sonnox eq vs PSP vs URS etc

[The Listener]

Originally Posted by Spartacus

Did u applied the same Q to all the plugs?

First yes, just to hear them vis a vis. Then I tried to match them to get more cancelling - with Nuendo4 this meant 0.1 less wide Q on all frequencies, Sonnox required even more bending - I don't recall which Q values - but I couldn't get closer, PlparEQ didn't want to cancel more dramatically at all - so I finally left the same Q as those original settings on Nuendo 2 EQ.

I provided the raw file, so you can try your own test on the same file - use the original settings for source EQ - let's say Cubase stock EQ and try to match Sonnox, PlparEQ or other beasts to it. I really wonder. We need some more audio!

[drpenguen]

Originally Posted by The Listener

-52db difference without a problem - this is quite some residue and it is obiously heard in the files.

-70db I don't know, but I couldn't achieve such similarity with my tests...

The point I might want to make is - there IS a difference between high quality and average digital EQs. It is most obvious in what you can achieve with some, that you can't with others. So I don't really see a revelation with those "nulling" tests. I still trust UAD emulation EQs, PlparEQ and similar to deliver sounds that I don't expect from Nuendo on-board EQ. (although I use it regularly for many puporses)

-52 no problem !!! WOW
i take your word for it, i am not here to argue with your personal abilities

EDIT : -52 on white noise is almost ( varies) 62 on music btw...

i am gonna make the hearing test harder for you

if i change coupe of EQ s on your mix by 0.01 would you be able to hear the difference ?
say Q 1.23 to Q 1.24 on any random channel...



sorry, back to subject

to ben honest, i have never used UAD plugs. i diont have any of them.

would you consider waves ren EQ, sonnox eq as high quality EQ
or are we talking about PLPAR EQ with quality set to 7
thats a different monster...
i once compared quality 1 through 7 on PLPAR EQ... and yes,
quality change worth the extra power. bottom end changes dramatically....
(bottom end is more important in my stuff)

but in the case of sonnox, waves cubase eq, there is nothing dramatic,
in-fact , they are almost identical, and even same at some settings,
well -70db and some perfect NULLs
gave me that idea...

[drpenguen]

Originally Posted by The Listener

PlparEQ didn't want to cancel more dramatically at all - so I finally left the same Q as those original settings on Nuendo 2 EQ.

my personal view :
PLPAR EQ 10 band is a king,
rest is whatever
i am not interested,


EDITc : sorry, i am repeating my self but linear phase EQ s are totally different.

[24-96 Mastering]

Originally Posted by drpenguen

remember,
his comment is for -52db null, and i think its a very honest opinion too.
do you think you can hear -70db null difference on two identical music?

1. Drpenguin, you're mixing numbers from white noise and music tests .

2. I feel my earlier post is used in a misleading way. I wrote that I can (repeatedly and blindly via ABX-testing) identify the difference in the test setup I described, but it's very, very hard. In your post, it sounds a bit like it's an honest confession of mine that proves a black/white point, but it's not. There's no need for generalisations; I believe the exact details make up the picture just fine.

3. The main point I was trying to get across is that a single (null difference level) number is meaningless, so please don't quote my experiences in a specific listening test in that simplified context.

[The Listener]

Originally Posted by drpenguen

my personal view :
PLPAR EQ 10 band is a king,
rest is whatever
i am not interested,


EDITc : sorry, i am repeating my self but linear phase EQ s are totally different.

OK, then we agree... PlparEQ is a special beast.

About teasing in the previous post, - -52db is quite audible and you don't have to be specially skilled.

About you changing my Qs - no, I would probably not notice 0.01 difference...

I also tested Waves Renaissance EQ and it is close to Sonnox, while Nuendo2 is not so good in the highs. (it didn't cancel well enough with either of them).

I am through with those tests. If you will post files with low, mid and high boost on Sonnox Oxford, Waves REQ and Cubase stock EQ that can cancel each other out by -70db I will check it, though...

[drpenguen]

Originally Posted by 24-96 Mastering

1. Drpenguin, you're mixing numbers from white noise and music tests .

2. I feel my comment is somewhat misquoted. I wrote can (repeatedly and blindly) identify the difference (which has a null residue with my test music of -58dB, white noise null of -52), but it's very hard.

