Digital EQ Fact & Myth.

[Crash]

what is that one you've linked? Have you tried sound forge?

The one linked is the unbuild filter of "Acoustic Analyzing System", or better to say the Realtime Analyzer part that among other comes with a the best test-signal generator (+filter) that I found for PC. And the best thing is that the test-signal Generator part is free to use.

this is actually something I was going to say about your earlier images - remember EQ8 also provides a gain boost as the Q increases - like a resonant filter on a synth, so it's possible that is why you are seeing the boost in the noise floor on the Spectrum - you may need to pull the gain down more to make up for it or use a 2nd filter as well

The noise-floor also happens with the lowest possible Q of 0.10 and all 8 EQ point at low cut.

If the EQ8 had a way to disable this boost and make it so that Q just adjusted the steepness of the slope then I may not even bother loading RenEQ at all

It has such a way and I already showed you how, mister. You need to watch the animated graphic on top of page 2.

I had to wrap my head around that one for a few minutes when I first needed a steep filter in Live, but it's dead easy. Here comes the technical part:

You can use the following Qs without getting gain boost of more than 0.01 dB (may be higher than 0.00 dB though, depending on the source signal).

High Pass: 0.71 (default)
Low Shelf: 4.12
Bell: 18 (0.00 dB!)
Notch: 18 (0.00 dB!)
High Shelf: 4.12
Low Pass: 0.71 (default)

To make the filter steeper simply add additional EQ points with the very same setting and think of them like poles. Here's another animated picture to demonstrate how it works.



It's far from convinient! If you want to change frequency you need to click through all the single EQs (or map a midi knob to all EQs).

I noticed something odd with Hi-Quality mode. It kind of allows you to set Q higher, but every few seconds one of the two stereo channels may get a gain-boot while the other does not. The higher the Q the more often it happens.

[forge]

If the EQ8 had a way to disable this boost and make it so that Q just adjusted the steepness of the slope then I may not even bother loading RenEQ at all

It has such a way and I already showed you how, mister. You need to watch the animated graphic on top of page 2.

I had to wrap my head around that one for a few minutes when I first needed a steep filter in Live, but it's dead easy. Here comes the technical part:

You can use the following Qs without getting gain boost of more than 0.01 dB (may be higher than 0.00 dB though, depending on the source signal).

To make the filter steeper simply add additional EQ points with the very same setting and think of them like poles. Here's another animated picture to demonstrate how it works.

It's far from convinient! If you want to change frequency you need to click through all the single EQs (or map a midi knob to all EQs).

I noticed something odd with Hi-Quality mode. It kind of allows you to set Q higher, but every few seconds one of the two stereo channels may get a gain-boot while the other does not. The higher the Q the more often it happens.

no that is what I meant - a way to do it with one filter

[Pathos V2]

the point is not to match the curves. This is possible pretty close.

the point with different digital minimal phase EQs and the emulations of loved hardware is the INTERACTION of the human turning the knob and the resulting curve...

SO it´s far more than the GUI...

[ikeaboy]

So basically the price tag for a lot of these plugs is paying for the GUI????

no not at all. Lots of info that makes me want a Waves REQ
having said that, the vast portion of the time I'm prefectly happy with Live's EQ8, and I use REQ mainly for when I want to be really surgical with hi/low pass filters as the Q has no gain boost, but occasionally I do use it for it's character

Thanks for clearing that up for me. Good luck to any charitable soul who wants to model the Waves REQ in a rack, I'll use it for one!

Not sure if its what you needing for the 'non-boosting low frequency cut' but I believe rubber filter is interesting (does ridiculous 364db slopes)

Rubberfilter provides up to 64th order butterworth filters to ‘rubber’ out frequencies. This plugin was made to be a static tool. Automation of any parameter is on your own risk.

http://www.savioursofsoul.de/Christian/?page_id=8

[forge]

Not sure if its what you needing for the 'non-boosting low frequency cut' but I believe rubber filter is interesting (does ridiculous 364db slopes)

Rubberfilter provides up to 64th order butterworth filters to ‘rubber’ out frequencies. This plugin was made to be a static tool. Automation of any parameter is on your own risk.

http://www.savioursofsoul.de/Christian/?page_id=8

cool, thanks I'll check it out

although really my point was more that if Live did hi/lo-pass slopes like the RenEQ then I might not bother with it - mainly because I use the RenEQ more for those surgical cuts - and I'm pretty happy with that

this is the thing with anything like this - it's all really subjective - similar to why someone will pay 3 grand for a 70s Les Paul special or whatever - people like the particular character and personality

for me personally I've never found myself in a position where I got attached to any of these kinds of analogue EQs so I have no need to try and recreate it

for me EQ is a much more functional thing used for carving up the sound sources to bring out or attenuate elements and make them fit together, so 80% of the time EQ8 is prefectly fine, and the other 20% when I need a steeper slope REQ does the job

so in a sense I agree with the original post that digital EQs are essentially the same at their core, because from my point of view and what I need they are, but not if you are specifically looking for the character and curves of certain analogue EQs

this is where timur/crash (and rhythminmind to a degree) is missing the point a little - you can do as many of these tests as you like, but the best advice is just just make some f$$kin music and use the ones you like the sound of!!!!

[moreofmorris]

Interesting post. Thanks for that...

I've been drawing my own conclusions on this... I've tried many different EQs and I've always found it pretty hard to hear the difference between the two. I seem to use the EQ8 alot, and then, for more notches etc... I like to use the Sonalksis EQ.

With the EQ8 and the Sonalksis EQ, I seem to have everything covered it seems.

