General discussion about phase deletion

kensuguro · Post by **kensuguro** » Thu Sep 28, 2006 5:28 am

take the EQ of your HWmixer i guess you have one and try it out
i can do this with a soundcraft and a PEQ4 too!!
and i guess soundcraft dont uses CWAtoms

I'd still think it's hard to find such a close match tho.. Sure, I can see them cancelling out to a certain extent, but not this much. I just think such a minute difference is kind of cutting it close, esp with the whole vintage emulation blurb.

I mean, there's even more of a difference between samplitude's eq versus peq4.

But I do think this hair splitting is kind of pointless so, I'm done with it.

[ This Message was edited by: kensuguro on 2006-09-28 06:30 ]

MCCY · Post by **MCCY** » Wed Oct 04, 2006 10:42 am

I heard here in the forum the thesis, that if an EQ uses same modules as PARA4EQ it should be able to phasedelete it to -150db on SCOPE.

This hypothesis is interesting, cause it were mathmatical correct if one could really achieve the 100% right settings on both EQs with the knobs which SFP offers to us. As I mentioned in my project-thread it seems a problem, that scope changes values (sorry I had not the time to go deeper in it &haven't tested it with my devices so far) in certain cases from one start to the other of a SFP-project. So let's forget about that totally confusing problem for a moment.

I have an EQ which is built out of the above mentioned modules:

"PulsderZeit 8X +/- 24db EQ"

I won't ask you: Hey, does it sound like the PARA4EQ? Because it should do, because it is built that way.

I ask:
Is anybody here able to phasedelete it to -150 db, all bands on, settings on all bands. I didn't try it so far. Would be interesting point in this diescussion?

Martin

P.S. ah, I forgot. The device can be set with values. This should not be allowed. Values have to be dialed in with ignoring the panel

. I think this could avoid the problem of exactly hitting the right values under "SFP-conditions" with a non value labeled device.
Hm. I should modificate it a bit, so that it suits our needs for such a test. I will do that when I have more time (in some weeks). Don't know if I enabled presets for it. We have all the time in the world to discuss all our points here.

[ This Message was edited by: MCCYRANO on 2006-10-04 12:58 ]

djmicron · Post by **djmicron** » Wed Oct 04, 2006 1:13 pm

i'm interested in where have you found the peq4 schematics, because i'm not able to open the original device with the sdk software...

MCCY · Post by **MCCY** » Wed Oct 04, 2006 2:08 pm

I haven't seen it... or look below... But is PEQ4 other than what we have in SDK? It's so sad that we can't talk about that in detail because of the NDA.

You think we don't have PEQ 'on board'? If you say no I can't prove anything... Just with phasedeleting again... who wants that again raises his hands please

But... oh, I forgot. You remember Pulsar 1.0? I won't say more... Can you imagine how I build ZAM+? - Which has been official allowed by creamware to post b.t.w.

Martin

[ This Message was edited by: MCCYRANO on 2006-10-04 15:09 ]

djmicron · Post by **djmicron** » Wed Oct 04, 2006 2:26 pm

i'm sorry, i have had not time to test devices in the latest months, but i'll do.
It was my curiosity, because by reading your phase deletion threads it seemed that you've based it on the peq4.
I know that pulsar 1 devices are unprotected.
In my opinion, phase deletion with different eq's based on scope sdk, is very easy to obtain if the eq is used in flat mode, without saturation on its output and using absolute values.

MCCY · Post by **MCCY** » Wed Oct 04, 2006 2:35 pm

Yes, but imagine an EQ which has different knobvalues (achieving in an area of 0,000...) because of limited knobpositions in reality.

Imagine doubled EQs to achieve 24 or bit less db and resulting differing knob-positions.

Imagine there are other variations in SCOPE concerning DSP-distribution (referring to astroman et alia) etc. (I can't be precise at that point with changing values, because I have to do some further testing, you can read about my experiences of changing values from opening a project two times with differing values in the project-thread).

Imagine slightly different settings on an EQ in every band. You will not achieve -150db, just because of the layout, not because of sound.

"Pulsderzeit 24db" is 16 parametric stereobands, each 12db. So one need multiple PARA4EQs. One needs absolutely precise positioning on DSPs, one need 1000% corresponding values... I say: It's not possible. Will try it later.

