The Bodzio Ultimate Equalizer

[soongsc]

Originally Posted by pawelp

There is one controversial aspect of phase correction which is to be precise non-minimumum phase correction. The only physical way it can be realised is to introduce a preresponse. And here comes a question if it is good given the fact that there is high end CD equipment where an upgrade is offered to make linear phase non-linear to get rid of a similar pre-response. The argument goes it "smears" the transients. I have brought it up on the SoundEasy forum but Bohdan Raczynski of SE for whatever reason did not like this rather interesting question.

The most important issue is "what is being equalized?".
My experience is that equalizing near field measurements generate more convincing playback. Equalization of 1M measurements result in more damped sound, but the image focus is a bit smeared. I have not tried equalizing in room response, but I would expect more damped sound and more smear of image focus. For this reason, I had asked whether UE3 will allow only equalizing low passed response, and the answer was YES.

[dlr]

Originally Posted by rdnjvn

The only thing that "smears" transients, is phase inaccuracies at one of two points, the source material, of which we audiophiles have no control over except what we choose to buy, and the sound at the listeners ears after the source material has been processed by the audio chain between the source and listeners ears.

Unless I miss my guess, he's referring to the impulse response, the time domain (hence the term used, "preresponse") , not the frequency domain as you are here. His reference to linear/non-linear phase may be confusing. The way I think it should be stated is linear phase vs. minimum phase (i.e. "non-linear"). The term non-linear isn't usually used in this context.

That and the term transient response is also referring to the impulse response, not music "transients". Different context, different aspect altogether.

If what he says is correct I get the impression that there is "correction" for the "correction". The goal being linear phase and its implementation not being perfect, someone else thinks that it introduces errors that need correcting, i.e., return the output back to minimum-phase in the frequency domain so that it "corrects" the problem in the time domain.

My assumption may be wrong, but that's how I read it at the moment. More detail of the "fix" would help, but then I'm not very well versed on details of DSP.

Dave

[dlr]

Originally Posted by soongsc

... I had asked whether UE3 will allow only equalizing low passed response, and the answer was YES.

All versions of the UE going back to the one in SoundEasy V17 EQ only what you specify. If all you do is enter a lowpass, that's what it equalizes. However, I don't believe that you could then use any of the phase linearization capability since that requires that both highpass and lowpass be specified in the target. I suppose, though, that it's possible to "unwrap" the excess-phase to the point that only the highpass phase remains. It seems to me that improving the phase response at the low end would be more beneficial given that we are more sensitive to phase at the low end.

Dave

[rdnjvn]

I guess it would be a good idea to define what I mean by phase distortion:
phase distortion or phase-frequency distortion is distortion that occurs when a filter's phase response is not linear over the frequency range of interest, that is, the phase shift introduced by a circuit or device is not directly proportional to frequency
Taken from here: Phase distortion - Wikipedia, the free encyclopedia

[soongsc]

Originally Posted by dlr

All versions of the UE going back to the one in SoundEasy V17 EQ only what you specify. If all you do is enter a lowpass, that's what it equalizes. However, I don't believe that you could then use any of the phase linearization capability since that requires that both highpass and lowpass be specified in the target. I suppose, though, that it's possible to "unwrap" the excess-phase to the point that only the highpass phase remains. It seems to me that improving the phase response at the low end would be more beneficial given that we are more sensitive to phase at the low end.

Dave

I think what I was addressing are separation of speaker equalization vs room equalization. Certainly want the phase to be flat, but in addition, in order to preserve the image focus, we want to be sure that what we are equalizing is not a combination response of waves with time difference exceeding one wave length.