3. The main point I was trying to get across is that a single number is meaningless, so please don't quote my experiences in a specific listening test with single number.

did you read post 83....

this post was continuation of 83....

if you jump to last page, yes it looks odd... and i would like it either.
i didnt see any reason to re-post 83 and continue discussion.

but, my apologies

[24-96 Mastering]

Originally Posted by drpenguen

but, my apologies

Thanks and no worries, just don't want to be misunderstood.

[drpenguen]

Originally Posted by 24-96 Mastering

Thanks and no worries, just don't want to be misunderstood.

SORRY actually its post 88

its almost full post of yours,...

[LPK]

Originally Posted by The Listener

-52db difference without a problem - this is quite some residue and it is obiously heard in the files.

-70db I don't know, but I couldn't achieve such similarity with my tests...

The point I might want to make is - there IS a difference between high quality and average digital EQs. It is most obvious in what you can achieve with some, that you can't with others. So I don't really see a revelation with those "nulling" tests.

No, it's not about different quality .. it's about difference.
You might have a difference at -70db (which I doubt very highly anybody could hear), but this has nothing to do with quality.
Even if you take EQs with more different curves, which you can only "null" at -20db, it's still not a qualitative difference. They are just different and you may like one better for one situation and the other one for another situation. Or you always like one more than another, but other people might dissagree.



About linear phase EQs:
You can obviously not null them against minimum phase EQs because of the different phase-response. But still for most cases the difference is totally inaudible. Test yourself blind!
You can only hear a difference to minimum phase if there's audible ringing. Audible ringing accours with higher Q values and/or lower frequencies. So the often preached "better sound of LP EQs" with bright boosts in the highs is just bogus. Test yourself blind!
And LP not always sounds better. It depends on the task, if the ringing-behaviour of an LP EQ is suited better or worse.

Originally Posted by The Listener

-52db is quite audible and you don't have to be specially skilled.

The question is, if you can still hear it in a mix, which peaks at 0db.

[The Listener]

My point in all this story would just be that despite those nulling tests, that don't proove much, I know that I could not "master" with Cubase stock EQ or even something like Sonnox Oxford, which is audibly more "precise" than Cubase stock eq... What sounded best to me in the digital realm was PlparEQ. They DO sound different and it IS as question of "quality". It is just better than let's say Cubase stock EQ in terms what you can do with it and how do the results sound... It is not a matter of preference. You can say that choosing between the sound of some Alesis reverb or Lexicon480 or Bricasti is a matter of preference, but it is also a matter of quality (meaning - how pleasant it sounds and how it acts in the mix).

But hey, if you are capable of mixing or mastering with Cubase stock eqs or similar - cool. I hear or "feel" (while working) the difference in comparison with the PlparEQ, UAD Neve, Pultec, Helios EQ, without which I could not mix well. I guess many people hear that difference, otherwise nobody would buy and use those plug-ins.

[winrar]

Originally Posted by The Listener

OK, then we agree... PlparEQ is a special beast.

About teasing in the previous post, - -52db is quite audible and you don't have to be specially skilled.

)

its not same thing with EQ comparison but,

we have placed -52db digital glitch, kick, short white noise clips and click sound randomly all over the one Jeff Buckley song.
( Andy Wallace is our hero btw )
and
nobody noticed anything
even with HD650 headphones.

[The Listener]

Originally Posted by winrar

its not same thing with EQ comparison but,

we have placed -52db digital glitch, kick, short white noise clips and click sound randomly all over the one Jeff Buckley song.
( Andy Wallace is our hero btw )
and
nobody noticed anything
even with HD650 headphones.

What is the case here is that the residue - the difference between two EQs is let's say - -52db... on its own it is easily audible, and in the mix it is not some "sound" that is added to the mix, but it is the difference in its sound. Which is quite a different thing.

Anyway - to each its own, use what you think makes sense to use. But those "nulling" tests didn't proove what they were meant to proove (to me, at least).

