Linear Phase EQ - out of phase???

[Paps4]

Hi,

 

I'll get straight to the point... I was eq'ing a kick with the lin. phase eq from logic 8. I tried a fairly sharp q with a boost around 50-60 Hz....

 

I was surprised to hear that the kick's frequency boost came AFTER the sound... There was a delay. With the normal eq of logic, no such problem...

 

Anyone has experienced this problem in logic studio 8????

 

Very very unprofessional....

[mattrixx]

What you are probably hearing is the natural decay of the kick being enhanced, as the bulk of the 50-60Hz content of the sample is most likely going to be in its decay cycle. experiment with shortening the decay of the sample, or applying a fade to it.

[Paps4]

this definitely has nothing to do with the natural decay... there is an inherent delay in the boosted frequencies from this plugin... I use waves lin-eq as well as psp's lin eq, and no such artifacts exist.... Logic's linear eq seems to be having this response due to a bug....

 

If it was the inherent delay of the kick at such frequencies, it would be heard with logic's normal eq as well, so that is out of the question...

 

Try it out your self... give a more extreme amount of boost so you can hear it... it is WAY out of sync...

[DaniGordon]

Me too.

At first I thought what you were hearing was the natural sound of the kick drum enhanced by the eq boost.

But if you say it's not it...

 

So, can you please upload the audio here?

 

Can be an MP3 example with an EQ plugin that you say this does not happen and one with Logic's linear phase eq.

 

Zip the mp3 files and upload it here so we can listen ourselves.

[Paps4]

ok, will do so tomorrow...

[Paps4]

i actually did it now... here's the three files

 

Kick Nothing.mp3 is the clear kick hit

Kick Lin EQ.mp3 is the Linear EQ (problematic) file

Kick Chan EQ.mp3 is the Channel EQ (without) the problem

 

To make the problem more audible i boosted 24 db at 490 hz (or 540 don't remember) and with a q of 2.00

 

It is clear that the lin eq almost seems like a 'reverse' effect before the kick, while the chan eq actually simply boosts the frequency with no delay...

 

any thoughts???

 

http://www.nick-p.com/Kick%20Chan%20EQ.mp3

http://www.nick-p.com/Kick%20Lin%20EQ.mp3

http://www.nick-p.com/Kick%20Nothing.mp3

[fader8]

It is clear that the lin eq almost seems like a 'reverse' effect before the kick, while the chan eq actually simply boosts the frequency with no delay...

Paps,

This is not unusual for linear phase EQ's. They get their phase linearity by sending the signal through a normal phase filter twice. First in the forward direction, and then again in reversed direction and that essentailly "normalizes" the phase back to zero.

 

EQ's ring, but linear EQ's ring in both time directions for this reason. In both cases, linear phase or not, the ringing gets worse as you use high Q values, (steeper slopes) and larger boosts or cuts. The Waves Linear is a little better behaved, but not much. It's just something you need to be aware of with EQ.

 

For more info, click here

[DaniGordon]

It is clear that the lin eq almost seems like a 'reverse' effect before the kick, while the chan eq actually simply boosts the frequency with no delay...

Paps,

This is not unusual for linear phase EQ's. They get their phase linearity by sending the signal through a normal phase filter twice. First in the forward direction, and then again in reversed direction and that essentailly "normalizes" the phase back to zero.

 

EQ's ring, but linear EQ's ring in both time directions for this reason. In both cases, linear phase or not, the ringing gets worse as you use high Q values, (steeper slopes) and larger boosts or cuts. The Waves Linear is a little better behaved, but not much. It's just something you need to be aware of with EQ.

 

For more info, click here

 

Nothing like someone who knows this s****

 

[pantomimeHorse]

 

Nothing like someone who knows this s****

 

 

And fader is nothing like someone who .................

 

 

(Sorry, R. Couldn't resist that one !)

 

-

C

[Paps4]

hi again,

 

i'm fully aware of how and why eq's ring, and the way linear eq's work.

 

The problem is, the the time domain in which logic's linear equalizer is shifting the audio material, to avoid phase problems, is utterly unusable in any detail work.

 

working with bigger q's is the only solution with that equalizer. I have no such problems from waves, perhaps due to the limited values and range of values that exist with that specific waves eq. The waves is not only much better, as it is also usable, something which i cannot say for logic's lin eq.

 

My problem is, that i like to create "universal" sessions, since the studio i work with does not have waves, and would like to just open the session and "go". Yet i'm forced to render the audio now with waves lineq, before going there... very annoying.... and very badly designed eq...

[Jope]

I just listened to your examples and find it quite clear what happens. As mentioned above, the ringing of the filters takes place in two directions: After the punch... And before the punch. That's what causes that swish that sounds somewhat like the mallet speeding through the air before hitting the drum. The ringing has to be that symmetric to avoid phase shifting. I would recommend the use of a regular EQ for sounds that have to be punchy.

[mattrixx]

Upon further inspection, I agree with what has already been discussed... It aint what it should be, but as has also been mentioned, the use of wider Q's does seem to help, which I guess, goes against the grain of the whole idea of a potentially surgical EQ such as the LinEQ. Mmmm

[pantomimeHorse]

 

........ but linear EQ's ring in both time directions .....

 

 

Hey, man !

Does this mean we can use LP EQ plugins to get next week's

lottery draw BEFORE we buy the ticket ......... ?

