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

Re: Acourate versus Tact

Expand Messages
  • Robert
    This is correct--it will time-align corner woofers(if they have indeed got this feature working correctly--cf TCDK s remarks in the archives about long ago).
    Message 1 of 11 , Mar 4, 2010
    • 0 Attachment
      This is correct--it will time-align corner woofers(if they
      have indeed got this feature working correctly--cf TCDK's
      remarks in the archives about long ago).
      But this is not at all the kind of phase correction I was
      discussing, which I am quite sure the tact does not do(unless it has been radically changed since I worked with it--I do not own one).
      Acourate does phase correction on the two channels in some considerable detail.
      UB can explain this far better than I, obviously.

      But believe me, it is different! Time coordinating woofers and main speakers is a far cry from doing detailed phase correction. Indeed the former is not really phase correction in the usual sense at all.
      It is just lining up the timing.


      REG

      --- In regsaudioforum@yahoogroups.com, "Martin Wax M.D." <mbw817@...> wrote:
      >
      > Robert,
      > Unless I am sadly uninformed, the TacT certainly does provide some key phase correction, as least when it is set up for using main speakers plus subs. Part of the set up uses impulse measurements so that the TacT can retard the signal from the mains which allows the sound to arrive at the listener the same time as it does for the subs, at least for the crossover frequency selected. For example, in my system, the mains are delayed 7.40 and 7.45 msec for the L and R speakers respectively compared to the subs. I have no idea if the Tact provides any phase correction when it is used to perform DSP EQ for full range speakers. Perhaps others familiar with the TacT can comment?
      >
      > It sure makes sense that phase correction would be desirable, and toward that end, I'd love to try Acourate. When it comes in a box with balanced ins and outs, I'm there. I've also heard that the DEQX does phase correction, yet we generally hear little about that unit. If anybody has familiarity with it, especially in comparison to a TacT or Acourate, please chime in.
      > Marty
      >
      >
      >
      > ________________________________
      > From: Robert <regonaudio@...>
      > To: regsaudioforum@yahoogroups.com
      > Sent: Wed, March 3, 2010 9:47:02 PM
      > Subject: [regsaudioforum] Acourate versus Tact
      >
      >
      >
      > I am sure UB will be willing to amplify on this, but it is important
      > to note that there is a fundamental difference.
      > Tact is an EQ system. Period (unless it has been changed).
      > It will get channels to match in frequency response in whatever sense
      > it is measuring frequency response(which one never really knows, I think).
      >
      > BUT in no sense will Tact make the channel match in terms of phase.
      >
      > Acourate however matches them up in phase as well as amplitude.
      > (It also phase linearizes each channel separately, but this is less important for what I am talking about).
      > Stereo imaging depends very much on phase differentials, below 1000 Hz anyway.
      >
      > So if the channels do not match in terms of phase, the imaging will be wrong even if they match in amplitude response in some sense or another.
      >
      > What is the effect of lack of phase matching? It makes things sound like spaced omnis, sort of indefinite and imprecise.
      >
      > Acourate will always (in my experience) give much better imaging, except in the rare instance where your room and speaker set up is EXACTLY symmetric. And almost no one's is!
      >
      > Tonally, the differences may not be large. But if imaging is your thing, Acourate will do better.
      >
      > If you want to see something surprising, try the test with Acourate where you are looking at the (fairly) early arrival with one speaker wired in reverse polarity and with playing a mono test signal.
      > The kind of cancellation that one gets from the Gradient Helsinki 1.5 without correction in the higher frequencies will go way down in the frequency range ,usually.
      >
      > Is this important to you? I don't know. But it is definitely real!
      > and definitely audible in terms of imaging especially.
      >
      > REG
      >
    • Uli Brueggemann
      It seems I need to jump into this discussion a bit :-) IMO it is quite simple but often simple things are difficult. Imagine a stupid RC lowpass filter. It
      Message 2 of 11 , Mar 4, 2010
      • 0 Attachment
        It seems I need to jump into this discussion a bit :-)

        IMO it is quite simple but often simple things are difficult.
        Imagine a stupid RC lowpass filter. It will show up a minphase behaviour for sure. Now it is already quite astonishing that the inverse filter of this lowpass --- also is of time minphase. So the convolution of two minphase filters, each one introducing phase errors,  can result in a perfect response without phase error.
        Great. A minphase filter corrects the phase response.
        And as TacT filters are minphase filters they correct the phase. Happy world.