[drpenguen]

Originally Posted by The Listener

those "nulling" tests didn't proove what they were meant to proove (to me, at least).

you are right in a way... i didnt get the information that i was looking for either you really need to set up test system for it and spend a lot of time... i was only messing about... and to be honest, dont bother use the ones you are happy with... anyway; i made more comparisons, and still doing, whenever i get some free time... all i have seen is things are getting more complex then ever and really without seeing their code ( i have a BA in computer information science), its hard to judge, which one is responding in which case, and how... over the whole spectrum, some frequencies are really not suitable for some EQs. as their Q curve is not responding the way you think they are... but on some frequencies, it really dont matter what eq you use, you get the same result. very strange coding they have... i think some companies, to cut the cost of programming, they choose wider frequency gaps, and some, spent more money on coding and they have narrower gaps... ( its not that simple,just to simplify things ) but when their frequency gaps overlaps, infinitive NULL, is not a problem between any min phaseEQ.... ( i dont have all the EQs, but algorithm is same) simulating EQs are.... mmmm.... compared vs real thing, they are way out of order... i have access to NEVE and SSL rooms, they sound close, if you pass the DA through the ssl or NEVE, but then its not a fair comparison anymore... as you are already adding a lot of character from the desks.. we tried as much as close as possible to NULL on desk with simulated EQ, then saved that setting and and played through same DA by hard by pass desk... A/B they are not even close...... sorry for the long and mixed post.... hope it makes sense... no objection to what you are saying tho.... happy EQings , .

[LPK]

Originally Posted by The Listener

They DO sound different and it IS as question of "quality". It is just better than let's say Cubase stock EQ in terms what you can do with it and how do the results sound...

Assuming the CubaseEQ has something against curve-warping near the nyquist-frequency, that's wrong. There is no qualitative difference in sound. Numerous people in forums and even some developers know that, since they actually scientifically tested it (well, I guess most developers will know that even without testing).
If you can reduce the differences of 2 different EQs to a degree, where it's not audible anymore, how can there be a qualitative difference in sound?

Originally Posted by The Listener

]I hear or "feel" (while working) the difference in comparison with the PlparEQ, UAD Neve, Pultec, Helios EQ, without which I could not mix well.

That feeling comes from the GUI and possibly the curves (a Pultec won't have standard-curves I believe). But this has still nothing to do with sound-quality.
Well, the PLPar is a linear phase EQ .. that might be better or worse depending on the situation. But it's not more clean or more HQ than minimum phase EQs.

Originally Posted by The Listener

]I guess many people hear that difference, otherwise nobody would buy and use those plug-ins.

Yes, they do "hear" that difference and I'll now explain you why:
1. They don't know anything about digital audio-processing or how an EQ works.
2. The devs bring out new plugins and always tell you some stories about why their plugin should be better than other ones.
[On one side that's meaningless stuff like "analogue sound", "super-clean 64bit internal processing", "component modeling" or "mastering precision". On the other side it's about stuff, which influences the way you work with it rather than the pure "sound-quality".]
3. Some charge like 500€ (or more) for their EQ .. so "it has to be better".

Originally Posted by The Listener

What is the case here is that the residue - the difference between two EQs is let's say - -52db... on its own it is easily audible, and in the mix it is not some "sound" that is added to the mix, but it is the difference in its sound. Which is quite a different thing.

Yes, because this difference in its sound is even harder to hear than "some sound".



Edit:
Ok well, to reformulate it (correcting it):
If you look very deep into it, there actually are sound-qualitative differences, BUT they are never audible. I tested LOADS of digital EQs .. not a single one had audibly bad quality .. it was always below the dynamic range of a 24bit audio-stream. Every EQ plugin I tested was transparent/clean!
There are a few ones, which can apply additional color in form of saturation (and/or noise), but that's it. It's something additional and not the standard and it does not mean "better quality" after all.