-

Fuckin' hell ! Count me in ! ! ! !

 

 

-

C

[fader8]

working with bigger q's is the only solution with that equalizer.

You might try working with the same Q first, and level too.

 

To match the tightest Q the Waves Linear can achieve, (6.50) set Logic's Linear Phase Q to .93, and then you need to drop the level of the band by almost 3dB. The Waves Linear is pretty wak when it comes to level. It's output rarely (actually never) matches what you set in the plug-in.

 

I have no such problems from waves, perhaps due to the limited values and range of values that exist with that specific waves eq. The waves is not only much better, as it is also usable, something which i cannot say for logic's lin eq.

Apples and oranges. Unless you set them both up very carefully using an analyzer, it's impossible to compare them. If you do set them up to be equal, you'll find the Waves is only slightly better regarding artifacts.

 

My problem is, that i like to create "universal" sessions, since the studio i work with does not have waves, and would like to just open the session and "go". Yet i'm forced to render the audio now with waves lineq, before going there... very annoying.... and very badly designed eq...

Not bad, just different, so it requires a different approach. As it would if you chose something else, like the Neon or Firium, or even Algorithmix, etc.

 

As Jope mentioned, linear phase eq's are not a good choice if you want to keep things punchy. Also, in general, boosting any more than a couple dB with eq of any kind isn't good practice if you're trying to maintain a semblance of accurate reproduction. It's always better to cut other bands around it because these artifacts are mostly going to to be audible near the filter's cutoff frequencies.

 

Now, for synthesis sound design, Logic's linear eq is a lot more fun.

[Jope]

 

........ but linear EQ's ring in both time directions .....

 

 

Hey, man !

Does this mean we can use LP EQ plugins to get next week's

lottery draw BEFORE we buy the ticket ......... ?

 

The trick is possible only at the expense of latency - the whole audio stream is being delayed, which doesn't matter in playback, but is annoying when you play realtime.

Meaning you can only read in the paper that your numbers were drawn and you had bought the ticket before... If this was the case.

[fader8]

Does this mean we can use LP EQ plugins to get next week's lottery draw BEFORE we buy the ticket ......... ?

Just set the lookahead to +7 days.

[davidpye]

Does the OP DEFINITELY have Plugin Delay Compensation switched to All in Audio Preferences?

[pantomimeHorse]

...you can only read in the paper that your numbers were drawn and you had bought the ticket before... If this was the case.

 

Damn, Jope ! really thought we were on to a winner, there ...........

 

..... Just set the lookahead to +7 days.

 

I tried that. "Not enough memory."

 

 

-

C

[fader8]

Here's a comparison of Waves tightest Q at 516Hz, boosted 24dB like your sample mp3's. Logic's Lin Phase EQ is set up to match. Input was a perfect 700Hz, 2 cycle burst, so everything outside the shading is created by the EQ.

 

Is the Waves better? Yup, but not by much.

[Paps4]

When i want to give more click to a kick, and i do it with logic's lin eq, i hear a delay in the boosted frequencies...

 

Waves lin eq has never done that, no matter what the boost is, and what my ears hear is what counts. So maybe not that much difference in artifacts, but there is a big difference in the 'placing' of the phase aligned samples...

 

I'm sorry to say, but logic's lin-eq just won't cut it for me...

[Holger Lagerfeldt]

Waves LP sucks ass from a straw too.

 

I use Flux Epure and actually prefer not using a linear phase EQ.

[David Nahmani]

what my ears hear is what counts

 

Totally agree. Then why use the lin eq, when channel eq sounds much tighter for this purpose?

[fader8]

Waves LP sucks ass from a straw too.

 

I use Flux Epure and actually prefer not using a linear phase EQ.

LOL, I'll second that. That's what I was tactfully attempting to suggest in my last post. I'm glad we can rely on you to be blunt when it's needed, lager!

 

Then why use the lin eq, when channel eq sounds much tighter for this purpose?

I think a lot of people are under the misconception that if something is linear phase, it has to be better. While that may be true for amplifiers, it is most certainly not the case for equalizers, as you know.

 

I don't like to use lin phase EQ's either, except where they're really useful or necessary, eg parallel processing. But for straight track use, or even tweaking a whole mix, there's always a better sounding choice. Interestingly, some of the boutique hardware eq emulations out there are highly prized for their rather wild phase response. All EQ's have their limitations and not one of them is "perfect", but then, a big part of this craft is understanding the limitations of these tools and working within them, or sometimes actually using them to our advantage.

 

Sometimes the science part is important. But as I'm sure you'll agree, we shouldn't let it drag us around by the nose!

[bmdaugherty]

I really think of a lin phase eq for tone shaping. I only use it where necessary. I like them on vocals because I cant stand unnatural vocals. (vox with phase distortion are the worst) I also like to use them when mastering. In bothcases, minimalism is practiced at every point. Actually, I tend to use the scale all of the way zoomed in, so that you can only see +/-6 dB. I dont think I have ever used more than +6 or -6 in a mix, unless im going for an effect, in which case I dont use the lin phase.

 

If you want a perfect lin phase eq, get a Weiss. http://www.weiss.ch/eq1/eq1.html

I have NEVER heard such a great sounding, truely surgical eq. It's truely amazing. (I would be interested to see one of fader8's tests run on that for comparison.)