Loading ...
Sorry, an error occurred while loading the content.
 

Re: Staffelt article

Expand Messages
  • plinque_fish
    Genelec make DSP pretty easy with their DSP speakers and Autocal software. Watch the Genelec demo online at youtube. http://www.youtube.com/watch?v=bVXdwGc6Dh0
    Message 1 of 22 , Sep 30, 2010
      Genelec make DSP pretty easy with their DSP speakers and Autocal software.
      Watch the Genelec demo online at youtube.
      http://www.youtube.com/watch?v=bVXdwGc6Dh0

      Joel

      --- In regsaudioforum@yahoogroups.com, Ken Holder <ken_holder@...> wrote:
      >
      > At 07:12 PM 9/29/2010, Robert wrote:
      >
      >
      > >I wonder how many of you have tried this. to take one set of
      > >speakers and then EQ another to sound as much alike as possible using pink
      > >noise(and perhaps measurements in the bass especially) and then
      > >comparing on music materials.
      >
      >
      > I've found I don't have the patience to fiddle with
      > third-octave EQ for more than, oh, 5 seconds before
      > going insane. But that's just me.
      >
      > Ken Holder
      > Just a Poor, Old, Simple, Country, Music-Lover
      >
    • Tom Mallin
      Jeff, as I recall REG s response to this issue of the Gradient experiment, he said they found a loudspeaker which was flat enough in frequency response without
      Message 2 of 22 , Oct 1, 2010
        Jeff, as I recall REG's response to this issue of the Gradient experiment, he said they found a loudspeaker which was flat enough in frequency response without EQ to meet the requirements of the test and show that nothing else mattered much. He said that Salmi said it was the original B&W 801. I think REG also said that he thought the Spendor SP 1/2 might qualify, given how little difference a DSP correction system makes in its tonal response, in his experience.



        >>> "jeff" <jeffstakehifi@...> 9/30/2010 11:45 PM >>>
        The Staffelt idea might fit with the Gradient results. How was the EQ done in the Gradient experiment?
        jeff stake

        --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@...> wrote:
        >
        >
        > I had to cut this short since someone
        > was waiting to talk to me. But
        > it is really interesting.
        > The point is that if tmbre is really
        > determined by a frequency response,
        > a single one, with one "degree of freedom",
        > or anything close to that, then one
        > should be able to make it whatever one
        > wants, in particular to make it "right"
        > whatever right means, with EQ --even one third
        > octave EQ and surely with DSP EQ.
        >
        > I know people here have done a good deal of experimenting
        > with this in terms of trying to make speakers
        > sound similar by "correcting" them with DSP devices
        > of an automatic sort.
        >
        > But my thought is that this is perhaps not quite the right thing
        > to do. Perhaps one should be instead working directly
        > with EQ to make the two different kinds of speakers
        > SOUND alike using, say , pink noise.
        >
        > Various people have mentioned(including me in TAS on occasion)
        > that because of the odd weightings used by the ear/brain
        > that this may lead to something different from the attempt
        > to match things by adjusting measurements using models of the
        > ear/brain to be the same. This latter depends on
        > the model's correctness. But EQing by listening might work
        > better.
        >
        > I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
        > noise(and perhaps measurements in the bass especially) and then
        > comparing on music materials.
        >
        > This is quite different from say a Toole type of evaluation where
        > one adjusts nothing but attempts to isolate factors. In the Toole
        > type experiment, it would be possible that the several factors
        > he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
        > he isolated as desirable simply acted in the setups he was using to get the total factor to be right.
        >
        > There would be no way to know this one way or the other except to try
        > adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
        > by a pre-existing model.
        >
        > Why not try? Could it be that timbre, which as I am sure we all
        > have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
        > Or at least could this be close to true?
        >
        > One thing that is worth noticing here a priori is that
        > when one EQs, even by small amounts, large timbre changes
        > happen. It is then something alarmingly like common sense
        > to suppose that if one got the RIGHT EQ, one might be able
        > to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
        > to mind that one might be able to move it to exactly where it ought to be!
        >
        > Could the audio industry have failed to notice this? Of course they
        > could! The ear/brain's threshold for hearing frequency response
        > changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
        > that it was simply overlooked, the extent to which micro EQ
        > could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
        > he measured a bunch of SOLID STATE amplifiers into real world
        > speaker loads and found their frequency response to vary a lot more
        > than the amount needed for them to be audibly distinguishable.
        > Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
        > was and is to what extent all the other described things
        > were actually matters of frequency response heard and described
        > as other things!)
        >
        > REG
        >
        > --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@> wrote:
        > >
        > >
        > > I have been reading an article from 1984 by
        > > Henrik Staffelt from Danish Technical University(where
        > > I used to work in the summers).
        > > I have not finished it but the essential
        > > point is that one third octave steady state
        > > at the ear canal entrance tells the story on
        > > timbre.
        > > If so, one ought to be able to get things right
        > > by EQ....more on this later
        > >
        > > REG
        > >
        >




        ------------------------------------

        Yahoo! Groups Links
      • jeff
        Thanks Tom. I had not remembered that there was no EQ involved. best jeff
        Message 3 of 22 , Oct 1, 2010
          Thanks Tom.
          I had not remembered that there was no EQ involved.
          best
          jeff

          --- In regsaudioforum@yahoogroups.com, "Tom Mallin" <tmallin@...> wrote:
          >
          > Jeff, as I recall REG's response to this issue of the Gradient experiment, he said they found a loudspeaker which was flat enough in frequency response without EQ to meet the requirements of the test and show that nothing else mattered much. He said that Salmi said it was the original B&W 801. I think REG also said that he thought the Spendor SP 1/2 might qualify, given how little difference a DSP correction system makes in its tonal response, in his experience.
          >
          >
          >
          > >>> "jeff" <jeffstakehifi@...> 9/30/2010 11:45 PM >>>
          > The Staffelt idea might fit with the Gradient results. How was the EQ done in the Gradient experiment?
          > jeff stake
          >
          > --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@> wrote:
          > >
          > >
          > > I had to cut this short since someone
          > > was waiting to talk to me. But
          > > it is really interesting.
          > > The point is that if tmbre is really
          > > determined by a frequency response,
          > > a single one, with one "degree of freedom",
          > > or anything close to that, then one
          > > should be able to make it whatever one
          > > wants, in particular to make it "right"
          > > whatever right means, with EQ --even one third
          > > octave EQ and surely with DSP EQ.
          > >
          > > I know people here have done a good deal of experimenting
          > > with this in terms of trying to make speakers
          > > sound similar by "correcting" them with DSP devices
          > > of an automatic sort.
          > >
          > > But my thought is that this is perhaps not quite the right thing
          > > to do. Perhaps one should be instead working directly
          > > with EQ to make the two different kinds of speakers
          > > SOUND alike using, say , pink noise.
          > >
          > > Various people have mentioned(including me in TAS on occasion)
          > > that because of the odd weightings used by the ear/brain
          > > that this may lead to something different from the attempt
          > > to match things by adjusting measurements using models of the
          > > ear/brain to be the same. This latter depends on
          > > the model's correctness. But EQing by listening might work
          > > better.
          > >
          > > I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
          > > noise(and perhaps measurements in the bass especially) and then
          > > comparing on music materials.
          > >
          > > This is quite different from say a Toole type of evaluation where
          > > one adjusts nothing but attempts to isolate factors. In the Toole
          > > type experiment, it would be possible that the several factors
          > > he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
          > > he isolated as desirable simply acted in the setups he was using to get the total factor to be right.
          > >
          > > There would be no way to know this one way or the other except to try
          > > adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
          > > by a pre-existing model.
          > >
          > > Why not try? Could it be that timbre, which as I am sure we all
          > > have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
          > > Or at least could this be close to true?
          > >
          > > One thing that is worth noticing here a priori is that
          > > when one EQs, even by small amounts, large timbre changes
          > > happen. It is then something alarmingly like common sense
          > > to suppose that if one got the RIGHT EQ, one might be able
          > > to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
          > > to mind that one might be able to move it to exactly where it ought to be!
          > >
          > > Could the audio industry have failed to notice this? Of course they
          > > could! The ear/brain's threshold for hearing frequency response
          > > changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
          > > that it was simply overlooked, the extent to which micro EQ
          > > could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
          > > he measured a bunch of SOLID STATE amplifiers into real world
          > > speaker loads and found their frequency response to vary a lot more
          > > than the amount needed for them to be audibly distinguishable.
          > > Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
          > > was and is to what extent all the other described things
          > > were actually matters of frequency response heard and described
          > > as other things!)
          > >
          > > REG
          > >
          > > --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@> wrote:
          > > >
          > > >
          > > > I have been reading an article from 1984 by
          > > > Henrik Staffelt from Danish Technical University(where
          > > > I used to work in the summers).
          > > > I have not finished it but the essential
          > > > point is that one third octave steady state
          > > > at the ear canal entrance tells the story on
          > > > timbre.
          > > > If so, one ought to be able to get things right
          > > > by EQ....more on this later
          > > >
          > > > REG
          > > >
          > >
          >
          >
          >
          >
          > ------------------------------------
          >
          > Yahoo! Groups Links
          >
        • ymm
          REG didin t think much of the B W 801 when he listened to them in his rooms. Yip
          Message 4 of 22 , Oct 1, 2010
            REG didin't think much of the B W 801 when he listened to them in his rooms.