        BUT:
        typically we use the LRC filters in the crossovers to separate a one-way signal into two, three or even more ways. Maybe a subwoofer, a woofer, a midrange driver and a tweeter. For each of the crossovers, typically of type minphase, we could create an individual minphase correction filter.
        This theoretical assumption does not work.
        Because the sound wave arriving at our ear is the sum of the individual driver signals. And surprisingly we have to acknowledge that the sum of the minphase signals is no longer of type minphase. Summing up the crossovers results in an excess phase behaviour. And if we apply just a minphase correction filter the excessphase (or allpass) behaviour remains.

        We all know the typical step response plots of speakers, tweeter first, then the midrange and with some delay the bass driver. See all the Stereophile speaker tests for example. A minphase correction filter can optimize the amplitude response but the typical step response will remain.
        You can easily test this by yourself. Run a TacT system and measure the final step response with an independant measurement system. You will see. A Lyngdorf RoomPerfect system even introduces more excessphase delays. (Do you really wonder why these systems do not allow to re-measure with active correction?)

        Furthermore in a room the reflections will also be added on to the direct signal. So finally we will get undesired phase responses, in worst case pretty different between the channels. A blurred image as a final result.

        The only chance to correct this is to apply also an excessphase correction beside the minphase correction. Nothing else.
        Also the typical TacT discussion about adjusting the delays between the drivers does not solve the problem. This may help a bit but it does not heal.

        Please also study my whitepaper http://www.acourate.com/XOWhitePaper.doc about the behaviour of crossover filters.

        Best, Uli




        On Thu, Mar 4, 2010 at 7:32 PM, Robert <regonaudio@...> wrote:
         

        This is correct--it will time-align corner woofers(if they
        have indeed got this feature working correctly--cf TCDK's
        remarks in the archives about long ago).
        But this is not at all the kind of phase correction I was
        discussing, which I am quite sure the tact does not do(unless it has been radically changed since I worked with it--I do not own one).
        Acourate does phase correction on the two channels in some considerable detail.
        UB can explain this far better than I, obviously.

        But believe me, it is different! Time coordinating woofers and main speakers is a far cry from doing detailed phase correction. Indeed the former is not really phase correction in the usual sense at all.
        It is just lining up the timing.

        REG




      • Uli Brueggemann
        oops, a typo ... On Thu, Mar 4, 2010 at 8:52 PM, Uli Brueggemann
        Message 3 of 11 , Mar 4, 2010
        • 0 Attachment
          oops, a typo
          --- also is of type minphase

          On Thu, Mar 4, 2010 at 8:52 PM, Uli Brueggemann <uli.brueggemann@...> wrote:
          It seems I need to jump into this discussion a bit :-)

          IMO it is quite simple but often simple things are difficult.
          Imagine a stupid RC lowpass filter. It will show up a minphase behaviour for sure. Now it is already quite astonishing that the inverse filter of this lowpass --- also is of time minphase. So the convolution of two minphase filters, each one introducing phase errors,  can result in a perfect response without phase error.
          Great. A minphase filter corrects the phase response.
          And as TacT filters are minphase filters they correct the phase. Happy world.

          BUT:
          typically we use the LRC filters in the crossovers to separate a one-way signal into two, three or even more ways. Maybe a subwoofer, a woofer, a midrange driver and a tweeter. For each of the crossovers, typically of type minphase, we could create an individual minphase correction filter.
          This theoretical assumption does not work.
          Because the sound wave arriving at our ear is the sum of the individual driver signals. And surprisingly we have to acknowledge that the sum of the minphase signals is no longer of type minphase. Summing up the crossovers results in an excess phase behaviour. And if we apply just a minphase correction filter the excessphase (or allpass) behaviour remains.

          We all know the typical step response plots of speakers, tweeter first, then the midrange and with some delay the bass driver. See all the Stereophile speaker tests for example. A minphase correction filter can optimize the amplitude response but the typical step response will remain.
          You can easily test this by yourself. Run a TacT system and measure the final step response with an independant measurement system. You will see. A Lyngdorf RoomPerfect system even introduces more excessphase delays. (Do you really wonder why these systems do not allow to re-measure with active correction?)