[drpenguen]

Originally Posted by LPK

Edit:
Ok well, to reformulate it (correcting it):
If you look very deep into it, there actually are sound-qualitative differences, BUT they are never audible. I tested LOADS of digital EQs .. not a single one had audibly bad quality .. it was always below the dynamic range of a 24bit audio-stream. Every EQ plugin I tested was transparent/clean!
There are a few ones, which can apply additional color in form of saturation (and/or noise), but that's it. It's something additional and not the standard and it does not mean "better quality" after all.

aha

i was guessing that involving of quality was well below dynamic range of a 24bit audio-stream too,
but never had any evidence of it... it was only a logical guess from my tests...

[The Listener]

Originally Posted by LPK

If you can reduce the differences of 2 different EQs to a degree, where it's not audible anymore, how can there be a qualitative difference in sound?
----------------

Yes, because this difference in its sound is even harder to hear than "some sound".

.

The first part of quote is the one that has not been sufficiently demonstrated - to which I objected. Maybe you can post some links to those forums or threads on GS that include sufficient proof and examples of those claims. I am more than willing to embrace "the truth", if it prooves to be like that.

About the difference in sound - the difference in sound is easier to hear than some additional "hidden" sound. If you have a more muddy midrange that would result in a -52db of residue it is clearly audible, or if the highs have completely different "sparkle" or "air" that would result in a -52db residue in the nulling test is again a very obvious difference in sound.

Again - I am interested in sound, I am a user, not a developer, I am not interested in code, but if there would be some large-scale scam of selling the same code with different GUIs for overrated price, I would be glad to avoid spending money on too expensive EQs, but I haven't seen/heard a clear demonstration/the proof, yet.

edit:

I realized that there was much comparing between Waves RenEQ and Sonnox Oxford in this thread - I also realised that they are very close and similar - and they can very well cancel on some frequencies, but I can't draw a generalisation out of this that all of the digital EQs are the same. I use niether of those two, only tested them, although they seem to be both quite usable. A couple of years ago I mixed one world music album with a collegues system with Waves RenEQ alone and it worked well...
PlparEQ couldn't be nulled with those EQs, and also not the emulation EQs of UAD which I use and have a distinct character. So, to me it is not un-important which plug-in EQ you choose as you seem to imply.

I am sure you can cancel out two high-end transparent analog EQs, too, with the right adjustment... but what does that help us? You have to pay similar amount of money for both and maybe you like the functionality or "curves" of one over the other, although you can achieve similar results with both.

[LPK]

Originally Posted by The Listener

The first part of quote is the one that has not been sufficiently demonstrated - to which I objected. Maybe you can post some links to those forums or threads on GS that include sufficient proof and examples of those claims. I am more than willing to embrace "the truth", if it prooves to be like that.

Dunno if there's a thread in this forum with proper examples .. I guess you'll find something at KVR.
But I can assure you, that I for example nulled the PSP Neon with the Fruity Parametric EQ2 down to -62dbFS (with a full-range signal peaking at 0dbFS).
I did an 18db bell boost on 4004kHz (while the PSP Neon is showing a Q value of 0.63) on both EQss for that comparison.

And the only reason, that it doesn't null completely is because both EQs have a different raster for their settings. So it's not possible to totally select the same Q values on both EQs.

Originally Posted by The Listener

About the difference in sound - the difference in sound is easier to hear than some additional "hidden" sound. If you have a more muddy midrange that would result in a -52db of residue it is clearly audible, or if the highs have completely different "sparkle" or "air" that would result in a -52db residue in the nulling test is again a very obvious difference in sound.

Test yourself blind. You'll fail.

And then those EQs don't add different "air or sparkle". They might have a different curve, which you might like better or not, but again:
This has NOTHING to do with sound-quality. The sound-quality of every digital EQ plugin I've ever tested is just absolutely clean and transparent. There are no artefacts, no "hard highs" and no "special sparkle/air".

Originally Posted by The Listener

I would be glad to avoid spending money on too expensive EQs, but I haven't seen/heard a clear demonstration/the proof, yet.

What kind of proof do you want to see/hear now?
Just say it.

Originally Posted by The Listener

PlparEQ couldn't be nulled with those EQs, and also not the emulation EQs of UAD which I use and have a distinct character. So, to me it is not un-important which plug-in EQ you choose as you seem to imply.