But I have always wondered about certain things with EQ, and this post has cleared it up nicely...

[Crash]

this is where timur/crash (and rhythminmind to a degree) is missing the point a little - you can do as many of these tests as you like, but the best advice is just just make some f$$kin music and use the ones you like the sound of!!!!

Absolutely right... if your budget allows to get other options. That's the point here, to make people who are on a tighter budget feel better with the EQs they've got at hand. Sure it would be nice to have more options for convinience.

"Don't feel too bad about having to use EQ8, you can get the same results. Better save your money on more fundamentally different sounding gear (like Compressors, Reverbs, Mics)."

[forge]

this is where timur/crash (and rhythminmind to a degree) is missing the point a little - you can do as many of these tests as you like, but the best advice is just just make some f$$kin music and use the ones you like the sound of!!!!

Absolutely right... if your budget allows to get other options. That's the point here, to make people who are on a tighter budget feel better with the EQs they've got at hand. Sure it would be nice to have more options for convinience.

"Don't feel too bad about having to use EQ8, you can get the same results. Better save your money on more fundamentally different sounding gear (like Compressors, Reverbs, Mics)."

actually I am very happy with the Ableton EQ and especially the compressor since Live 7 - Claes has done a superb job with the compressor and in many cases I even prefer it to other high end plug-ins

and as I said before, aside from the steepness of slopes on hi/lo-pass filters I'm mostly fine with EQ8 as well

[Crash]

I didn't mean to say the Compressor or EQ is bad.

I meant to say that using another Compressor, Reverb or Mic will likely have more impact on your sound than using another digital EQ. At least for me, if I need to shelve out money on gear then a digital EQ is not among the first choices. (I think a better Reverb for vocals would be a good investment for me or I just need to learn better control of Live's Reverb. )

[forge]

I wasn't suggesting that, just pointing out that I have never thought there was any reason to doubt the Ableton EQ8 or Compressor

but I agree Reverb would probably be top of the list for external

[Crash]

I'm still pulled back and forth between hardware and software for my live vocal effects. Right now I'm still using hardware EQ and FX (especially for reverb), but going all software is more and more tempting. Once I decide on that step it's gonna be hard for me to find a fitting software reverb that doesn't cost a grand. But EQ wise I pretty don't worry.

[fader8]

Well, an interesting thread.

I've been keeping a blog for a while now that focuses on this issue. Since I just added the artifact plot for Live 7's EQ8 plug, I thought I'd chime in on this here. Click here for the blog. I used Spectrafoo to calibrate the transfer functions shown on the phase page, and also to calibrate every EQ before running the artifact tests.

All properly designed Digital plug-in Parametric EQ's are the same.[/b]

I know your sitting there saying but hey my Waves SSL, URS, Vox, blah blah sounds completely different. This is why.

OK, I'll agree, to a point, but there's a bit more to it than that. I did all my tests using a -6dB bell filter so that I wasn't looking at any artificial saturation that might be added during boost. Also, I rarely ever use EQ boost so it's more relevant to the kind of mixing and mastering that I do.

With emulation plugins you have EQ+X. X = Saturation, bandwidth limits, softclips, & so on. Now take away X what do you have? Just the EQ.

What I'm looking at is the difference in many EQ's phase response and ringing, (or pre-ringing in some cases). From SSL: Though it is not possible for us to detect phase in isolation, the phase shift of an EQ can often be the cause of the ‘magical’
sound of that device once the EQ’d signal is added back into the mix.

I've always agreed with this and decided to plot the phase characteristics of several "emulations", notably the "colourful" Neve, Helios and Pultecs of the UAD suite. Indeed these have very different phase characteristics, as well as some overall bandwidth characteristics that resemble their analog counterparts. Check out the plots I ran on them on my Phase page. I started with showing a few high quality minimal phase EQs to show that indeed they can be made (much to your point) very, very similar.

So spend your $ on the tools that actually make a difference.

Thank you, sir. My point as well here, and one that many should share.

everyone likes null tests right =)

For EQ's you need to qualify that a bit further. Because of the varying degrees of ring and pre-ring with various EQ's, if there isn't enough transient and dynamic content to make this phenomenon audible, it will likely not show up in a null test either. But the way to find out is by using some extreme settings on very dynamic material.

Emulation plugins for the most part are Hype.

It really depends on how far the developers go with it. If they only emulate the Q control and gain tapers, then that's a bit lame. Some go a step further to be faithful to the gain-Q relationship and filter symmetry. But what impresses me about the UAD stuff is that they go the extra and include the original circuits phase and bandwidth attributes. And it shows!

[Grappadura]

Interesting points, even though I didn´t understand the ringing and the phasing part going back into the mixer.

Hmm, I want a UAD...

[fader8]

Interesting points, even though I didn´t understand the ringing and the phasing part going back into the mixer.

It simply reinforces that you really shouldn't EQ in isolation, ie with your track solo'd. You're doing two things when you EQ, changing the magnitude distribution of the signal and altering the phase linearity. Listening to the EQ'd track by itself, you won't notice this phase difference, but turn the rest of the mix on and now that track has a different phase relationship to the other instruments.

For example, if you swap polarity on a drumkit mic to clean up comb filtering from leakage, and then go through and EQ those channels, you may find that you need to swap the polarity back again on that mic.

Hard slope EQ with big boosts or cuts can easily shift the phase over 100 degrees at the cutoff freq. It can also exacerbate the filter ringing, turning a nice snappy tom or snare into mush. So, it's good to be somewhat informed about this stuff.