Normal Phase & Inverse Phase And PA +/-

[SeVeReD]

At the moment I've only tried PA minus and normal phase.  Do I remember reading something about switching to PA plus if I switch to inverse phase?

Also, as stated, I've been using PA minus with normal phase and all sounds well.  Should I try PA plus with normal phase just to listen to the difference?

[PeterSt]

Do I remember reading something about switching to PA plus if I switch to inverse phase?

Yes. Officially the both should be anti-poles. So, Absolute Phase = Plus goes along with Phase Alignment = Minus. And, Absolute Phase = Minus (= Inverted) goes along with Phase Alignment = Plus.

Should I try PA plus with normal phase just to listen to the difference?

No. It is merely that I myself can't oversee the effect of something being inverted in your chain, you still use Absloute Phase (Plus) which thus turns into Inverted Phase somewhere (Minus) and now how to "counteract" that. Should be with Ph=Plus. But, this is because you just *have* it (inverted) like that. The real solution would be Invert the Phase (minus) which is to go along with Ph=Plus).

See ? I can't follow that.

And then there is the phenomenon of "at a certain Q1 needing to invert the phase" (a bit of your subject Dave). Here it becomes dangerous because we can only guess why that happens (but again this is about "alignment"). And since this is a non-discrete phenomenon (merely a natural one born from physics) I really can't predict how Phase Alignment has to work upon that.
Ask me 10 years from now ...

Peter

[SeVeReD]

And then there is the phenomenon of "at a certain Q1 needing to invert the phase" (a bit of your subject Dave). Here it becomes dangerous because we can only guess why that happens (but again this is about "alignment"). And since this is a non-discrete phenomenon (merely a natural one born from physics) I really can't predict how Phase Alignment has to work upon that.
Ask me 10 years from now ...

Peter

Hahaha But yes,,, our stereos really are about being phase sure to get the best sound stage/SQ out of them... the closer XXHE gets to doing that the better it sounds.  I need to sleep, but maybe soon I'll list the settings I'm using to get there... but at this time they involve Special, +1 PA intensity, minimize,,,, many other settings, and not  touching anything (vol control or anything), shutting off the amps before the last song stops.... to listen to the best and not hurt the system.. later Peter, but it is quite something if you're willing to jump through the hoops.  Thanks,,, and ah
not recommended to anyone but the mentally insane.

[phantomax]

Officially the both should be anti-poles. So, Absolute Phase = Plus goes along with Phase Alignment = Minus. And, Absolute Phase = Minus (= Inverted) goes along with Phase Alignment = Plus.

With Z7-3 I could play all the four combinations without problem. But now I can't do it: playing non invert phase with PA plus ends with a fading distorted sound and a loud crack after a couple of minutes.

Maxi

[PeterSt]

Maxi,

With Z7-3 I could play all the four combinations without problem.

I hope it won't come across as strange to you, but I say : impossible. And when I am right on this, it can only mean that before you used a situation where PA actually did not do a thing at all. Native Hires is the example.

I don't say it is exactly this, but that fact that you could use (realy engaged) all combinations before is impossible.

Generally : increase your Q1xQ1FxDevBufSize total.
Does that help ?

Regards,
Peter

[phantomax]

Hi Peter,

I am playing the two versions with the same settings and the conclusions are:

  - With 9Z 7-4 I can't play with IPhase on whatever the   PA is + or -. With PA- the problem is the related before: after a while the sound distorts and vanishes. With PA+ the following message appears: "24 bit second stage crack detect - 78174". With IPhase off  all seems to be OK with both PA+ and  PA-.

  - Re checked 9Z 7-3 and I have none of the above mentioned problems. And all signs suggests that PA is working properly.

  I haven't tried increase Q1xQ1FxDevBufSize total yet because I had a narrow margin to avoid plops .

Best regards

Maxi

[PeterSt]

Hey Maxi,

Don't ever think I don't believe you, but I don't understand one bit of it. You must be playing under conditions I just did not capture and checked (so mind you, all which now is said to work has been the most extensively checked, after first creating test data which allowed for that checking).

That crack detect being triggered already is an indication that something is severely wrong.

ends with a fading distorted sound and a loud crack after a couple of minutes

I only now see this "couple of minutes". This indicates that you most probably use an SFS of under 8 and nothing stops at all (until ??).

