Equavescent equalizer release

VST, AU, AAX, CLAP, etc. Plugin Virtual Effects Discussion
Post Reply New Topic
RELATED
PRODUCTS

Post

Trying to download, not working, maybe my Internet sucks.
--After silence, that which comes nearest to expressing the inexpressible is music.

-Aldous Huxley

Post

bmanic wrote:Image

This is how weird the bell is at 16dB/octave setting (whatever that means when it comes to bells? How steep the slope is?).

At 8db/octave, 40dB and 100dB the bell is much more symmetrical which makes me believe there might be something buggy going on with the 16dB/oct setting.

Also, the shelves shape changes slightly depending on if you set it in linear phase or normal minimum phase mode.

Indeed, this is one strange EQ.
Yep, but we know that it works using precalculated FIRs (remember, they take 2 weeks to generate...), so maybe the curve above is what happens when the EQ adds up two FIRs together, like I did here with two bell curves: http://www.wolframalpha.com/input/?i=0. ... %3D-3+to+4
Image

Maybe if people in this thread think this EQ sounds so good it says something about weird asymmetrical shapes?
Developer of Photosounder (a spectral editor/synth), SplineEQ and Spiral

Post

A_SN wrote:
bmanic wrote:Image

This is how weird the bell is at 16dB/octave setting (whatever that means when it comes to bells? How steep the slope is?).

At 8db/octave, 40dB and 100dB the bell is much more symmetrical which makes me believe there might be something buggy going on with the 16dB/oct setting.

Also, the shelves shape changes slightly depending on if you set it in linear phase or normal minimum phase mode.

Indeed, this is one strange EQ.
Yep, but we know that it works using precalculated FIRs (remember, they take 2 weeks to generate...), so maybe the curve above is what happens when the EQ adds up two FIRs together, like I did here with two bell curves: http://www.wolframalpha.com/input/?i=0. ... %3D-3+to+4

Maybe if people in this thread think this EQ sounds so good it says something about weird asymmetrical shapes?
just out of curiousity:
why are most curves symetrical in the first place? i mean apart form nyquist warped IIRs and for historical/analog prototype reasons, is there any good reason why the symetrical curves where adopted and propagated? i mean, with spectral processing we can have alsmost any different shapes we want, no? i mean, is there any (psycho)accoustic/musical reason for the symmetry ?
i seem to prefer the classic lin-phase ones it seems , but that may be for reasons of familiarity (do they use classic analog style curves btw.?).

i'm still waiting for the EQ names btw. :P

cheers!

jm

Post

probably because they're way easier to create and understand.
I don't know what to write here that won't be censored, as I can only speak in profanity.

Post

A_SN wrote: Yep, but we know that it works using precalculated FIRs (remember, they take 2 weeks to generate...
stop it please...this is not funny...
because this EQ really IS revolution!...he has spent 2 weeks to generate IR...I have spent 2 days repeatedly downloading it and installing it exactly according to given instructions...and results : frozen cubase or crashed wavelab...or telling me that cannot find data blabla... :shrug:
and what is revolutionary about this eq? ...its easy - I dont know any other EQ plugin on the planet/be it with asymetrical or symetrical curves/ which takes 2 weeks to generate FIRs PLUS 2 days to download and install these FIRs PLUS 2 seconds to crash all my DAWs...

Post

kvaca, your troubles are familiar to me. Stability is an issue with this plugin, but it's still beta.

IMHO, three things need to happen for this to be a viable commercial product:

1. Stability must be greatly improved.
2. GUI should be revamped.
3. Load/save times should be greatly reduced (at least, this is a problem for me on Studio One / Win7 x64).

Until then, I consider this to be an unfinished product. But I do hope it becomes finished soon, because I really want to use it in place of most of my current EQ tools. Whether due to weird curves or whatever, it does things nothing else can do IME.

And really, if it's all just about the curves, I would think it should be a simple matter for some developer somewhere to replicate those in a much more efficient plugin, even if it's just in a stepped form a la API EQs. Have at it, devs!

Cheers,
Eddie
The future exists in all directions.

Post

Eddie TX wrote:And really, if it's all just about the curves, I would think it should be a simple matter for some developer somewhere to replicate those in a much more efficient plugin, even if it's just in a stepped form a la API EQs. Have at it, devs!

Cheers,
Eddie
Well, I think you can replicate those kind of curves if for instance in SplineEQ (can't think of any other EQ that has a slope at the nodes like that, that's kind of a specificity of cubic splines and their control points) by tweaking the slope at the node (for instance you make a bell curve by raising a node, then tilt it one way to make it asymmetrical). Not sure if that would do the trick sonically, also it's all linear phase, but that might be worth trying. I wouldn't go about implementing anything like that in a different way, so you can consider it my best shot at it.
Developer of Photosounder (a spectral editor/synth), SplineEQ and Spiral

Post

Personally I think the asymmetric bells sound very weird, especially at tight Q values. They are moderately usable when they are wide. The asymmetric and strange shelves on the other hand sound really good.