          Furthermore in a room the reflections will also be added on to the direct signal. So finally we will get undesired phase responses, in worst case pretty different between the channels. A blurred image as a final result.

          The only chance to correct this is to apply also an excessphase correction beside the minphase correction. Nothing else.
          Also the typical TacT discussion about adjusting the delays between the drivers does not solve the problem. This may help a bit but it does not heal.

          Please also study my whitepaper http://www.acourate.com/XOWhitePaper.doc about the behaviour of crossover filters.

          Best, Uli





          On Thu, Mar 4, 2010 at 7:32 PM, Robert <regonaudio@...> wrote:
           

          This is correct--it will time-align corner woofers(if they
          have indeed got this feature working correctly--cf TCDK's
          remarks in the archives about long ago).
          But this is not at all the kind of phase correction I was
          discussing, which I am quite sure the tact does not do(unless it has been radically changed since I worked with it--I do not own one).
          Acourate does phase correction on the two channels in some considerable detail.
          UB can explain this far better than I, obviously.

          But believe me, it is different! Time coordinating woofers and main speakers is a far cry from doing detailed phase correction. Indeed the former is not really phase correction in the usual sense at all.
          It is just lining up the timing.

          REG





        • Robert
          Perhaps I could add to clarify this. There is not much to suggest that the non-minimum phase behavior of speakers does much of anything audibly --as such. It
          Message 4 of 11 , Mar 4, 2010
          • 0 Attachment
            Perhaps I could add to clarify this.
            There is not much to suggest that the non-minimum phase behavior of speakers does much of anything audibly --as such. It does a little(alters sounds of transients mostly). But it does not do a lot and not much to imaging either--as such.

            What DOES do a whole lot to imaging is phase mismatching between the channels. This causes the imaging to become erratic and blurred.
            It is like having a little bit of the totally diffuse quality that arises when the speakers are in verse polarity one relative to the other.

            Of course it does not hurt to make the speakers linear phase,too--
            in principle that is better--but the BIG DEAL in imaging is the phase-match between the channels(which Acourate provides).

            REG

            PS No offense to UB or anyone else, but engineers are in love with minimum phase speakers in my view for pictorial reasons--the graphs of the acoustic output look like the acoustic input.

            But the ear is not the eye. There is a lot of evidence of a scientific sort(as well as the casual listening sort that occurs in audio magazines, where almost no speakers are minimum phase but apparently no one much cares one way or the other) that on music material the non-minimum phase nature of most speakers is all but inaudible. This has been tested many times. (It is easy with DSP since one can make an all pass filter easily that linearizes phase and switch it in and out).

            But the phase match between channels--that is another story.