You can theoretically null them as well. For the PLPar you have to consider the linear phase thing.
You might need more bands of a standard EQ to match one band of a certain UAD EQ because the UAD EQs offer non-standard-curves I think.
For the PLPar I'm not sure ... I see it has different bell-types, but I guess those are probably standard-ones (maybe first, second, third order ... some parametric EQs offer semething like that).

Whatever ...
You don't even have to null them to see, if they're doing anything special or not. Just use the VSTPluginAnalyser from Christian Budde and look if those EQs do anything different than pure, clean EQing. As far as I know they don't, and so their actual sound-quality is not better or worse.

Maybe you should just talk with some actual plugin developers about that if you're interested. They might be able to explain it better than me. I assume most of them are honest enough to tell you the truth.

[drpenguen]

Originally Posted by LPK

Maybe you should just talk with some actual plugin developers about that if you're interested.

good idea,

i can now get big null values (more tan -70 db) on shareware compressors with tiny addition of Eq

[The Listener]

LPK:

But I can assure you, that I for example nulled the PSP Neon with the Fruity Parametric EQ2 down to -62dbFS (with a full-range signal peaking at 0dbFS).
I did an 18db bell boost on 4004kHz (while the PSP Neon is showing a Q value of 0.63) on both EQss for that comparison.
--------

That is what I say "not enough" - only one band, reasidue still -62dbFS "loud" and not even in the "critical" high and low shelving...

LPK:

And then those EQs don't add different "air or sparkle". They might have a different curve, which you might like better or not, but again:
This has NOTHING to do with sound-quality. The sound-quality of every digital EQ plugin I've ever tested is just absolutely clean and transparent. There are no artefacts, no "hard highs" and no "special sparkle/air".
-----------------

If you boost - they do ADD. Well, the sound IS the curve, the algorithm, the "behaviour" of the EQ at hand and it has everything to do with quality. (just not the excellence of electro-technical engineering, but the excellence of coding... - what's the difference?)


LPK:

What kind of proof do you want to see/hear now?
Just say it.
--------------
The same three or four band low, mid, high-mid, high treatment bewteen something like PlparEQ or even Sonnox, UAD Neve 1081, Algorithmix Red and stock Cubase EQ and similar to null completely or at least "sufficiently at -70dbFS. And some audio clips and written settings to back it up.

--------------
You can theoretically null them as well. For the PLPar you have to consider the linear phase thing.
You might need more bands of a standard EQ to match one band of a certain UAD EQ because the UAD EQs offer non-standard-curves I think.
For the PLPar I'm not sure ... I see it has different bell-types, but I guess those are probably standard-ones (maybe first, second, third order ... some parametric EQs offer semething like that).

Whatever ...
You don't even have to null them to see, if they're doing anything special or not. Just use the VSTPluginAnalyser from Christian Budde and look if those EQs do anything different than pure, clean EQing. As far as I know they don't, and so their actual sound-quality is not better or worse.
----------------------

Who cares about "theoretically" if the real life scenario doesn't work that way. I might be a theoretical millionaire, but I am not.

Well, I prefer to listen, not null or analyze and there is "something special" in those EQs I mentioned. But if the nulling and analyzing would show something I would appreciate it... As I objected - those nulling tests weren't actually nulling.

And btw - it is part of "quality" that you can just have a normal musical idea - like "more air" or "sparkle" and to choose a 12 or 15khz shelving filter on let's say(UAD) Neve and voila! what you imagined happened, without spending much time getting the same with experimenting and multiple curves on a less well made (coded) EQ (if it is possible at all).

LPK:

Maybe you should just talk with some actual plugin developers about that if you're interested.
----------------------
Well, they try to sell it, so.... I doubt they would say - hey, yeah, you can do the same basically with any digital EQ...

You would probably get such talk from the developers:

(from the website) "PLParEQ is a 10-band Phase-Linear Parametric Equalizer of the highest quality. Each filter may assume any of many different filter characteristics, and operate in either traditional phase-warping mode, or our phase-linear mode. It uses the same internal DSP core as all of our other high-end products. Audio streams are treated in either stereo or mono. Individual filters can be applied to either or both stereo channels, middle only, or side only.
Phase Linear Operation is achieved by processing your sound in both the forward-time and reverse-time directions through classic filters - all in realtime. This completely removes the phase warping caused by IIR filtering, and applies their roll-off twice. So each filter type becomes two: one for classic IIR filtering, and the other for Phase-Linear operation."