            Yip

            From: Tom Mallin <tmallin@...>
            To: regsaudioforum@yahoogroups.com
            Sent: Friday, 1 October 2010 18:39:24
            Subject: [regsaudioforum] Re: Staffelt article

             

            Jeff, as I recall REG's response to this issue of the Gradient experiment, he said they found a loudspeaker which was flat enough in frequency response without EQ to meet the requirements of the test and show that nothing else mattered much. He said that Salmi said it was the original B&W 801. I think REG also said that he thought the Spendor SP 1/2 might qualify, given how little difference a DSP correction system makes in its tonal response, in his experience.

            >>> "jeff" <jeffstakehifi@...> 9/30/2010 11:45 PM >>>
            The Staffelt idea might fit with the Gradient results. How was the EQ done in the Gradient experiment?
            jeff stake

            --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@...> wrote:
            >
            >
            > I had to cut this short since someone
            > was waiting to talk to me. But
            > it is really interesting.
            > The point is that if tmbre is really
            > determined by a frequency response,
            > a single one, with one "degree of freedom",
            > or anything close to that, then one
            > should be able to make it whatever one
            > wants, in particular to make it "right"
            > whatever right means, with EQ --even one third
            > octave EQ and surely with DSP EQ.
            >
            > I know people here have done a good deal of experimenting
            > with this in terms of trying to make speakers
            > sound similar by "correcting" them with DSP devices
            > of an automatic sort.
            >
            > But my thought is that this is perhaps not quite the right thing
            > to do. Perhaps one should be instead working directly
            > with EQ to make the two different kinds of speakers
            > SOUND alike using, say , pink noise.
            >
            > Various people have mentioned(including me in TAS on occasion)
            > that because of the odd weightings used by the ear/brain
            > that this may lead to something different from the attempt
            > to match things by adjusting measurements using models of the
            > ear/brain to be the same. This latter depends on
            > the model's correctness. But EQing by listening might work
            > better.
            >
            > I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
            > noise(and perhaps measurements in the bass especially) and then
            > comparing on music materials.
            >
            > This is quite different from say a Toole type of evaluation where
            > one adjusts nothing but attempts to isolate factors. In the Toole
            > type experiment, it would be possible that the several factors
            > he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
            > he isolated as desirable simply acted in the setups he was using to get the total factor to be right.
            >
            > There would be no way to know this one way or the other except to try
            > adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
            > by a pre-existing model.
            >
            > Why not try? Could it be that timbre, which as I am sure we all
            > have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
            > Or at least could this be close to true?
            >
            > One thing that is worth noticing here a priori is that
            > when one EQs, even by small amounts, large timbre changes
            > happen. It is then something alarmingly like common sense
            > to suppose that if one got the RIGHT EQ, one might be able
            > to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
            > to mind that one might be able to move it to exactly where it ought to be!
            >
            > Could the audio industry have failed to notice this? Of course they
            > could! The ear/brain's threshold for hearing frequency response
            > changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
            > that it was simply overlooked, the extent to which micro EQ
            > could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
            > he measured a bunch of SOLID STATE amplifiers into real world
            > speaker loads and found their frequency response to vary a lot more
            > than the amount needed for them to be audibly distinguishable.
            > Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
            > was and is to what extent all the other described things
            > were actually matters of frequency response heard and described
            > as other things!)
            >
            > REG
            >
            > --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@> wrote:
            > >
            > >
            > > I have been reading an article from 1984 by
            > > Henrik Staffelt from Danish Technical University(where
            > > I used to work in the summers).
            > > I have not finished it but the essential
            > > point is that one third octave steady state
            > > at the ear canal entrance tells the story on
            > > timbre.
            > > If so, one ought to be able to get things right
            > > by EQ....more on this later
            > >
            > > REG
            > >
            >

            ------------------------------------

            Yahoo! Groups Links


          • robert jorgensen
            Note that B&W 801 has a very irregular dispersion pattern. Greetings from Brussels Robert
            Message 5 of 22 , Oct 1, 2010
              Note that B&W 801 has a very irregular dispersion pattern.

              Greetings from Brussels

              Robert

              On Fri, Oct 1, 2010 at 5:20 PM, ymm <yipmangmeng@...> wrote:


              REG didin't think much of the B W 801 when he listened to them in his rooms.

              Yip
            • Tom Mallin
              Which will count in a listening room, but not in an anechoic situation such as in the Gradient experiment. ... Note that B&W 801 has a very irregular
              Message 6 of 22 , Oct 1, 2010
                Which will count in a listening room, but not in an anechoic situation such as in the Gradient experiment.

                >>> robert jorgensen <robert.jorgensen@...> 10/1/2010 11:07 AM >>>
                Note that B&W 801 has a very irregular dispersion pattern.

                Greetings from Brussels

                Robert

                On Fri, Oct 1, 2010 at 5:20 PM, ymm <yipmangmeng@...> wrote:


                REG didin't think much of the B W 801 when he listened to them in his rooms.

                Yip
              • Robert
                I think that the crucial thing here is that he claims that he has isolated WHAT DETERMINES THE PERCEIVED TIMBRE. And it is ONE DIMENSIONAL--it is a single
                Message 7 of 22 , Oct 2, 2010
                  I think that the crucial thing here is that
                  he claims that he has isolated WHAT DETERMINES THE PERCEIVED TIMBRE.

                  And it is ONE DIMENSIONAL--it is a single curve.

                  This means you can make it anything you like with EQ. He says this explicitly--that one third octave EQ is enough if you do it right.