            --- In regsaudioforum@yahoogroups.com, Uli Brueggemann <uli.brueggemann@...> wrote:
            >
            > It seems I need to jump into this discussion a bit :-)
            >
            > IMO it is quite simple but often simple things are difficult.
            > Imagine a stupid RC lowpass filter. It will show up a minphase behaviour for
            > sure. Now it is already quite astonishing that the inverse filter of this
            > lowpass --- also is of time minphase. So the convolution of two minphase
            > filters, each one introducing phase errors, can result in a perfect
            > response without phase error.
            > Great. A minphase filter corrects the phase response.
            > And as TacT filters are minphase filters they correct the phase. Happy
            > world.
            >
            > BUT:
            > typically we use the LRC filters in the crossovers to separate a one-way
            > signal into two, three or even more ways. Maybe a subwoofer, a woofer, a
            > midrange driver and a tweeter. For each of the crossovers, typically of type
            > minphase, we could create an individual minphase correction filter.
            > This theoretical assumption does not work.
            > Because the sound wave arriving at our ear is the sum of the individual
            > driver signals. And surprisingly we have to acknowledge that the sum of the
            > minphase signals is no longer of type minphase. Summing up the crossovers
            > results in an excess phase behaviour. And if we apply just a minphase
            > correction filter the excessphase (or allpass) behaviour remains.
            >
            > We all know the typical step response plots of speakers, tweeter first, then
            > the midrange and with some delay the bass driver. See all the Stereophile
            > speaker tests for example. A minphase correction filter can optimize the
            > amplitude response but the typical step response will remain.
            > You can easily test this by yourself. Run a TacT system and measure the
            > final step response with an independant measurement system. You will see. A
            > Lyngdorf RoomPerfect system even introduces more excessphase delays. (Do you
            > really wonder why these systems do not allow to re-measure with active
            > correction?)
            >
            > Furthermore in a room the reflections will also be added on to the direct
            > signal. So finally we will get undesired phase responses, in worst case
            > pretty different between the channels. A blurred image as a final result.
            >
            > The only chance to correct this is to apply also an excessphase correction
            > beside the minphase correction. Nothing else.
            > Also the typical TacT discussion about adjusting the delays between the
            > drivers does not solve the problem. This may help a bit but it does not
            > heal.
            >
            > Please also study my whitepaper
            > http://www.acourate.com/XOWhitePaper.docabout the behaviour of
            > crossover filters.
            >
            > Best, Uli
            >
            >
            >
            >
            > On Thu, Mar 4, 2010 at 7:32 PM, Robert <regonaudio@...> wrote:
            >
            > >
            > >
            > > This is correct--it will time-align corner woofers(if they
            > > have indeed got this feature working correctly--cf TCDK's
            > > remarks in the archives about long ago).
            > > But this is not at all the kind of phase correction I was
            > > discussing, which I am quite sure the tact does not do(unless it has been
            > > radically changed since I worked with it--I do not own one).
            > > Acourate does phase correction on the two channels in some considerable
            > > detail.
            > > UB can explain this far better than I, obviously.
            > >
            > > But believe me, it is different! Time coordinating woofers and main
            > > speakers is a far cry from doing detailed phase correction. Indeed the
            > > former is not really phase correction in the usual sense at all.
            > > It is just lining up the timing.
            > >
            > > REG
            > >
            > >
            > >
            >
          • Tom Mallin
            For the record, I have not heard an Acourate system so I have no reason to doubt that Acourate makes a system sound better than the TacT RCS does, as most if
            Message 5 of 11 , Mar 4, 2010
            • 0 Attachment
              For the record, I have not heard an Acourate system so I have no reason to doubt that Acourate makes a system sound better than the TacT RCS does, as most if not all of those who have heard both say.

              But I guess I don't understand, or just missed in the previous discussion, how Acourate and TacT differ in terms of the phase match between the two channels which REG says is the BIG DEAL. Both the TacT and Acourate time correct the arrival of signals of left and right speakers and subs. The TacT may not correct for the arrival time of main speaker sound versus subwoofer sound for that channel as correctly as Acourate does, but the TacT does the same processing for both channels. The TacT's lack of accounting for digital latency in the sub/main crossover also would not apply if the system just consists of left and right main speakers. The TacT and Acourate also both flatten the frequency response of left and right speakers to the same degree.

              If we discount Acourate's ability to make the speakers minimum phase, what is Acourate's advantage in terms of phase match between the two stereo channels?


              >>> "Robert" <regonaudio@...> 3/4/2010 3:21 PM >>>

              Perhaps I could add to clarify this.
              There is not much to suggest that the non-minimum phase behavior of speakers does much of anything audibly --as such. It does a little(alters sounds of transients mostly). But it does not do a lot and not much to imaging either--as such.

              What DOES do a whole lot to imaging is phase mismatching between the channels. This causes the imaging to become erratic and blurred.
              It is like having a little bit of the totally diffuse quality that arises when the speakers are in verse polarity one relative to the other.

              Of course it does not hurt to make the speakers linear phase,too--
              in principle that is better--but the BIG DEAL in imaging is the phase-match between the channels(which Acourate provides).

              REG

              PS No offense to UB or anyone else, but engineers are in love with minimum phase speakers in my view for pictorial reasons--the graphs of the acoustic output look like the acoustic input.

              But the ear is not the eye. There is a lot of evidence of a scientific sort(as well as the casual listening sort that occurs in audio magazines, where almost no speakers are minimum phase but apparently no one much cares one way or the other) that on music material the non-minimum phase nature of most speakers is all but inaudible. This has been tested many times. (It is easy with DSP since one can make an all pass filter easily that linearizes phase and switch it in and out).