I doubt I will ask them if I can do the same with any generic digital EQ... And anyway - I experienced it in action. If you proove it otherwise, I might sell all that unnecessary crap I use.

[LPK]

Originally Posted by The Listener

If you boost - they do ADD. Well, the sound IS the curve, the algorithm, the "behaviour" of the EQ at hand and it has everything to do with quality. (just not the excellence of electro-technical engineering, but the excellence of coding... - what's the difference?)

...

And btw - it is part of "quality" that you can just have a normal musical idea - like "more air" or "sparkle" and to choose a 12 or 15khz shelving filter on let's say(UAD) Neve and voila! what you imagined happened, without spending much time getting the same with experimenting and multiple curves on a less well made (coded) EQ (if it is possible at all).

The curve is a matter of taste and has nothing to do with "the excellence of coding". It's a decision a developer makes .. standard-curves, curves of a Pulteq/API/SSL/whatever or own non-standard curves.
One is not more qualitative than another.

And this -again- has nothing to do with any kind of special sparkling. It's still absolutely purely, clean and non-magic digital signal processing without any non-linearities.
EQs are not less wellmade, because you might like another one better.
There are loads of people (me included), who still use a standard parametric EQ for almost everything, just because they always get immediatly exactly what they expected/want. Just because they know and like that tool like you might like your UAD-EQs.

Originally Posted by The Listener

Maybe you should just talk with some actual plugin developers about that if you're interested.
----------------------
Well, they try to sell it, so.... I doubt they would say - hey, yeah, you can do the same basically with any digital EQ...

You obviously haven't been much around at KVR-forums, where some developers will tell you exactly what I tried to tell you in this thread.

Originally Posted by The Listener

(from the website) "PLParEQ is a 10-band Phase-Linear Parametric Equalizer of the highest quality. Each filter may assume any of many different filter characteristics, and operate in either traditional phase-warping mode, or our phase-linear mode. It uses the same internal DSP core as all of our other high-end products. Audio streams are treated in either stereo or mono. Individual filters can be applied to either or both stereo channels, middle only, or side only.
Phase Linear Operation is achieved by processing your sound in both the forward-time and reverse-time directions through classic filters - all in realtime. This completely removes the phase warping caused by IIR filtering, and applies their roll-off twice. So each filter type becomes two: one for classic IIR filtering, and the other for Phase-Linear operation."

Well, what they write there is just a description of which features their EQ offers. You just don't really seem to understand it and so you draw wrong conclusions out of it.



If you want to believe, that that some digital EQs have an absolute better quality than most other ones, just do it. But I still wonder, why people are even thinking something like that without any prove for it on their side (if you know a bit about DSP, there's not even an indication for it).
Since this discussion seems to go in circles, I'm out of it now. If you really want to know more I recommend to join KVR forums and talk with some helpful an honest EQ-developers like Christian Budde.

[everyday]

Maybe I have to try harder but these eqs don't null...

MDW EQ
Digirack EQIII
EMI Mastering Pack EQ
Eiosis Air EQ

Testing only a few eqs doesn't mean every eq plugins in the world all null...

If someone can make MDW and DIGI rack eq null each other, that means George Massenberg is a liar!!!

Prove it!!!


T...

Originally Posted by drpenguen

so far, i have rested

sonnox EQ
waves ren EQ
URS Fulltec EQ
PSP neon (liner phase switched off)
SSL EQ ( -58 db on meters which is inaudible for me in mix or mastering)
Cubase channel EQ ( the stinky one )

they all NULL i am afraid.... only difference is their Q value.... nothing else...


i am more than happy to test your favorite mastering EQ plug,
against my stinky cubase EQ.
make sure they have demo available for download

[The Listener]

Yes, just proove it. LPK, you are talking about the technical theory, I am talking about "artistic" value. Quality, sparkle, "magic", etc. is in how they work in someone's hands.

