design, using all that is available as respective tools. If you let these tools and guidelines own you and ignore the psychoacoustic connection, the risk end up being an amazingly average speaker and there are far too many out there DIY and Retail.

    IMHO, FWIW, Whatever...

    I was inspired at one point to publish my connections with pretty graphs and charts, however since they were not DBT, the life got sucked out of that inspiration....:rolleyes:
    .

    Comment


    • #3
      Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

      Is the BBC dip used for the same reasoning?
      Some explanations I've read on the BBC dip (not saying they're true):
      1) To make the speakers easier to listen to (especially up close in a home environment)
      2) Improve response off-axis when using a larger driver(8"+) across the midrange.
      3) To have a "perceived" flat response.
      4) Designed to satisfy the average consumer's sound preferences.

      Comment


      • #4
        Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

        Read this interesting thread on the idea of "Flat response"

        http://www.diyaudio.com/forums/multi...eo-system.html

        Comment


        • #5
          Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

          It's heresy around here to bring up topics like this. This implies there is no standard of perfection for design. The recording chain, the room acoustics, the relative SPL level at which you listen and who mastered will all probably have a much larger effect than 57th order harmonic distortion or gold/platinum/silver/mink oil/ capacitors. And, don't dare touch the tone controls! :rolleyes:

          BITD most better quality gear had a "loudness" button designed to work in conjunction with the volume control to compensate for these equal loudness curves. (Basically boosting the lower end at lower volumes) Once remote controls became standard, a lot of gear did away with it.

          It's there, it's real, but it's kind of like Aunt Tillie's little drinking problem -- we just don't talk about it. In a way it make sense to leave it out of the discussion, everybody's listening environment is unique, so discussions center around issues that are a bit more universal.
          Lou's Speaker Site [speakers.lonesaguaro.com]
          "Different" is objective, "better" is subjective. Taste is not a provable fact.
          Where are you John Galt? I may not be worthy, but I'm ready.

          Comment


          • #6
            Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

            But speaker design in general is room independant so people either design based on that idea that accuracy requires a flat on and off axis response or they do not? Neither is more right, its completely a choice.

            Dont touch the tone control is funny :D I completely set my in room curve with a DCX as much as I like, I first will try to build flat response speakers first :D

            Comment


            • #7
              Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

              The problem I see with this theory is that we hear live music with the same imperfect hearing as we do any other sound. So designing our speakers to reproduce a frequency response that fills in, so to speak, our lacking areas of hearing would alter the way music sounds to us naturally.

              If you want to design your speaker to sound like anything specific - design it to sound like the speakers the sound engineer used to mix the final recording. That would be as true to what they want you to hear as anything... though it might not be that good

              Comment


              • #8
                Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                Originally posted by pypes View Post

                If you want to design your speaker to sound like anything specific - design it to sound like the speakers the sound engineer used to mix the final recording. That would be as true to what they want you to hear as anything... though it might not be that good
                not to mention remember to listen always in the near field, or the mid-field, or the far-field depending on the album/and or song as you can have multiple mixers, but why not just got for the mastering engineers speakers.

                P.S. 80-90db for mastering playback
                David

                Comment


                • #9
                  Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                  Think about it this way: the equal loudness contour is how we hear OUR WHOLE WORLD, so when a band writes or records a song, the song in itself is "voiced" to sound appropriate based on what our ears like. Or, if it's an orchestra or band, the conductor will "voice" the players so that it sounds "balanced" to our ears (we need more tuba than oboe, if you get what I mean). A mix engineer will usually "voice" a mix this way, as well; we like to hear full bass, uncluttered midrange, and a sprinkling of treble.

                  The job of the speaker designer is to simply play back what the recording engineer created, so it should (theoretically) be flat.

                  Question 4: While everyone looks for a speaker to be ultimately flat. If I listen to music at lower volume levels then the average sound engineer do I want to custom tailor a x-over to my specific dB listening preference to match the perceved human hearing response?
                  The "reference level" I was taught back in audio engineering class was 85 dB. Supposedly, that's the volume to mix at. So it stands to reason that that's teh volume to listen at. However, 85 dB is actually kind of loud for normal households and apartments, especially for long periods of time. So most people listen quieter. When you listen below 85 dB, then yes, the bass will sound thinner (some say the recording sounds "lifeless"), and that's where that ol' "loudness button" came in. I don't know if I agree that the loudness button is extinct; I think it just goes by different names now ("MAXXXbass" or somesuch)

                  When I did the Overnight Sensations, they actually were voiced around 70-75 dB because I wanted them for "nearfield." This voicing probably explains why people tend to like them so much for quiet, casual, nearfield listening.
                  Isn't it about time we started answering rhetorical questions?

                  Paul Carmody's DIY Audio Projects
                  Twitter: @undefinition1

                  Comment


                  • #10
                    Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                    And there it is...
                    Originally posted by Paul Carmody View Post
                    The job of the speaker designer is to simply play back what the recording engineer created, so it should (theoretically) be flat.
                    Our psychoacoustic response represented by ISO 226 is unconsciously 'built in' to our instruments and our performances by those performing them and/or mastering them.

                    C
                    Curt's Speaker Design Works

                    "It is the mark of an educated mind to be able to entertain a thought without accepting it."
                    - Aristotle

                    Comment


                    • #11
                      Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                      Originally posted by pypes View Post
                      The problem I see with this theory is that we hear live music with the same imperfect hearing as we do any other sound. So designing our speakers to reproduce a frequency response that fills in, so to speak, our lacking areas of hearing would alter the way music sounds to us naturally.