              But the phase match between channels--that is another story.
            • yvl222
              Hi Uli, I must be lucky. Using 4th order crossovers implemented in my 2150 amps and time aligning them with a 1 Hz square wave, I could get a very good step
              Message 6 of 11 , Mar 6, 2010
              • 0 Attachment
                Hi Uli,

                I must be lucky. Using 4th order crossovers implemented in my 2150 amps and time aligning them with a 1 Hz square wave, I could get a very good step response with both the 2.2x and 2150s doing room correction calculations.

                Victor

                --- In regsaudioforum@yahoogroups.com, Uli Brueggemann <uli.brueggemann@...> wrote:
                >
                > It seems I need to jump into this discussion a bit :-)
                >
                > IMO it is quite simple but often simple things are difficult.
                > Imagine a stupid RC lowpass filter. It will show up a minphase behaviour for
                > sure. Now it is already quite astonishing that the inverse filter of this
                > lowpass --- also is of time minphase. So the convolution of two minphase
                > filters, each one introducing phase errors, can result in a perfect
                > response without phase error.
                > Great. A minphase filter corrects the phase response.
                > And as TacT filters are minphase filters they correct the phase. Happy
                > world.
                >
                > BUT:
                > typically we use the LRC filters in the crossovers to separate a one-way
                > signal into two, three or even more ways. Maybe a subwoofer, a woofer, a
                > midrange driver and a tweeter. For each of the crossovers, typically of type
                > minphase, we could create an individual minphase correction filter.
                > This theoretical assumption does not work.
                > Because the sound wave arriving at our ear is the sum of the individual
                > driver signals. And surprisingly we have to acknowledge that the sum of the
                > minphase signals is no longer of type minphase. Summing up the crossovers
                > results in an excess phase behaviour. And if we apply just a minphase
                > correction filter the excessphase (or allpass) behaviour remains.
                >
                > We all know the typical step response plots of speakers, tweeter first, then
                > the midrange and with some delay the bass driver. See all the Stereophile
                > speaker tests for example. A minphase correction filter can optimize the
                > amplitude response but the typical step response will remain.
                > You can easily test this by yourself. Run a TacT system and measure the
                > final step response with an independant measurement system. You will see. A
                > Lyngdorf RoomPerfect system even introduces more excessphase delays. (Do you
                > really wonder why these systems do not allow to re-measure with active
                > correction?)
                >
                > Furthermore in a room the reflections will also be added on to the direct
                > signal. So finally we will get undesired phase responses, in worst case
                > pretty different between the channels. A blurred image as a final result.
                >
                > The only chance to correct this is to apply also an excessphase correction
                > beside the minphase correction. Nothing else.
                > Also the typical TacT discussion about adjusting the delays between the
                > drivers does not solve the problem. This may help a bit but it does not
                > heal.
                >
                > Please also study my whitepaper
                > http://www.acourate.com/XOWhitePaper.docabout the behaviour of
                > crossover filters.
                >
                > Best, Uli
                >
                >
                >
                >
                > On Thu, Mar 4, 2010 at 7:32 PM, Robert <regonaudio@...> wrote:
                >
                > >
                > >
                > > This is correct--it will time-align corner woofers(if they
                > > have indeed got this feature working correctly--cf TCDK's
                > > remarks in the archives about long ago).
                > > But this is not at all the kind of phase correction I was
                > > discussing, which I am quite sure the tact does not do(unless it has been
                > > radically changed since I worked with it--I do not own one).
                > > Acourate does phase correction on the two channels in some considerable
                > > detail.
                > > UB can explain this far better than I, obviously.
                > >
                > > But believe me, it is different! Time coordinating woofers and main
                > > speakers is a far cry from doing detailed phase correction. Indeed the
                > > former is not really phase correction in the usual sense at all.
                > > It is just lining up the timing.
                > >
                > > REG
                > >
                > >
                > >
                >
              • Uli Brueggemann
                Victor, you can download the logsweep recorder program www.acourate.com/AcourateLSR2Setup.exe Then you can record the pulse responses of your system. Can you
                Message 7 of 11 , Mar 6, 2010
                • 0 Attachment
                  Victor,

                  you can download the logsweep recorder program www.acourate.com/AcourateLSR2Setup.exe
                  Then you can record the pulse responses of your system. Can you send them?