I have not seen, heard the proof, sorry. They just don't null as you promote - or give us a link where they do it. Just you making statements is not a proof.

I'm out of the circle, too, until someone offers some real example.
----------

Why did I even bother... I checked KVR, I guess quite some don't agree with the simplistic judgement "all digital EQs are equal", including the esteemed Aleksey Vaneev of Voxengo, the producers of some of the finest plug-ins.

He said this among other things that you can read in the original thread at KVR:

"First of all, from your tests "null" applies to frequency and phase response, that are average measures. Secondly, making a comparison using a stationary signal will only reveal how EQs deal with such exact stationary signal, this comparison won't cover "second order" reactions that make a difference e.g. between American Deluxe and Mexican Standard Strat electric guitar (at least how I understand this). Both these guitars produce mostly identical stationary tone I believe."

There are other guys there, also some developers like Christian Budde that don't really agree with that proposition and offer some logical arguments and also some others who value UAD EQs much, like me. What did you want me to read at KVR, again?

There is also quite some logical fallacy or misguiding statements in the thread. The thread "Digital EQ fact & myth", starts with the statement that the three types of EQs are different:


(from here: KVR)

"1. Parametric/Series - "PEQ".
(These are the most common track EQ's, Most channelstrip EQ's, Waves stuff, Most Native DAW eq's, it's never ending. )

2. parallel - More uncommon in the digital world. Algorithmix has it.

3. Linear phase - Mastering EQ

These 3 types will be different.

But PEQ's are the most commonly used/emulated.

You can take any clean PEQ and create all the characteristics of any other EQ. It is the "X" that is variable. The Q setting is also a developer decision. You can't compare EQ's by matching dials & numbers. You have to use analyzers & match curves."


And then it goes on later in the thread and disregards this with a generalisation, that you can replicate anything with anything. Disregarding, that many think as "better" about exactly those EQs that simulate "+X" which cannot be separated from the EQ - it is part of it! And also regard as better the linear phase EQs, which are obviously a different beast.

So, where is the enlightenment in this? What a waste of time this was!


And some more of KVR for zi end:

"mustgroove wrote:

Don't get me wrong, the UAD plugins definitely have an x factor of some kind, they give an instant "wow" whenever you use them... all I've discovered is that it isn't distortion or harmonics or anything of that kind.

The whole distortion/harmonics and "warming" thing needs to be rethought IMHO. I think that whole theory arose at the dawn of the digital audio era, when some people were scratching their heads as to why all this digital gear didn't sound as "good" as the old analogue stuff... But to me it seems like just a theory - these UAD EQs kick total ass over everything else I've ever used... and none of them have any harmonic distortion to speak of. Clearly something else is going on.


When I was doing my assessments of various plugins, I found the UAD 1073, Pultec and 88RS the most difficult to null to any degree. (The Cambridge seems to be a standard BiQuad)
My suspicion is that there is something like Allpass filtering going on somewhere in these UAD plugs which produces phase shift without attenuation across the spectrum. I could never get the bass end and the treble end to both cancel to my satisfaction at the same time - sometimes I couldn't get the bass end to cancel at all...........

K"


And just for humor purposes: every person on Earth shares 99.99% of the same genetic code with all other people. And I am sure you would not say that all girls seem equal to you, except for the "+X" factor, but besides that - you can "null" them.

[drpenguen]

Originally Posted by everyday

Maybe I have to try harder but these eqs don't null...

MDW EQ
Digirack EQIII
EMI Mastering Pack EQ
Eiosis Air EQ

Testing only a few eqs doesn't mean every eq plugins in the world all null...

If someone can make MDW and DIGI rack eq null each other, that means George Massenberg is a liar!!!

Prove it!!!