                  I think it is important to understand here that what is RIGHT is secondary. Once you know what counts, then you can decide separately
                  what is right. Actually he has some ideas on what sounds flat ,too.

                  But only old fashioned audio people who are still thinking of buying equipment with fixed frequency response need to worry about what is right.

                  If it is really true that one can isolate a single curve that determines timbre,than one is in business.

                  I have of course been trying to persuade everyone to find this out for themselves for what seems like years. Namely I have been urging people to experiment with using 1/3 octave EQ to see how similar
                  two different speakers can be made to sound!! I do not think too
                  many of you have tried this very carefully. But this article ought to encourage you to do so.

                  Another feature of this article is that it DIRECTLY contradicts
                  the current prevailing belief that the ear's perception of timbre
                  is based on time windows that decrease in the higher frequencies.
                  He says this is WRONG. It is the steady state measurement that counts.

                  This does NOT mean that the steady state should measure flat(he talkes about how the amount of diffuse field is important and one has to roll down the top if there is more diffuse field to get the same listening result as if there were more direct and less diffuse field because of the ear's greater response to highs in diffuse field sound, as we have often discussed here).

                  I have in all honesty always been a little sceptical of the tendency of people in psychoacoustics to take the locational precedence effect and make it apply to timbre as well. This is so clearly wrong in concert halls. The timbre at distance is a LOT different from the direct arrival, even in the top where the detection time window is short. I think this tendency to over-rate direct arrival arises from the combination of the obsession of psychoacoustics people with localization(because it is so easy to do experiments on!) together with a general failure of nonmusicians to understand that timbre is essentially the ONLY thing about music that matters(except dynamics range but that is automatically ok in most audio today).
                  Psychoacoustic people (with some exceptions) are inclined to like tests of things that do not involve timbre perception because it is so indefinite. One can say when it matches, but how to specify it otherwise...Zwicker worked a lot on quantifying various words like "sharpness", but it is tough to get enough words to cover things.

                  So there is a lot to think about here. Meanwhile, really, anyone
                  interested in this--or in audio in general--ought to be
                  trying to make different speakers sound alike by listening and EQing.
                  Forget the automatic systems--they are all based on short time windows in the top. Just listen to pink noise, shifting back and forth between the two speakers(you have to match the volumes of course) and try to make it sound the same. The claim is that once this is done, music will sound much the same on the two speaker, too.

                  I am not claiming science for my own experience here, because I have mostly done this in circumstances where the direct arrival is predominant so that one is not really detecting whether the room sound and the direct arrival are separated. But my experience is that one can get the speakers quite close!

                  How does this fit in with Toole-sim? Just fine. The point of Toole-ism is that the early side wall reflection mixes in and is part of the steady state response at the ear. He never discusses whether one can adjust for this because he never adjusts things by EQ. (He was
                  really only working to try to find guidelines for how speakers should
                  be made to sell well--whjt no EQ allowed)

                  REG

                  --- In regsaudioforum@yahoogroups.com, "Tom Mallin" <tmallin@...> wrote:
                  >
                  > Yes, this is very interesting. But I'm wondering how one would know when the pink noise sounds "correct"? I'm sure one could micro-correct the response of speaker B to match the sound of speaker A when both A and B are playing pink noise and they are placed right next to each other. But how would you know if Speaker A is reproducing the sound correctly? Can we get nearer than just picking a speaker (like M40s) which long term experience with music playback convinces you is as correct in timbre as any other?
                  >
                  > >>> "Robert" <regonaudio@...> 9/29/2010 9:12 PM >>>
                  >
                  > I had to cut this short since someone
                  > was waiting to talk to me. But
                  > it is really interesting.
                  > The point is that if tmbre is really
                  > determined by a frequency response,
                  > a single one, with one "degree of freedom",
                  > or anything close to that, then one
                  > should be able to make it whatever one
                  > wants, in particular to make it "right"
                  > whatever right means, with EQ --even one third
                  > octave EQ and surely with DSP EQ.
                  >
                  > I know people here have done a good deal of experimenting
                  > with this in terms of trying to make speakers
                  > sound similar by "correcting" them with DSP devices
                  > of an automatic sort.
                  >
                  > But my thought is that this is perhaps not quite the right thing
                  > to do. Perhaps one should be instead working directly
                  > with EQ to make the two different kinds of speakers
                  > SOUND alike using, say , pink noise.
                  >
                  > Various people have mentioned(including me in TAS on occasion)
                  > that because of the odd weightings used by the ear/brain
                  > that this may lead to something different from the attempt
                  > to match things by adjusting measurements using models of the
                  > ear/brain to be the same. This latter depends on
                  > the model's correctness. But EQing by listening might work
                  > better.
                  >
                  > I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
                  > noise(and perhaps measurements in the bass especially) and then
                  > comparing on music materials.
                  >
                  > This is quite different from say a Toole type of evaluation where
                  > one adjusts nothing but attempts to isolate factors. In the Toole
                  > type experiment, it would be possible that the several factors
                  > he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
                  > he isolated as desirable simply acted in the setups he was using to get the total factor to be right.
                  >
                  > There would be no way to know this one way or the other except to try
                  > adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
                  > by a pre-existing model.
                  >
                  > Why not try? Could it be that timbre, which as I am sure we all
                  > have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
                  > Or at least could this be close to true?
                  >
                  > One thing that is worth noticing here a priori is that
                  > when one EQs, even by small amounts, large timbre changes
                  > happen. It is then something alarmingly like common sense
                  > to suppose that if one got the RIGHT EQ, one might be able
                  > to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
                  > to mind that one might be able to move it to exactly where it ought to be!
                  >
                  > Could the audio industry have failed to notice this? Of course they
                  > could! The ear/brain's threshold for hearing frequency response
                  > changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
                  > that it was simply overlooked, the extent to which micro EQ
                  > could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
                  > he measured a bunch of SOLID STATE amplifiers into real world
                  > speaker loads and found their frequency response to vary a lot more
                  > than the amount needed for them to be audibly distinguishable.
                  > Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
                  > was and is to what extent all the other described things
                  > were actually matters of frequency response heard and described
                  > as other things!)
                  >
                  > REG
                  >
                  > --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@> wrote:
                  > >
                  > >
                  > > I have been reading an article from 1984 by
                  > > Henrik Staffelt from Danish Technical University(where
                  > > I used to work in the summers).
                  > > I have not finished it but the essential
                  > > point is that one third octave steady state
                  > > at the ear canal entrance tells the story on
                  > > timbre.
                  > > If so, one ought to be able to get things right
                  > > by EQ....more on this later
                  > >
                  > > REG
                  > >
                  >
                  >
                  >
                  >
                  > ------------------------------------
                  >
                  > Yahoo! Groups Links
                  >
                • Robert
                  Their pattern is irregular. But it did not work on whether I could EQ away the effect (in a given room) of the hole in the power response at the top of the
                  Message 8 of 22 , Oct 2, 2010
                    Their pattern is irregular. But it did not work on whether
                    I could EQ away the effect (in a given room)
                    of the hole in the power response at the top of the midrange
                    driver.

                    I learn things too! It took me a long time to liberate
                    myself completely from the idea that one ought to be adjusting things,
                    not accepting them as they are.