                  Uli

                  On Sat, Mar 6, 2010 at 9:32 AM, yvl222 <yvl222@...> wrote:
                   

                  Hi Uli,

                  I must be lucky. Using 4th order crossovers implemented in my 2150 amps and time aligning them with a 1 Hz square wave, I could get a very good step response with both the 2.2x and 2150s doing room correction calculations.

                  Victor



                  --- In regsaudioforum@yahoogroups.com, Uli Brueggemann <uli.brueggemann@...> wrote:
                  >
                  > It seems I need to jump into this discussion a bit :-)
                  >
                  > IMO it is quite simple but often simple things are difficult.
                  > Imagine a stupid RC lowpass filter. It will show up a minphase behaviour for
                  > sure. Now it is already quite astonishing that the inverse filter of this
                  > lowpass --- also is of time minphase. So the convolution of two minphase
                  > filters, each one introducing phase errors, can result in a perfect
                  > response without phase error.
                  > Great. A minphase filter corrects the phase response.
                  > And as TacT filters are minphase filters they correct the phase. Happy
                  > world.
                  >
                  > BUT:
                  > typically we use the LRC filters in the crossovers to separate a one-way
                  > signal into two, three or even more ways. Maybe a subwoofer, a woofer, a
                  > midrange driver and a tweeter. For each of the crossovers, typically of type
                  > minphase, we could create an individual minphase correction filter.
                  > This theoretical assumption does not work.
                  > Because the sound wave arriving at our ear is the sum of the individual
                  > driver signals. And surprisingly we have to acknowledge that the sum of the
                  > minphase signals is no longer of type minphase. Summing up the crossovers
                  > results in an excess phase behaviour. And if we apply just a minphase
                  > correction filter the excessphase (or allpass) behaviour remains.
                  >
                  > We all know the typical step response plots of speakers, tweeter first, then
                  > the midrange and with some delay the bass driver. See all the Stereophile
                  > speaker tests for example. A minphase correction filter can optimize the
                  > amplitude response but the typical step response will remain.
                  > You can easily test this by yourself. Run a TacT system and measure the
                  > final step response with an independant measurement system. You will see. A
                  > Lyngdorf RoomPerfect system even introduces more excessphase delays. (Do you
                  > really wonder why these systems do not allow to re-measure with active
                  > correction?)
                  >
                  > Furthermore in a room the reflections will also be added on to the direct
                  > signal. So finally we will get undesired phase responses, in worst case
                  > pretty different between the channels. A blurred image as a final result.
                  >
                  > The only chance to correct this is to apply also an excessphase correction
                  > beside the minphase correction. Nothing else.
                  > Also the typical TacT discussion about adjusting the delays between the
                  > drivers does not solve the problem. This may help a bit but it does not
                  > heal.
                  >
                  > Please also study my whitepaper
                  > http://www.acourate.com/XOWhitePaper.docabout the behaviour of
                  > crossover filters.
                  >
                  > Best, Uli

                  >
                  >
                  >
                  >
                  > On Thu, Mar 4, 2010 at 7:32 PM, Robert <regonaudio@...> wrote:
                  >
                  > >
                  > >
                  > > This is correct--it will time-align corner woofers(if they
                  > > have indeed got this feature working correctly--cf TCDK's
                  > > remarks in the archives about long ago).
                  > > But this is not at all the kind of phase correction I was
                  > > discussing, which I am quite sure the tact does not do(unless it has been
                  > > radically changed since I worked with it--I do not own one).
                  > > Acourate does phase correction on the two channels in some considerable
                  > > detail.
                  > > UB can explain this far better than I, obviously.
                  > >
                  > > But believe me, it is different! Time coordinating woofers and main
                  > > speakers is a far cry from doing detailed phase correction. Indeed the
                  > > former is not really phase correction in the usual sense at all.
                  > > It is just lining up the timing.
                  > >
                  > > REG
                  > >
                  > >
                  > >
                  >


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