                      If you want to design your speaker to sound like anything specific - design it to sound like the speakers the sound engineer used to mix the final recording. That would be as true to what they want you to hear as anything... though it might not be that good
                      While I agree completely that music reproduction should try to imitate a good live stage performance the main difference between listening at home and at a concert is exactly the reason I bring up the topic. The difference being about 30dB give or take. The curve becomes far more flat at the "Live Concert" listening levels... What about listening at home at 65dB? Which is where my question regarding designing a custom crossover for your listening dB prefference comes into play.

                      Comment


                      • #12
                        Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                        Great timing for this thread!.... since I was about to start something similar. I'd thought about hearing being not sensitive to bass linearly with volume level. I'm voicing a build and was really happy with the sound... so smooth. But Most of the time I listen is at lower volume levels. The other day I got to turn it up while auditioning it for a friend - since the wife wasn't home. It was a bass thick muddy mess, really disappointing, and shocking how different it sounds at higher volume levels. I've since adjusted the XO for less bass and it sounds better when turned up though I feel I'm almost starting from scratch again!

                        Comment


                        • #13
                          Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                          Originally posted by Ryan_M View Post
                          Great timing for this thread!.... since I was about to start something similar. I'd thought about hearing being not sensitive to bass linearly with volume level. I'm voicing a build and was really happy with the sound... so smooth. But Most of the time I listen is at lower volume levels. The other day I got to turn it up while auditioning it for a friend - since the wife wasn't home. It was a bass thick muddy mess, really disappointing, and shocking how different it sounds at higher volume levels. I've since adjusted the XO for less bass and it sounds better when turned up though I feel I'm almost starting from scratch again!
                          Hello Mr. Compromise ;) Seriously though, everyone makes certain compromises and this is a common one. One of several real advantages we have as builders is choosing the compromise we like best.

                          You can read about your specific example on a lot of speaker reviews: "As soon as I turned it up, it transformed the speaker etc etc".

                          Just be sure that the changes you are making will make you happy at your preferred listening levels and location!

                          Driscoll is voiced for reasonable and prudent listening levels in the near field. It did alright in Iowa in the large venue at high volumes, but it is much better in its intended install.
                          Don't listen to me - I have not sold any $150,000 speakers.

                          Comment


                          • #14
                            Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                            Originally posted by lamski View Post
                            Question 1: does the software or measuring equipment take this into consideration at all? Obviously a microphone will measure perfectly but our human ears do not hear the same way a mic does.
                            No, it merely compares input versus acoustic output under controlled conditions.

                            Originally posted by lamski View Post
                            Question 2: I assume the measuring equipment measures test tones and not pre-mixed music that may have already been mixed to take this into account?
                            Correct, typically, though it is possible to use music as the input.

                            Originally posted by lamski View Post
                            Question 3: When a music album is mixed, does the sound engineer take this into account and if so, at what SPL does he mix at? Listening to anything other then his same "mixing" SPL will not be heard as the sound engineer intended and will be "off".
                            Also correct.

                            Originally posted by lamski View Post
                            Question 4: While everyone looks for a speaker to be ultimately flat. If I listen to music at lower volume levels then the average sound engineer do I want to custom tailor a x-over to my specific dB listening preference to match the perceved human hearing response?
                            Flat is a better strategy, then EQ as required by listening vs. mastering level. Yes, it MAY be different for every album, even cut, depending upon how anal you want to be about this....

                            Comment


                            • #15
                              Re: Theory and possible debate: Human Hearing Perception (x-over for your dB preferen

                              Originally posted by Paul Carmody View Post
                              Think about it this way: the equal loudness contour is how we hear OUR WHOLE WORLD, so when a band writes or records a song, the song in itself is "voiced" to sound appropriate based on what our ears like. Or, if it's an orchestra or band, the conductor will "voice" the players so that it sounds "balanced" to our ears (we need more tuba than oboe, if you get what I mean). A mix engineer will usually "voice" a mix this way, as well; we like to hear full bass, uncluttered midrange, and a sprinkling of treble.

                              The job of the speaker designer is to simply play back what the recording engineer created, so it should (theoretically) be flat.


                              The "reference level" I was taught back in audio engineering class was 85 dB. Supposedly, that's the volume to mix at. So it stands to reason that that's teh volume to listen at. However, 85 dB is actually kind of loud for normal households and apartments, especially for long periods of time. So most people listen quieter. When you listen below 85 dB, then yes, the bass will sound thinner (some say the recording sounds "lifeless"), and that's where that ol' "loudness button" came in. I don't know if I agree that the loudness button is extinct; I think it just goes by different names now ("MAXXXbass" or somesuch)

                              When I did the Overnight Sensations, they actually were voiced around 70-75 dB because I wanted them for "nearfield." This voicing probably explains why people tend to like them so much for quiet, casual, nearfield listening.
                              Totally agree. The loudness button was a completely genius idea that came about because of the equal loudness curves. listen at 90db everything sounds great. But if you listen at 80db... "hey where'd the bass go??" no problem, just hit that little loudness button and voila! the bass is back. Now if they could design a feature to add onto one of those auto calibration programs, like ypao for yamaha, that would continuously adjust the lower octaves levels to sound best with whatever volume your playing at... that would be cool.
                              It's not how far you go, it's how go you far http://techtalk.parts-express.com/co...es/biggrin.gif

                              Comment

                              Working...
                              X