Martin

P.S. Pulsderzeit is not a good device b.t.w. I have done MUCH better ones. One mighty problem is with having distortions controlled only via 1 (!!!!) input gain. In MOVEQ+ this problem (and many more) will be eliminated.

[ This Message was edited by: MCCYRANO on 2006-10-04 15:41 ]

MCCY · Post by **MCCY** » Wed Oct 04, 2006 3:09 pm

Ahhh.... now I finally have an exact example of what I'm talking about:

Setting exactly -10 db in PEQ4 is fine by the values.

Setting exactly -10db with keyboard (arrows) or knobmovement (mouse) is not possible on Pulsderzeit 24!!!! I can only get -10.02 or -09.98. This difference alone leads to impossibility to delete to -150db. I don't know if this alone... I am from my first short try not able to phasedelete it and these differences in values lead for sure to the impossibility of getting to -150 db.

So I have to also state that these are values which should lead to audible sounddifferences sometime when adding them and still the question is: When will they be hearable?
is someone able to hear a difference of 0,02db in loudness in one EQ-band? Could be, because it leads to a deletion to ONLY -56db!!!!!

I see: We will never get to a point of clearness in this case. So I see with these differences it is a wonder if possible to go down to -90db and below, but maybe it could be sounddifferences which lead to audible differences sometimes.

Yes, I see this problematic and I am curious what interesting other things we will find & maybe prove. I have still hope, to have some blind testing up sometimes here from people who are interested in possibilities and limits of our ears.

More fast tests (with two PARA4EQs):
Gaindifference from 0,1db in one ParaEQ set to 2000HZ -10db (-10,1), Q0,7 leads to only - 42db cancellation.

0,001db difference to -83db cancellation, so we now know in which dimensions this game playes. And these are differences on ONE Band, which occure neccessarily on devices with different knob-value-access-possibilities.

0,00001db (you can set these values with para EQ: just ignore that they won't display after typing - as one can see: they work) difference in this specific band leads to -119db canellation.

0,000001db => -124db

Isn't this interesting? Just these facts, aren't they intersting for themselves? These are some basics for listening and engineering and we can talk about such things. Not about any specific company. Let's just talk about such interesting things!

I could post examples for blind-testing, if anybody's interested! It would be fun to test our beloved ears!

P.S. These results were achieved with a song. I think white noise won't be too different. Will try that later.

[ This Message was edited by: MCCYRANO on 2006-10-05 01:29 ]

Post by **hubird** » Wed Oct 04, 2006 4:04 pm

impressive results

katano · Post by **katano** » Wed Oct 04, 2006 4:05 pm

it would be more fun to see your move-EQ finished

kensuguro · Post by **kensuguro** » Wed Oct 04, 2006 7:03 pm

btw, I'll also add that after I finished my round of tests, I did a very, VERY detailed listening comparison, at a cancellation of -60db... and could hear an EXTREMELY minute difference between the 2. I wouldn't say it's a significant difference, surely nothing that is sound character defining in anyway, but "audible", in a clinical sense.

But yeah, I realized from my tests, that there is a very high possibility of certain values just not being able to be obtained through arrow key browsing.

MCCY · Post by **MCCY** » Thu Oct 05, 2006 12:26 am

Yes, kensurgo, now we are a lot closer to the questions that araised in my head.

What are we able to hear?

I can underline your results. I think in that region you mentioned we are at the border of abilities of our ears. But still we can hear differences!

Martin

P.S. MOVEQ+ being finished does not only depend on me. In fact there are involved several persons & circumstances (which is good, so it still grows)... From circuit I can say that I made some great progressions again (I'm at 327% of what I expected in the beginning now

)and would release it tomorrow if I could.

[ This Message was edited by: MCCYRANO on 2006-10-05 01:35 ]

Shroomz~> · Post by **Shroomz~>** » Fri Oct 06, 2006 2:22 pm

The Krypton valve technology is taking considerable time to realise visually

MCCY · Post by **MCCY** » Fri Oct 06, 2006 2:54 pm

Shroomz, the Krypton thing should be kept secret till final release!!!

This posting deletes itself within 20 seconds.

[ This Message was edited by: MCCYRANO on 2006-10-06 15:55 ]

Shroomz~> · Post by **Shroomz~>** » Sat Oct 07, 2006 4:02 am

Of course!! How stupid of me!

So should I not tell anyone that your plugins have special mind-altering powers??