                    REG
                    --- In regsaudioforum@yahoogroups.com, "Tom Mallin" <tmallin@...> wrote:
                    >
                    > Which will count in a listening room, but not in an anechoic situation such as in the Gradient experiment.
                    >
                    > >>> robert jorgensen <robert.jorgensen@...> 10/1/2010 11:07 AM >>>
                    >
                    >
                    > Note that B&W 801 has a very irregular dispersion pattern.
                    >
                    > Greetings from Brussels
                    >
                    > Robert
                    >
                    >
                    > On Fri, Oct 1, 2010 at 5:20 PM, ymm <yipmangmeng@...> wrote:
                    >
                    >
                    >
                    >
                    > REG didin't think much of the B W 801 when he listened to them in his rooms.
                    >
                    >
                    > Yip
                    >
                  • Ken Holder
                    ... Is this article on-line somewhere? This is fascinating. I ve always thought sound reproduction took a big step backwards when tone controls disappeared
                    Message 9 of 22 , Oct 3, 2010
                      At 09:16 PM 10/2/2010, Robert wrote:

                      >But only old fashioned audio people who are still thinking of buying
                      >equipment with fixed frequency response need to worry about what is right.


                      Is this article on-line somewhere?

                      This is fascinating. I've always thought sound reproduction
                      took a big step backwards when tone controls disappeared
                      from preamps and band-level controls vanished from the backs
                      of speakers.

                      I don't have the patience to fiddle with .33/8v EQ but I
                      am going to try (((Acourate)))'s automatic adjustments
                      on my old pair of Realistic Minimus 7s here sometime soon.
                      (I've been sick all summer and haven't done much except
                      just listen with amazement at the improvement Uli's
                      software made to my main system -- just plain
                      astonishing, I'd say! And pleasing too, since even
                      not-so-great recordings tend to sound better (as in
                      "more natural").)

                      The TAS-inspired obsession with imaging for the last
                      20 or 30 years seems to have taken away the concern
                      with timbre, which -- as you say -- is THE MAIN THING
                      that really defines "high fidelity" (once noise,
                      distortion, and therefore dynamic range is taken
                      care of).

                      Dang, we live in exciting times, eh!

                      Ken Holder
                      Just a Poor, Old, Simple, Country, Music-Lover
                    • Richard Tuck
                      Hi EMI used to use the 801s in its listening, mixing and mastering rooms and I got to hear them in what must be ideal conditions. My impression compared to
                      Message 10 of 22 , Oct 5, 2010

                        Hi

                        EMI used to use the 801s in its listening, mixing and mastering rooms and I got to hear them in what must be ideal conditions.  My impression compared to the Rogers LS7s I had at the time was that they sounded a bit rough.  As opposed to transparent and sweet but for pop monitoring they could sink lots of power without blowing up.  It always amazed me that, considering all the R&D they did, they produced a rather poor product, people would use them at shows and you could tell them a mile away.

                        Richard

                         


                        From: regsaudioforum@yahoogroups.com [mailto:regsaudioforum@yahoogroups.com] On Behalf Of Tom Mallin
                        Sent: 01 October 2010 17:10
                        To: regsaudioforum@yahoogroups.com
                        Subject: Re: [regsaudioforum] Re: Staffelt article

                         

                        Which will count in a listening room, but not in an anechoic situation such as in the Gradient experiment.

                        >>> robert jorgensen <robert.jorgensen@...> 10/1/2010
                        11:07 AM >>>
                        Note that B&W 801 has a very irregular dispersion pattern.

                        Greetings from Brussels

                        Robert

                        On Fri, Oct 1, 2010 at 5:20 PM, ymm <yipmangmeng@...> wrote:

                         

                        REG didin't think much of the B W 801 when he listened to them in his rooms.

                         

                        Yip

                      • Richard Tuck
                        Hi Tom I ve never heard a description of pink noise but white noise is said to sound like heavy rain on long grass or steam escaping from a boiler. B&K made a
                        Message 11 of 22 , Oct 5, 2010

                          Hi Tom

                          I’ve never heard a description of pink noise but white noise is said to sound like heavy rain on long grass or steam escaping from a boiler.  B&K made a microphone calibrator which was a hopper filled with small ball bearings which fell onto a diaphragm, I don’t know how white or pink it was or if it could be scaled up to loudspeaker sound levels it might offer a reference.

                          Richard

                           


                          From: regsaudioforum@yahoogroups.com [mailto:regsaudioforum@yahoogroups.com] On Behalf Of Tom Mallin
                          Sent: 30 September 2010 03:18
                          To: regsaudioforum@yahoogroups.com
                          Subject: [regsaudioforum] Re: Staffelt article

                           



                          Yes, this is very interesting.  But I'm wondering how one would know when the pink noise sounds "correct"?  I'm sure one could micro-correct the response of speaker B to match the sound of speaker A when both A and B are playing pink noise and they are placed right next to each other.  But how would you know if Speaker A is reproducing the sound correctly?  Can we get nearer than just picking a speaker (like M40s) which long term experience with music playback convinces you is as correct in timbre as any other?

                          >>> "Robert" <regonaudio@...> 9/29/2010 9:12 PM >>>

                          I had to cut this short since someone
                          was waiting to talk to me. But
                          it is really interesting.
                          The point is that if tmbre is really
                          determined by a frequency response,
                          a single one, with one "degree of freedom",
                          or anything close to that, then one
                          should be able to make it whatever one
                          wants, in particular to make it "right"
                          whatever right means, with EQ --even one third
                          octave EQ and surely with DSP EQ.

                          I know people here have done a good deal of experimenting
                          with this in terms of trying to make speakers
                          sound similar by "correcting" them with DSP devices
                          of an automatic sort.

                          But my thought is that this is perhaps not quite the right thing
                          to do. Perhaps one should be instead working directly
                          with EQ to make the two different kinds of speakers
                          SOUND alike using, say , pink noise.

                          Various people have mentioned(including me in TAS on occasion)
                          that because of the odd weightings used by the ear/brain
                          that this may lead to something different from the attempt
                          to match things by adjusting measurements using models of the
                          ear/brain to be the same. This latter depends on
                          the model's correctness. But EQing by listening might work
                          better.

                          I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
                          noise(and perhaps measurements in the bass especially) and then
                          comparing on music materials.

                          This is quite different from say a Toole type of evaluation where
                          one adjusts nothing but attempts to isolate factors. In the Toole
                          type experiment, it would be possible that the several factors
                          he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
                          he isolated as desirable simply acted in the setups he was using to get the total factor to be right.

                          There would be no way to know this one way or the other except to try
                          adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
                          by a pre-existing model.

                          Why not try? Could it be that timbre, which as I am sure we all
                          have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
                          Or at least could this be close to true?

                          One thing that is worth noticing here a priori is that
                          when one EQs, even by small amounts, large timbre changes
                          happen. It is then something alarmingly like common sense
                          to suppose that if one got the RIGHT EQ, one might be able
                          to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
                          to mind that one might be able to move it to exactly where it ought to be!

                          Could the audio industry have failed to notice this? Of course they
                          could! The ear/brain's threshold for hearing frequency response
                          changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
                          that it was simply overlooked, the extent to which micro EQ
                          could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
                          he measured a bunch of SOLID STATE amplifiers into real world
                          speaker loads and found their frequency response to vary a lot more
                          than the amount needed for them to be audibly distinguishable.
                          Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
                          was and is to what extent all the other described things
                          were actually matters of frequency response heard and described
                          as other things!)