So no more guesses on which is which with my little A/B test? Shall I spill the beans? Here's the post once more just in case somebody missed it. Simple +10dB low shelf boost at around 65hz or something. One is Equavescent and the other is Equilibrium, matched by ear and phase reversal (rest of the null is around -30dBFS).

EQ _X_

EQ _Y_

original file
"Wisdom is wisdom, regardless of the idiot who said it." -an idiot

Post

As soon as possible our programmer will have an update available for Equavescent™.

I wanted to clarify that it is not possible to capture the Equavescent™ eq response with a match eq and this demonstrates that Equavescent™ does not break down well into traditional FFT components. I am developing and prototyping an alternative to FFT convolution to reduce the CPU load of Equavescent™.

Kind regards
lbarratt
www.barrattaudio.com

Post

I have read the whole thread now, and still don't know, what this EQ is actually about.

There is NO concrete information about the approach of this EQ.
What is meant by "absolute clarity"?
Every digital EQ offers "absolute clarity". So what are we talking about here? Less ringing? Some special kind of phase relations?
Or is it just about the curve-forms ... maybe some psycho-acoustic ideas in there?

Also I'd like to know to concept, which requieres to render IRs 2 weeks long.


I mean, it sounds interesting ... but until now we only got some very empty marketing-words and some very unconvincing subterfuge for high CPU load. :roll:

LBarratt wrote:I wanted to clarify that it is not possible to capture the Equavescent™ eq response with a match eq and this demonstrates that Equavescent™ does not break down well into traditional FFT components.
So it's not a linear process?

Could someone, who actually downloaded this plugin, please capture an IR of it, load it into an FFT-convolution-plugin and null it with the directly processed signal? :wink:

Post

Nokenoku wrote:
LBarratt wrote:I wanted to clarify that it is not possible to capture the Equavescent™ eq response with a match eq and this demonstrates that Equavescent™ does not break down well into traditional FFT components.
So it's not a linear process?
Precisely. This whole thing puts me somewhat in mind of the Bill Nye vs Ken Ham "debate". On one side, those who understand the sampling theorem and system theory (that talking about FFTs is entirely irrelevant; either it's a linear process that can be described entirely with an impulse response, or it's distorting), and those who don't (largely OP).

Now, I'll happily concede that those graphs look interesting, and the curves are probably really quite interesting.

I haven't downloaded it myself, and I don't intend to. I can state with certainty that either it's linear, and can be captured with an impulse response, or that it's causing distortion, because that's the DEFINITION of linearity. The universe would fall apart entirely if an exception could be made.

Still, I'm always interested in new curves. If anyone sends me screenshots of interestingly shaped curves, or interesting IRs, I'd be remiss not to take a look at them.

Dave.
[ DMGAudio ] | [ DMGAudio Blog ] | dave AT dmgaudio DOT com

Post

DaveGamble wrote:Precisely. This whole thing puts me somewhat in mind of the Bill Nye vs Ken Ham "debate". On one side, those who understand the sampling theorem and system theory (that talking about FFTs is entirely irrelevant; either it's a linear process that can be described entirely with an impulse response, or it's distorting), and those who don't (largely OP).

Now, I'll happily concede that those graphs look interesting, and the curves are probably really quite interesting.

I haven't downloaded it myself, and I don't intend to. I can state with certainty that either it's linear, and can be captured with an impulse response, or that it's causing distortion, because that's the DEFINITION of linearity. The universe would fall apart entirely if an exception could be made.

Still, I'm always interested in new curves. If anyone sends me screenshots of interestingly shaped curves, or interesting IRs, I'd be remiss not to take a look at them.

Dave.
Amen! I didn't want to pile on with all my previous criticism as I think I've done enough, but OP's latest message just adds to my deep annoyance with how he either has deep misconceptions about what he is doing or he's just throwing senseless jargon around to impress and confuse, probably both. Really though, "does not break down well into traditional FFT components" < this phrase is extremely annoying to read, causes nerd rage.
Developer of Photosounder (a spectral editor/synth), SplineEQ and Spiral

Post

A_SN wrote:Really though, "does not break down well into traditional FFT components" < this phrase is extremely annoying to read, causes nerd rage.
Indeed. That was the phrase that caused someone to call me to comment. It's self-sufficiently indicative of someone who doesn't understand the words they're using. I believe it was posted to both KVR and GS, no less.

Dave.
[ DMGAudio ] | [ DMGAudio Blog ] | dave AT dmgaudio DOT com

Post

Now we have a problem...snake oil that tastes good [sometimes]... :help:

Post

kylen wrote:Now we have a problem...snake oil that tastes good [sometimes]... :help:
No problem. The curves probably are quite interesting- the asymmetric peak is unusual to say the least. (Though it can be expressed as a trivial extension of the flat-top paper I wrote a while back where you play with the gains. In any case, it's not clear that that shape is what people are saying sounds good).

Also there are some suggestions that the precomputed IRs are truncated, and there appears to be a subset of people who very much like how that sounds. I added artificial truncation to EQuilibrium six months ago for this very reason.

Which shapes do sound really good, and what do they look like on a plugin analyser?

Dave.
[ DMGAudio ] | [ DMGAudio Blog ] | dave AT dmgaudio DOT com

Post Reply

Return to “Effects”