Sony Oxford at NAMM...going native!

[Kingston]

An eq doesn't all of a sudden 'improve' (once the sound breaks up) - my 'hearing' tests simply tell me if a sound is pleasant; that really is all I need to know.

hearing based A/B tests still have their place, no doubt. there simply isn't any other meaningful way to compare transient performance. filter graphs will tell very little, if at all on this.

and it's in my opinion the most important aspect of all filtering as well. although surprisingly phase shape walks hand in hand with it (and somewhat eases filter design because of it).

let's say 2 eqs use the same implementation. if you match the curves, it's highly logical that the phase response will be pretty similar. this should result in a similar sound.

sure you should use your ears as a final judgement...but if the curve and phase are the same (and both don't have any kind of distortion or noise) then i doubt that the ears won't confirm that they sound the same.

I was actually referring more to the EQ design. I've been practising some of that lore, and I simply can't get anything meaningful out of those graphs. I know what they mean but I have a different esthetic about it. It's always like "okay, this sounds a bit bland and sterile, and it's been done before, let's see what happens if I stick another all pass network in there, oh wait, maybe I should try it parallel with 1/4th of the 2rd order version included".. then it's listening test after listening test, and then some more because I had a typo in the code. ugh. This is a personal preference, but I couldn't care less of what the graphs show me, or what the feedback network and signal flow graph looks like.

[bmanic]

I think the next logical step for more "musical" EQs is to introduce non-linearities and slight dynamic variations (ala Voxengo Harmoni EQ).

Nebula2 also demonstrates this really nicely. The ear/brain clearly likes constant movement and not static things. Note: even a long semi-random pattern that is repeated in a cycle is after a short while interpreted by the brain as static..

however, this might go into audiophilia territory already and defeat the purpose of EQ as a tool.

Cheers!
bManic

[Kingston]

I think the next logical step for more "musical" EQs is to introduce non-linearities and slight dynamic variations (ala Voxengo Harmoni EQ).

Nebula2 also demonstrates this really nicely. The ear/brain clearly likes constant movement and not static things. Note: even a long semi-random pattern that is repeated in a cycle is after a short while interpreted by the brain as static..

I'm actually semi seriously studying some uncharted filtering network territories, and state space designs. It has become painfully obvious why too many EQs are clones of each other: they rarely dwell outside the biquad and simplistic less than 4th order designs. This has officially been done to death.

There is a whole world of EQ sound to be done with state spaces and creative filter networks, ie. done for the sake of the sound, not for purely technical reasons and relying on white papers, which seems to be the current fashion since the dawn of DAW age.

[bduffy]

I think the next logical step for more "musical" EQs is to introduce non-linearities and slight dynamic variations (ala Voxengo Harmoni EQ).

Nebula2 also demonstrates this really nicely. The ear/brain clearly likes constant movement and not static things. Note: even a long semi-random pattern that is repeated in a cycle is after a short while interpreted by the brain as static..

however, this might go into audiophilia territory already and defeat the purpose of EQ as a tool.

Cheers!
bManic

I thought that was something that a lot of these analog-modelling EQ's did already.

[Tronam]

Thanks for reminding me of Posihfopit. I'd somehow forgotten about it, but am now reminded of how excellent it truly is. I had been testing out the demo of the new Sonalksis MK2 and started comparing their LP filters in particular on percussion. What a glaring difference. No matter how I tried to match them, the posihfopit was always silky smooth and detailed, regardless of whether it was in analog or digital mode, whilst the SV-517 was surprisingly harsh and almost distorted sounding. I must be doing something wrong.

[Kingston]

What a glaring difference. No matter how I tried to match them, the posihfopit was always silky smooth and detailed, regardless of whether it was in analog or digital mode, whilst the SV-517 was surprisingly harsh and almost distorted sounding. I must be doing something wrong.

You aren't. That's very near exactly what I experienced with the two side by side. SV-517... indeed, surprisingly harsh describes it well. I'm always surprised to see the unanimous praise it gets around here.

Now that you've discovered Electri-q/posihfopit, definitely check out Nyquist EQ as well.

[Tronam]

I don't have the UAD-1 any longer, but I remember the Cambridge having excellent quality filters as well. I miss that EQ actually. It was a very usable general-purpose equalizer.

[original flipper]

HI

I agree about the Sonalksis eq - you go try adding some decent boost on any high frequencies and listen to the harshness, IMO simply a poor eq for being creative.

The Voxengo FREE Overtone GEQ is superior!

TBH the above mentioned Voxengo eq is a good example of what I am talking about - compare it with ANY eq at ANY price: add some 'welly' to the hi frequencies and listen ... it puts some $$$$ eq's to shame.

I dont care what is going on under the bonnet or on a graph - it simply pleases my ears!

Flipper.

[bduffy]

What a glaring difference. No matter how I tried to match them, the posihfopit was always silky smooth and detailed, regardless of whether it was in analog or digital mode, whilst the SV-517 was surprisingly harsh and almost distorted sounding. I must be doing something wrong.

You aren't. That's very near exactly what I experienced with the two side by side. SV-517... indeed, surprisingly harsh describes it well. I'm always surprised to see the unanimous praise it gets around here.