                          REG

                          --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@...> wrote:
                          >
                          >
                          > I have been reading an article from 1984 by
                          > Henrik Staffelt from Danish
                          Technical University (where
                          > I used to work in the summers).
                          > I have not finished it but the essential
                          > point is that one third octave steady state
                          > at the ear canal entrance tells the story on
                          > timbre.
                          > If so, one ought to be able to get things right
                          > by EQ....more on this later
                          >
                          > REG
                          >




                          ------------------------------------

                          Yahoo! Groups Links

                          <*> To visit your group on the web, go to:
                              http://groups.yahoo.com/group/regsaudioforum/

                          <*> Your email settings:
                              Individual Email | Traditional

                          <*> To change settings online go to:
                              http://groups.yahoo.com/group/regsaudioforum/join
                              (Yahoo! ID required)

                          <*> To change settings via email:
                              regsaudioforum-digest@yahoogroups.com
                              regsaudioforum-fullfeatured@yahoogroups.com

                          <*> To unsubscribe from this group, send an email to:
                              regsaudioforum-unsubscribe@yahoogroups.com

                          <*> Your use of Yahoo! Groups is subject to:
                              http://docs.yahoo.com/info/terms/


                        • Tom Mallin
                          I ve also long heard that such noise should sound like a shower running in the next room. With a little experience, one can hear peaks sounding like tones
                          Message 12 of 22 , Oct 5, 2010
                            I've also long heard that such noise should sound like a shower running in the next room.
                             
                            With a little experience, one can hear peaks sounding like tones sticking out through the noise.  The idea is to equalize so that the noise sounds uniform and none of these tones stick out.
                             
                            Dips are harder to hear, but, again, with some experience playing with something like 1/3-octave graphic EQ and creating suckouts of considerable depth, you can zero in on the sound of a lack of SPL at particular frequencies and EQ around that.
                             

                            >>> "Richard Tuck" <rtuck@...> 10/5/2010 5:33 AM >>>

                            Hi Tom

                            I’ve never heard a description of pink noise but white noise is said to sound like heavy rain on long grass or steam escaping from a boiler.  B&K made a microphone calibrator which was a hopper filled with small ball bearings which fell onto a diaphragm, I don’t know how white or pink it was or if it could be scaled up to loudspeaker sound levels it might offer a reference.

                            Richard

                             


                            From: regsaudioforum@yahoogroups.com [mailto:regsaudioforum@yahoogroups.com] On Behalf Of Tom Mallin
                            Sent: 30 September 2010 03:18
                            To: regsaudioforum@yahoogroups.com
                            Subject: [regsaudioforum] Re: Staffelt article

                             



                            Yes, this is very interesting.  But I'm wondering how one would know when the pink noise sounds "correct"?  I'm sure one could micro-correct the response of speaker B to match the sound of speaker A when both A and B are playing pink noise and they are placed right next to each other.  But how would you know if Speaker A is reproducing the sound correctly?  Can we get nearer than just picking a speaker (like M40s) which long term experience with music playback convinces you is as correct in timbre as any other?

                            >>> "Robert" <regonaudio@...> 9/29/2010 9:12 PM >>>

                            I had to cut this short since someone
                            was waiting to talk to me. But
                            it is really interesting.
                            The point is that if tmbre is really
                            determined by a frequency response,
                            a single one, with one "degree of freedom",
                            or anything close to that, then one
                            should be able to make it whatever one
                            wants, in particular to make it "right"
                            whatever right means, with EQ --even one third
                            octave EQ and surely with DSP EQ.

                            I know people here have done a good deal of experimenting
                            with this in terms of trying to make speakers
                            sound similar by "correcting" them with DSP devices
                            of an automatic sort.

                            But my thought is that this is perhaps not quite the right thing
                            to do. Perhaps one should be instead working directly
                            with EQ to make the two different kinds of speakers
                            SOUND alike using, say , pink noise.

                            Various people have mentioned(including me in TAS on occasion)
                            that because of the odd weightings used by the ear/brain
                            that this may lead to something different from the attempt
                            to match things by adjusting measurements using models of the
                            ear/brain to be the same. This latter depends on
                            the model's correctness. But EQing by listening might work
                            better.

                            I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
                            noise(and perhaps measurements in the bass especially) and then
                            comparing on music materials.

                            This is quite different from say a Toole type of evaluation where
                            one adjusts nothing but attempts to isolate factors. In the Toole
                            type experiment, it would be possible that the several factors
                            he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
                            he isolated as desirable simply acted in the setups he was using to get the total factor to be right.

                            There would be no way to know this one way or the other except to try
                            adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
                            by a pre-existing model.

                            Why not try? Could it be that timbre, which as I am sure we all
                            have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
                            Or at least could this be close to true?

                            One thing that is worth noticing here a priori is that
                            when one EQs, even by small amounts, large timbre changes
                            happen. It is then something alarmingly like common sense
                            to suppose that if one got the RIGHT EQ, one might be able
                            to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
                            to mind that one might be able to move it to exactly where it ought to be!

                            Could the audio industry have failed to notice this? Of course they
                            could! The ear/brain's threshold for hearing frequency response
                            changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
                            that it was simply overlooked, the extent to which micro EQ
                            could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
                            he measured a bunch of SOLID STATE amplifiers into real world
                            speaker loads and found their frequency response to vary a lot more
                            than the amount needed for them to be audibly distinguishable.
                            Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
                            was and is to what extent all the other described things
                            were actually matters of frequency response heard and described
                            as other things!)

                            REG

                            --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@...> wrote:
                            >
                            >
                            > I have been reading an article from 1984 by
                            > Henrik Staffelt from Danish Technical University (where
                            > I used to work in the summers).
                            > I have not finished it but the essential
                            > point is that one third octave steady state
                            > at the ear canal entrance tells the story on
                            > timbre.
                            > If so, one ought to be able to get things right
                            > by EQ....more on this later
                            >
                            > REG
                            >




                            ------------------------------------

                            Yahoo! Groups Links

                            <*> To visit your group on the web, go to:
                                http://groups.yahoo.com/group/regsaudioforum/

                            <*> Your email settings:
                                Individual Email | Traditional

                            <*> To change settings online go to:
                                http://groups.yahoo.com/group/regsaudioforum/join
                                (Yahoo! ID required)

                            <*> To change settings via email:
                                regsaudioforum-digest@yahoogroups.com
                                regsaudioforum-fullfeatured@yahoogroups.com

                            <*> To unsubscribe from this group, send an email to:
                                regsaudioforum-unsubscribe@yahoogroups.com

                            <*> Your use of Yahoo! Groups is subject to:
                                http://docs.yahoo.com/info/terms/


                          • laurie483000
                            Interesting, I heard big B&Ws (whether 801s or not can t remember) once or twice at shows a few years back and they sounded somewhat hard to my ears on the
                            Message 13 of 22 , Oct 5, 2010
                              Interesting, I heard big B&Ws (whether 801s or not can't remember) once or twice at shows a few years back and they sounded somewhat hard to my ears on the occasions I heard them. But I was inclined at the time to make allowances - usual caveats - program material quality unknown and room acoustics question marks. However, I also remember being impressed by some of their medium price models at the same shows.