So, please tell me :
- Your Attennuation level (without Normalization);
- The base material you play (like 16/44.1);
- SFS;
- Q1, Q1F, DevBufSize;
- Your DAC Needs setting.
- Your DAC Is setting.
- Your upsampling rate.

Please be careful not to only "think" about some situation, but that you precisely mention one consistent situation (say, one you are going to try right now to be certain).

If it is indeed your SFS which is under 8, then forget about all because then that is the culprit (mine). But check 8 or higher of course.

Regards,
Peter

[phantomax]

Hello Peter,
first of all when you say that is something wrong with my settings/system for sure you are (have to be) rigth. I was just pointing that weird situation.  I'd like to comment that with the same material and settings I think there were some inconsistency in the SQ, always very good but only sometimes magic.

You requested my actual settings:

- Attennuation level (without Normalization): mostly -9 because I have a gain problem not solved yet. But the problems persist with more attenuation.
- The base material you play: 16/44.1 mostly FLAC but also some WAV and mp3
- SFS: 60. I only could down to 4 with the actual Dev. buffer settings.
- Q1, Q1F, DevBufSize: 30, (x4), 1024. No plops at end with that settings but as I said before, I have a narrow margin there.   
- Your DAC Needs setting: 32 bits *
- Your DAC Is setting: 32 bits  192 Khz *
- Your upsampling rate: (x4) 176,4
 * I recognize I 've never been sure about that settings, especially about bits one.

Regards

Maxi

[PeterSt]

- Q1, Q1F, DevBufSize: 30, (x4), 1024. No plops at end with that settings but as I said before, I have a narrow margin there.

When ? before with 0.9z-7-3 ? or with 0.9z-7-4 ?


DAC Needs of 32 will be OK if it plays. DAC Is of 32 will be wrong when you have a 24 bit DAC. But not wrong to the sense that it matters anywhere. So, all OK here.

[phantomax]

- Q1, Q1F, DevBufSize: 30, (x4), 1024. No plops at end with that settings but as I said before, I have a narrow margin there.

When ? before with 0.9z-7-3 ? or with 0.9z-7-4 ?

With 0.9z-7-4 I haven't tried other than that.

[PeterSt]

Arrrghhh

Maxi, with that quote you imply that you don't have problems at that setting and with 0.9z-7-4. This is contrary to this whole forum topic in the first place ...

So ?

[phantomax]

Sorry Peter for bother with this but I have a lots of problems with 09.Z7-4 so I am going back to the safe side of 09.Z7-3. I am missing something probably because my low level english. I'll try it in the future after read carefully the questions involved.

Thank you anyway!

Best regards
Maxi

[phantomax]

Encouraged by the comments in the specific topic about the  Z7-4 SQ, I finally took the plunge to try it again, this time trough an old pair of headphones ( I was  so scared ). The question is  wathever the buffer size and SFS settings,  I can only play with IPhase off and PA- . All of the other three possibilities end after a while with  horrendous cracking noises.
I don't know because the new version itself improving or because, as Peter suggested, probably I had never PA working properly in my system, the question is that now the sound is SUPERB. The best so far.   

Maxi

[PeterSt]

Maxi,

Since you seem to be the only one reporting this, and including that you plainly tell that it doesn't work at all now ... are you sure you didn't mix up XXHighEnd.exe and Engine3.exe versions of different versions ?

Otherwise, can you, somehow, test with another D/A converter ? (could be mobo's spdif output)

Also, do you maybe have some other setting than the "normal" Arc Prediction and x times upsampling ? I must say it again, that "crack detect" tripping already is the indication of your cracking noises, but it means that the output of any conversion goes wrong. This *will* be because of something wrong on my side, but it can hardly be any normal setting people use for years by now ...

Let me know ...
Peter

[phantomax]

are you sure you didn't mix up XXHighEnd.exe and Engine3.exe versions of different versions ?

Hi Peter, I am not sure but I think not. Anyway, how I can check that?

Otherwise, can you, somehow, test with another D/A converter ? (could be mobo's spdif output)

I am using no usb or SPDIF because I am stealing the signal directly from balanced soundcard DAC chip legs (through  coupling caps) to a tube analog circuit that works as buffer and converts the balanced signal in SE. Perhaps this is the differential fact with "normal" DACs

I could try with  a Cambridge DacMagic via usb or SPDIF but I think it works only 16 bits 48 Khz via USB and 24 bits 96 Khz via SPDIF.

Maxi