T...

yes, you do have to try very hard.... hope none of them lin_phase EQ btw

i am not calling anybody lier.

there is something you are missing... some EQ s have very musical Q curves...
they dont need much tweaking to get good results. but some, you do need to tweak them to get same result.
and, it has nothing to the with the quality.
some suits one, other suits other task. becasue, we have so many EQ to choose from these days. people are not tweaking as in old days.

and now, people will jump,,, yea a lot of people we tweak....
and thats a total nonsense... as a result, you have zillion EQs to choose from...
and , they all are doing the same thing...
nothing so far matched OTB so far... if they were so great, so high quality, no difference vs the real thing..... how come no one managed the same dynamics, sweat tomes yet
show me one example ITB sounds better than OTB on any commercial CD...
not some premature GS demo stufff...
pros here, even comparing L1 to L2 , they think L2 sounds better than L1...
they are not even aware of that they both NULL 100%....
usually, people's opinion on plugs is ;
good presets = good plug
bad preset = bad plug
needs tweaking = bad
no tweaking = good

if you are happy with your plugs, just use them... but dont get in to the idea of
you have the best sounding highest quality EQs in this planet....

[Nordenstam]

Originally Posted by LPK

And then those EQs don't add different "air or sparkle". They might have a different curve, which you might like better or not, but again:
This has NOTHING to do with sound-quality. The sound-quality of every digital EQ plugin I've ever tested is just absolutely clean and transparent. There are no artefacts, no "hard highs" and no "special sparkle/air".

It's good to see that people draw different conclusions from using the same tools! Makes it all more interesting. Have also been poking around on plugs using vstanalyzer etc. It's been hard to find plugs that don't have relatively high levels of artifacts. Most shows stuff from about -120dB to -150dB on the FFT graphs. In comparison, the noise floor of flat 16 bit dither hovers around -130 on the FFT's.

This makes me assume that it's quite likely that people hear these things and do indeed make well informed digital EQ decisions based on what their ears tell them.

IMHO; Paying attention to signal integrity is as at least as important in digital as in analogue.

[LPK]

Originally Posted by The Listener

"[I]First of all, from your tests "null" applies to frequency and phase response, that are average measures. Secondly, making a comparison using a stationary signal will only reveal how EQs deal with such exact stationary signal, this comparison won't cover "second order" reactions that make a difference ...

1. I haven't used a static signal but a full mix.
2. Harmonics and stuff like that are not modeled in 99,9% of all digital EQs (including the UAD ones [you were just quoting mustgroove saying it by the way]).
The only difference between most EQs (and also most expensive and/or popular ones) is the curve. And how much you like the curves of an EQ is a matter of taste.

Originally Posted by The Listener

There are other guys there, also some developers like Christian Budde that don't really agree with that proposition and offer some logical arguments and also some others who value UAD EQs much, like me.

Christian Budde as well as every other guy who knows a bit about EQ plugins agrees with what I said. You abviously just still don't understand what I/they said.
The problem is you only understand what you want to understand to justify the investments you made.

[The Listener]

Originally Posted by LPK

1. I haven't used a static signal but a full mix.
2. Harmonics and stuff like that are not modeled in 99,9% of all digital EQs (including the UAD ones [you were just quoting mustgroove saying it by the way]).
The only difference between most EQs (and also most expensive and/or popular ones) is the curve. And how much you like the curves of an EQ is a matter of taste.

Christian Budde as well as every other guy who knows a bit about EQ plugins agrees with what I said. You abviously just still don't understand what I/they said.
The problem is you only understand what you want to understand to justify the investments you made.

Hi,

let's not argue in vain. I know whom I quoted, and also UAD stuff is modelled with anomalies, well their Neve 1081/1073, Pultec, etc. UA claim it themselves and it is also obvious audibly - they do sound particular, very different from the channel EQ in Nuende4 (which is great, btw. - similar to Sonnox, RenEQ - you would say - the same). And you cannot null them with that. Try and you'll see/hear.

About those guys agreeing with the proposition - everyone can go to KVR and read. You might say that someone doesn't understand, but it is clearly written. They had their reservations and didn't agree simplistically. Not much room for misunderstanding. And the initial poster himself stated that linear phase EQs and the ones modelled (+x) are different, so what is all the fuss about? You cannot disregard that, just becuase maybe 80% of EQs might be very similar and can be matched, but their functionality (GUI, curves, number of bands, MS posibilities, etc.) differs, which is still a "qualitative" category.

Well, I will leave that debate to guys at KVR and any new debater. I am tired.