                              Laurie


                              --- In regsaudioforum@yahoogroups.com, "Richard Tuck" <rtuck@...> wrote:
                              >
                              > Hi
                              >
                              > EMI used to use the 801s in its listening, mixing and mastering rooms and I
                              > got to hear them in what must be ideal conditions. My impression compared
                              > to the Rogers LS7s I had at the time was that they sounded a bit rough. As
                              > opposed to transparent and sweet but for pop monitoring they could sink lots
                              > of power without blowing up. It always amazed me that, considering all the
                              > R&D they did, they produced a rather poor product, people would use them at
                              > shows and you could tell them a mile away.
                              >
                              > Richard
                              >
                              >
                              >
                              > _____
                              >
                              > From: regsaudioforum@yahoogroups.com [mailto:regsaudioforum@yahoogroups.com]
                              > On Behalf Of Tom Mallin
                              > Sent: 01 October 2010 17:10
                              > To: regsaudioforum@yahoogroups.com
                              > Subject: Re: [regsaudioforum] Re: Staffelt article
                              >
                              >
                              >
                              > Which will count in a listening room, but not in an anechoic situation such
                              > as in the Gradient experiment.
                              >
                              > >>> robert jorgensen <robert.jorgensen@...> 10/1/2010 11:07 AM >>>
                              > Note that B&W 801 has a very irregular dispersion pattern.
                              >
                              > Greetings from Brussels
                              >
                              > Robert
                              >
                              > On Fri, Oct 1, 2010 at 5:20 PM, ymm <yipmangmeng@...> wrote:
                              >
                              >
                              >
                              > REG didin't think much of the B W 801 when he listened to them in his rooms.
                              >
                              >
                              >
                              > Yip
                              >
                            • Tom Mallin
                              If anyone is interested in going the 1/3-octave graphic equalizer route, you might try borrowing a Rane DEQ-60L. It is very easy to use and, as far as I know,
                              Message 14 of 22 , Oct 5, 2010
                                If anyone is interested in going the 1/3-octave graphic equalizer route, you might try borrowing a Rane DEQ-60L. It is very easy to use and, as far as I know, is the ONLY one which has so little interaction among adjacent bands that the measured modification of the frequency response looks like a graph of the way the sliders are arranged. What you see is what you get (WYSIWYG) in terms of altered response with the Rane, but not with any others. I have added back in to my photo album a couple of old pictures which show the difference between the "Perfect Q" technology Rane uses and the proportional and constant Q methods others use, in this case the Audient ASP 231. The two pictures show the two graphic equalizers with their sliders arranged to provide the same measured bass equalization of my (at that time Orion-based) system:

                                http://tinyurl.com/2v9x3fl
                                http://tinyurl.com/3xz7wh4

                                Ultimately I did not like the "sound" of the Rane, but I know that George Day and the person here who bought George's unit (Ted?) both think it is totally transparent in their systems. Yes, at about $900 new street price, it is one of the more expensive graphic EQs available (not as pricey as the Klark Teknik stuff, though), but it really is the easiest to use, I think.

                                For TacT owners/contemplators, I've suggested to Anthony/Boz that TacT add a dual channel 1/3-octave graphic EQ function with "Perfect Q" characteristics to the TacT software. Anthony says he likes the idea and will pursue it. The TacT currently offers dual channel 12-band parametric EQ, in addition to its customizable target curve methods of applying EQ.

                                Of course, for computer-based systems, I know there are free or inexpensive software offerings that can add 1/3-octave graphic EQ to such a system. Those using such systems may want to let us know their experiences. I would bet, however, that few if any of these offer the patented "Perfect Q" technology to allow WYSIWYG adjustment like the Rane. Here's more about "Perfect Q" from Rane's tech note:

                                http://www.rane.com/note154.html
                              • Robert
                                Of course any kind of verbal description is WAY too insensitive to cover the possibilities one by one. This is the inevitable limitation of review by writing
                                Message 15 of 22 , Oct 5, 2010
                                  Of course any kind of verbal description is WAY too insensitive
                                  to cover the possibilities one by one.

                                  This is the inevitable limitation
                                  of review by writing or talking. One has to HEAR!(or look at a graph)

                                  Supposing that we are talking about
                                  +-3 dB things over 30 1/3 octave bands and the threshold for audible shifts is less than 0.2 dB --well, you can see that one one is looking
                                  at something enormous in the way of numbers of possibilities.
                                  Each 1/3 octave band(imagine one fixed in level so there are 29 left)
                                  can have any one of -3, -2.8,-2.6, 0, +0.2, +0.4...,+3
                                  levels within that +- 3 dB nominal window. So there are 15 plus levels, 15 minus levels, and the 0 level for 31 levels total in that particular band. And then there are 29 bands(one being normalized for volume). So there are in that +3 , -3 dB window (wait for it)
                                  31 to the 29th power possibilities! = approx 1.8 x 10 to the 43rd power!.

                                  Even in a +_ 1 dB window, there are 11 possbilities per band
                                  and hence 11 to the 29th power= approx 1.6 times 10 to the 30th This is clearly a lot more possibilities than one is going to be able to cover with words
                                  and surely with words that give any precise idea!

                                  The idea that this could be covered by verbal description was clearly
                                  not thought through very carefully!! "Dreamers, they're nothing but dreamers" if they think verbal reviewing will get it nailed down.