Now that you've discovered Electri-q/posihfopit, definitely check out Nyquist EQ as well.

You know, as much as I've enjoyed working with the mk2 Sonalksis, I've been noticing a kind of smearing across my mixes that's giving me concern; I wonder if it's that distortion building up..?

[Kingston]

The Voxengo FREE Overtone GEQ is superior!

TBH the above mentioned Voxengo eq is a good example of what I am talking about - compare it with ANY eq at ANY price: add some 'welly' to the hi frequencies and listen ... it puts some $$$$ eq's to shame.

I dont care what is going on under the bonnet or on a graph - it simply pleases my ears!

Flipper.

I'm with you on this one. Keep in mind Aleksey has been doing filter and non-linearity development for quite the many years now so it could be said GEQ is the top of the heap right now.

I don't know exactly how it does it either, but I can't get it to sound bad! I can't say this about any other EQ, as they all seem to have their weak points. Logic would have me assume GEQ's weak point is that it's not tweakable enough, but the fact it's non-parametric never bothered me at all!

And hey, no digi transients or discernible aliasing even at 44.1khz. Here's where I tip my hat off to Aleksey's direction.

[leonard]

well, since you're talking about Voxengo, let me praise CurveeQ. Both the frequency matching and the presets. I had a fun mastering experiment, one click mastering. It was just a fun track, but it turned out about seven time nicer than it ever would if I was to tweak my way to aural joy. I think I want HarmonieQ, so I can say I have all their eq's.

[bmanic]

I think the next logical step for more "musical" EQs is to introduce non-linearities and slight dynamic variations (ala Voxengo Harmoni EQ).

Nebula2 also demonstrates this really nicely. The ear/brain clearly likes constant movement and not static things. Note: even a long semi-random pattern that is repeated in a cycle is after a short while interpreted by the brain as static..

however, this might go into audiophilia territory already and defeat the purpose of EQ as a tool.

Cheers!
bManic

I thought that was something that a lot of these analog-modelling EQ's did already.

Nope. Like URS, they just slap on a fancy GUI and make your placebo part of the brain think that they emulate proper distortion. They don't.

AFAIK only Voxengo OvertoneEQ, HarmoniEQ, Electri-Q and UAD-1 EQ's, except the basic one and the cambridge, (and Sonalksis EQ in a weird way.. ) model some non-linearities. Probably also Waves SSL thing as it has that "analogue" button but who knows.. oh and tritonedigital Hydratone has a waveshaper or something in there too.

Also the excellent free EQ22 from duncan parson has some distortion stage.

All the others.. URS, Waves RenEQ, Nyquist eq, etc etc. are simple equalizers with nothing fancy going on..

Cheers!
bManic

[bmanic]

What a glaring difference. No matter how I tried to match them, the posihfopit was always silky smooth and detailed, regardless of whether it was in analog or digital mode, whilst the SV-517 was surprisingly harsh and almost distorted sounding. I must be doing something wrong.

You aren't. That's very near exactly what I experienced with the two side by side. SV-517... indeed, surprisingly harsh describes it well. I'm always surprised to see the unanimous praise it gets around here.

Now that you've discovered Electri-q/posihfopit, definitely check out Nyquist EQ as well.

Like Kingston and myself (and numerous others) have said over and over again: Electri-q (posihfopit) + Nyquist + Voxengo OvertoneGEQ = all the friggin EQ you'll probably ever need and it absolutely KILLS a large chunk of the commercial competition.

Cheers!
bManic

[bugs]

What a glaring difference. No matter how I tried to match them, the posihfopit was always silky smooth and detailed, regardless of whether it was in analog or digital mode, whilst the SV-517 was surprisingly harsh and almost distorted sounding. I must be doing something wrong.

You aren't. That's very near exactly what I experienced with the two side by side. SV-517... indeed, surprisingly harsh describes it well. I'm always surprised to see the unanimous praise it gets around here.

Now that you've discovered Electri-q/posihfopit, definitely check out Nyquist EQ as well.

Like Kingston and myself (and numerous others) have said over and over again: Electri-q (posihfopit) + Nyquist + Voxengo OvertoneGEQ = all the friggin EQ you'll probably ever need and it absolutely KILLS a large chunk of the commercial competition.

Cheers!
bManic

Agreed, as far as EQs go, although I am also very fond of the Voxengo GlissEQ too. The Oxford Inflator, however, is in a class by itself with the possible exception of the Refined Audiometrics CLAS/NR.

[defjamm]

Nope. Like URS, they just slap on a fancy GUI and make your placebo part of the brain think that they emulate proper distortion. They don't.

you remind me of something. i had a discussion somewhere else and people didn't believe me that they aren't special. i'll do a blind matching of all of them with the free electri-q in low quality mode. i actually already did a lot of comparisons, and if you match the curves you'll get also the same phase response. they kind of use the same code on all of them, just the parameters are tuned to react differently.

All the others.. URS, Waves RenEQ, Nyquist eq, etc etc. are simple equalizers with nothing fancy going on..

which isn't bad, just people should know what they use and what they get (sometimes clean is what you want). for example reneq uses kind of the same/similar implementation as the build it samplitude eq, they null out to a pretty low volume if you match them.