                                  REG

                                  --- In regsaudioforum@yahoogroups.com, "Tom Mallin" <tmallin@...> wrote:
                                  >
                                  > I've also long heard that such noise should sound like a shower running in the next room.
                                  >
                                  > With a little experience, one can hear peaks sounding like tones sticking out through the noise. The idea is to equalize so that the noise sounds uniform and none of these tones stick out.
                                  >
                                  > Dips are harder to hear, but, again, with some experience playing with something like 1/3-octave graphic EQ and creating suckouts of considerable depth, you can zero in on the sound of a lack of SPL at particular frequencies and EQ around that.
                                  >
                                  >
                                  > >>> "Richard Tuck" <rtuck@...> 10/5/2010 5:33 AM >>>
                                  >
                                  >
                                  >
                                  > Hi Tom
                                  > I’ve never heard a description of pink noise but white noise is said to sound like heavy rain on long grass or steam escaping from a boiler. B&K made a microphone calibrator which was a hopper filled with small ball bearings which fell onto a diaphragm, I don’t know how white or pink it was or if it could be scaled up to loudspeaker sound levels it might offer a reference.
                                  > Richard
                                  >
                                  >
                                  >
                                  >
                                  > From: regsaudioforum@yahoogroups.com [mailto:regsaudioforum@yahoogroups.com] On Behalf Of Tom Mallin
                                  > Sent: 30 September 2010 03:18
                                  > To: regsaudioforum@yahoogroups.com
                                  > Subject: [regsaudioforum] Re: Staffelt article
                                  >
                                  >
                                  >
                                  > Yes, this is very interesting. But I'm wondering how one would know when the pink noise sounds "correct"? I'm sure one could micro-correct the response of speaker B to match the sound of speaker A when both A and B are playing pink noise and they are placed right next to each other. But how would you know if Speaker A is reproducing the sound correctly? Can we get nearer than just picking a speaker (like M40s) which long term experience with music playback convinces you is as correct in timbre as any other?
                                  >
                                  > >>> "Robert" <regonaudio@...> 9/29/2010 9:12 PM >>>
                                  >
                                  > I had to cut this short since someone
                                  > was waiting to talk to me. But
                                  > it is really interesting.
                                  > The point is that if tmbre is really
                                  > determined by a frequency response,
                                  > a single one, with one "degree of freedom",
                                  > or anything close to that, then one
                                  > should be able to make it whatever one
                                  > wants, in particular to make it "right"
                                  > whatever right means, with EQ --even one third
                                  > octave EQ and surely with DSP EQ.
                                  >
                                  > I know people here have done a good deal of experimenting
                                  > with this in terms of trying to make speakers
                                  > sound similar by "correcting" them with DSP devices
                                  > of an automatic sort.
                                  >
                                  > But my thought is that this is perhaps not quite the right thing
                                  > to do. Perhaps one should be instead working directly
                                  > with EQ to make the two different kinds of speakers
                                  > SOUND alike using, say , pink noise.
                                  >
                                  > Various people have mentioned(including me in TAS on occasion)
                                  > that because of the odd weightings used by the ear/brain
                                  > that this may lead to something different from the attempt
                                  > to match things by adjusting measurements using models of the
                                  > ear/brain to be the same. This latter depends on
                                  > the model's correctness. But EQing by listening might work
                                  > better.
                                  >
                                  > I wonder how many of you have tried this. to take one set of speakers and then EQ another to sound as much alike as possible using pink
                                  > noise(and perhaps measurements in the bass especially) and then
                                  > comparing on music materials.
                                  >
                                  > This is quite different from say a Toole type of evaluation where
                                  > one adjusts nothing but attempts to isolate factors. In the Toole
                                  > type experiment, it would be possible that the several factors
                                  > he isolated(on axis, smooth off axis) in fact combined in the fact into a single factor. In more detail, it could be that the factors
                                  > he isolated as desirable simply acted in the setups he was using to get the total factor to be right.
                                  >
                                  > There would be no way to know this one way or the other except to try
                                  > adjusting one speaker to match another by adjustment and see to what extent this is possible. But one would need to do this by ear not
                                  > by a pre-existing model.
                                  >
                                  > Why not try? Could it be that timbre, which as I am sure we all
                                  > have observed dominates the scene(things like dynmaic linearity are really nonissues with modern equipment) is really a matter of a SINGLE adjustment--as long as one did the adjustment right?
                                  > Or at least could this be close to true?
                                  >
                                  > One thing that is worth noticing here a priori is that
                                  > when one EQs, even by small amounts, large timbre changes
                                  > happen. It is then something alarmingly like common sense
                                  > to suppose that if one got the RIGHT EQ, one might be able
                                  > to move the timbre to sounding essentially right. If one can move it so very very much--as one can--the possibility surely comes
                                  > to mind that one might be able to move it to exactly where it ought to be!
                                  >
                                  > Could the audio industry have failed to notice this? Of course they
                                  > could! The ear/brain's threshold for hearing frequency response
                                  > changes is so much smaller than the actual frequency response tolerance of any speaker whatever(there is no speaker that is +-0.1 dB flat without electronic adjustment) , it is quite easy to imagine
                                  > that it was simply overlooked, the extent to which micro EQ
                                  > could alter things to sound alike. (This was pointed out about amplifiers by Tom Holman in issue 26 of The Absolute Sound, where
                                  > he measured a bunch of SOLID STATE amplifiers into real world
                                  > speaker loads and found their frequency response to vary a lot more
                                  > than the amount needed for them to be audibly distinguishable.
                                  > Of course the reviewers had been carrying on about the observed sonic differences--and diffferences there no doubt were. The question
                                  > was and is to what extent all the other described things
                                  > were actually matters of frequency response heard and described
                                  > as other things!)
                                  >
                                  > REG
                                  >
                                  > --- In regsaudioforum@yahoogroups.com, "Robert" <regonaudio@> wrote:
                                  > >
                                  > >
                                  > > I have been reading an article from 1984 by
                                  > > Henrik Staffelt from Danish Technical University(where
                                  > > I used to work in the summers).
                                  > > I have not finished it but the essential
                                  > > point is that one third octave steady state
                                  > > at the ear canal entrance tells the story on
                                  > > timbre.
                                  > > If so, one ought to be able to get things right
                                  > > by EQ....more on this later
                                  > >
                                  > > REG
                                  > >
                                  >
                                  >
                                  >
                                  >
                                  > ------------------------------------
                                  >
                                  > Yahoo! Groups Links
                                  >
                                • Ted Rook
                                  Yes the Rane is about as easy as it gets with 1/3rd octave graphic EQ, I did buy Georges unit and have never regretted it, but there is something I have to add
                                  Message 16 of 22 , Oct 5, 2010
                                    Yes the Rane is about as easy as it gets with 1/3rd octave graphic EQ, I did buy Georges unit
                                    and have never regretted it, but there is something I have to add which is that one must use it
                                    with a PC based, or Mac based, on-screen software RTA package, I use TrueAudio's system
                                    that was a snip at $100 for 1/24th octave analyzer and full function generator.

                                    As a bonus the DEQ60L has two extras that increase its usefulness for me. There are high
                                    pass and low pass filters independent of the graphic 1/3rd octave, I find the high pass filter
                                    set at 30Hz does the job of keeping LP playback subsonic garbage out of my amp and
                                    speakers and suppressing subsonic resonance.

                                    The second bonus is a set of three sliders for simple overall EQ control of the whole
                                    spectrum, arranged as high low and mid sliders. Having used pink noise to EQ room effects
                                    out of the system (by use of cut only) I use these three sliders to apply tonal balance for my
                                    preference of the relative balance between the lows mids and highs, it varies a little over time
                                    but basically this is where I do the target slope adjustment, a slope down from the bass to the
                                    treble which I find most realistic and subjectively pleasing with most sources, especially the
                                    symphony orchestra. Having three sliders that control the whole spectrum makes it possible
                                    to experiment with the sound tonal balance in ways that aren't possible with the full set of 30
                                    sliders.

                                    Ted



                                    On 5 Oct 2010 at 11:45, Tom Mallin wrote:

                                    > If anyone is interested in going the 1/3-octave graphic equalizer
                                    > route, you might try borrowing a Rane DEQ-60L. It is very easy to use
                                    > and, as far as I know, is the ONLY one which has so little interaction
                                    > among adjacent bands that the measured modification of the frequency
                                    > response looks like a graph of the way the sliders are arranged. What
                                    > you see is what you get (WYSIWYG) in terms of altered response with
                                    > the Rane, but not with any others. I have added back in to my photo
                                    > album a couple of old pictures which show the difference between the
                                    > "Perfect Q" technology Rane uses and the proportional and constant Q
                                    > methods others use, in this case the Audient ASP 231. The two
                                    > pictures show the two graphic equalizers with their sliders arranged
                                    > to provide the same measured bass equalization of my (at that time
                                    > Orion-based) system:
                                    >
                                    > http://tinyurl.com/2v9x3fl
                                    > http://tinyurl.com/3xz7wh4
                                    >
                                    > Ultimately I did not like the "sound" of the Rane, but I know that
                                    > George Day and the person here who bought George's unit (Ted?) both
                                    > think it is totally transparent in their systems. Yes, at about $900
                                    > new street price, it is one of the more expensive graphic EQs
                                    > available (not as pricey as the Klark Teknik stuff, though), but it
                                    > really is the easiest to use, I think.
                                    >
                                    > For TacT owners/contemplators, I've suggested to Anthony/Boz that TacT
                                    > add a dual channel 1/3-octave graphic EQ function with "Perfect Q"
                                    > characteristics to the TacT software. Anthony says he likes the idea
                                    > and will pursue it. The TacT currently offers dual channel 12-band
                                    > parametric EQ, in addition to its customizable target curve methods of
                                    > applying EQ.
                                    >
                                    > Of course, for computer-based systems, I know there are free or
                                    > inexpensive software offerings that can add 1/3-octave graphic EQ to
                                    > such a system. Those using such systems may want to let us know their
                                    > experiences. I would bet, however, that few if any of these offer the
                                    > patented "Perfect Q" technology to allow WYSIWYG adjustment like the
                                    > Rane. Here's more about "Perfect Q" from Rane's tech note:
                                    >
                                    > http://www.rane.com/note154.html
                                    >
                                    >
                                    >
                                    >
                                    > ------------------------------------
                                    >
                                    > Yahoo! Groups Links
                                    >
                                    >
                                    >
                                  • Tom Mallin
                                    I agree with everything Ted says about the ease of use and functionality of the Rane DEQ-60L. The overall control of frequency balance of the three sliders
                                    Message 17 of 22 , Oct 6, 2010
                                      I agree with everything Ted says about the ease of use and functionality of the Rane DEQ-60L.  The overall control of frequency balance of the three sliders has an effect similar to that of the "Tilt" control present on some Quad preamps. 
                                       
                                      I disagree about the desirability of the low and high pass filters.  These filters are NOT defeatable.  The lowest bass setting is 15 Hz.  Thus, the unit always induces a bass rolloff below 20 Hz, even if you are using subwoofers.  Those of us who are not trying to reproduce vinyl recordings might rather have the correct low frequency phase response which affects the sound of bass higher up (cf., the KEF experiments REG has referred to).  If these filters were defeatable, I'd have no complaint about them, but as is, this is a minus for my system.
                                       
                                      You only need the measurement package if you want to measure the initial frequency response of your system and what the Rane is doing to modify that response.  I like to be able to do this as well, and I use Liberty's SynRTA to do this.
                                       
                                      But as REG is saying, we really aren't trying to achieve measured flatness.  We are trying to adjust the response to sound subjectively correct (or at least quite reasonable) for a given recording or set of recordings.  Subjective naturalness may not correspond to measured flatness or even measured smoothness (as in a frequency response which is a tilted straight line).  If this is what we are attempting to do, you don't really have to measure the response before or after at all.
                                       
                                      I do think, however, that if you are trying to use a 1/3-octave graphic EQ to subjectively adjust the response to sound correct or at least reasonable, it will help enormously to have a unit which, like the Rane, does not have a lot of interaction among adjacent (or even two away) sliders.  Making an adjustment in one band with the Rane does not adjust the response elsewhere.  With other units, there is quite of bit of "chasing your tail involved in the adjustment process as making an adjustment in one band significantly affects the response in at least two other bands.  That patented "Perfect Q" lack of band overlap is really helpful in zeroing in on a desired response characteristic.  At least that was my experience.

                                      >>> "Ted Rook" <rooknrol@...> 10/5/2010 4:18 PM >>>
                                      Yes the Rane is about as easy as it gets with 1/3rd octave graphic EQ, I did buy Georges unit
                                      and have never regretted it, but there is something I have to add which is that one must use it
                                      with a PC based, or Mac based, on-screen software RTA package, I use TrueAudio's system
                                      that was a snip at $100 for 1/24th octave analyzer and full function generator.

                                      As a bonus the DEQ60L has two extras that increase its usefulness for me. There are high
                                      pass and low pass filters independent of the graphic 1/3rd octave, I find the high pass filter
                                      set at 30Hz does the job of keeping LP playback subsonic garbage out of my amp and
                                      speakers and suppressing subsonic resonance.

                                      The second bonus is a set of three sliders for simple overall EQ control of the whole
                                      spectrum, arranged as high low and mid sliders. Having used pink noise to EQ room effects
                                      out of the system (by use of cut only) I use these three sliders to apply tonal balance for my
                                      preference of the relative balance between the lows mids and highs, it varies a little over time
                                      but basically this is where I do the target slope adjustment, a slope down from the bass to the
                                      treble which I find most realistic and subjectively pleasing with most sources, especially the
                                      symphony orchestra. Having three sliders that control the whole spectrum makes it possible
                                      to experiment with the sound tonal balance in ways that aren't possible with the full set of 30
                                      sliders.

                                      Ted



                                      On 5 Oct 2010 at 11:45, Tom Mallin wrote:

                                      > If anyone is interested in going the 1/3-octave graphic equalizer
                                      > route, you might try borrowing a Rane DEQ-60L.  It is very easy to use
                                      > and, as far as I know, is the ONLY one which has so little interaction
                                      > among adjacent bands that the measured modification of the frequency
                                      > response looks like a graph of the way the sliders are arranged.  What
                                      > you see is what you get (WYSIWYG) in terms of altered response with
                                      > the Rane, but not with any others.  I have added back in to my photo
                                      > album a couple of old pictures which show the difference between the
                                      > "Perfect Q" technology Rane uses and the proportional and constant Q
                                      > methods others use, in this case the Audient ASP 231.  The two
                                      > pictures show the two graphic equalizers with their sliders arranged
                                      > to provide the same measured bass equalization of my (at that time
                                      > Orion-based) system:
                                      >
                                      > http://tinyurl.com/2v9x3fl
                                      > http://tinyurl.com/3xz7wh4
                                      >
                                      > Ultimately I did not like the "sound" of the Rane, but I know that
                                      > George Day and the person here who bought George's unit (Ted?) both
                                      > think it is totally transparent in their systems.  Yes, at about $900
                                      > new street price, it is one of the more expensive graphic EQs
                                      > available (not as pricey as the Klark Teknik stuff, though), but it
                                      > really is the easiest to use, I think.
                                      >
                                      > For TacT owners/contemplators, I've suggested to Anthony/Boz that TacT
                                      > add a dual channel 1/3-octave graphic EQ function with "Perfect Q"
                                      > characteristics to the TacT software.  Anthony says he likes the idea
                                      > and will pursue it. The TacT currently offers dual channel 12-band
                                      > parametric EQ, in addition to its customizable target curve methods of
                                      > applying EQ.
                                      >
                                      > Of course, for computer-based systems, I know there are free or
                                      > inexpensive software offerings that can add 1/3-octave graphic EQ to
                                      > such a system.  Those using such systems may want to let us know their
                                      > experiences.  I would bet, however, that few if any of these offer the
                                      > patented "Perfect Q" technology to allow WYSIWYG adjustment like the
                                      > Rane.  Here's more about "Perfect Q" from Rane's tech note:
                                      >
                                      > http://www.rane.com/note154.html
                                      >
                                      >
                                      >
                                      >
                                      > ------------------------------------
                                      >
                                      > Yahoo! Groups Links
                                      >
                                      >
                                      >




                                      ------------------------------------

                                      Yahoo! Groups Links

                                      <*> To visit your group on the web, go to:
                                          http://groups.yahoo.com/group/regsaudioforum/

                                      <*> Your email settings:
                                          Individual Email | Traditional

                                      <*> To change settings online go to:
                                          http://groups.yahoo.com/group/regsaudioforum/join
                                          (Yahoo! ID required)

                                      <*> To change settings via email:
                                          regsaudioforum-digest@yahoogroups.com
                                          regsaudioforum-fullfeatured@yahoogroups.com

                                      <*> To unsubscribe from this group, send an email to:
                                          regsaudioforum-unsubscribe@yahoogroups.com

                                      <*> Your use of Yahoo! Groups is subject to:
                                          http://docs.yahoo.com/info/terms/

                                    Your message has been successfully submitted and would be delivered to recipients shortly.