About Equalization.....

[betowelch]

Hi Charlie and All:

Well....this a complicated question but, after some emails changed with Charlie I want to open this important question.

I'm trying to re-recorded some of my Youtube videos just only in audio form and with a better sound quality. So I will need some mixing acknowledgment with use of EQ plugins and just for that I'm trying to learn some basics fundamentals of EQ

I understand some things and others absolutely not. I understand that, if you don't have experience with EQ corrections it will be more easy to you use only the Graphic Equalization. You will correcting some frequencies and in a basic "trial-error" process you can give a improvement in your sound.
But it seems that the Parametric EQ it's a better tool for that as with it you can improve not only in boost or cut some frequencies. You will alter too other variables as width... OK..then, the parametric EQ it will a better than graphic...but this one it's a lot more complicated (to someone without experience like me) to starting what will be the center frequencies to select and to correct, what the width, etc, etc.

There are some rules to do this?

In a practical example, in this weekend I recorded TFYL and I want to improved it with EQ plugin through the Sound Forge....With a Graphic EQ plugin I could done and see the graphic and did a lot of "trial and errors" tentatives...but wasn't possible to do the same just only with the Parametric EQ plugin....Really I don't know for where we can to start...what center frequency should be selected? What it could be the best width to be select?


PS: We're talking of lead guitar equalization as the BT it's all right with it. So I should to focalized the guitar range freqencies, from nearly 200 Hz untill 3 KHz.... But there is an starting point reccomended?

Cheers from Brazil!
Roberto

[Charlie Hall]

Hi Roberto,
If you get the free Voxengo Span VST plugin you will be able to see the frequencies of any track.
The basis of a good EQ for a whole track is to have the area from the lowest frequency ie. kick or bass drum, at around 30-40Hz up to the accepted limit of human hearing of 20KHz.
The ideal recording curve is flat up to 500Hz, then should fall off at approximately 6dB per octave, meaning that at 1000Hz (1KHz) the energy should be lower by 6dB, then at 2kHz, another 6dB lower, and so on. It is rare to see a recording that follows this exact curve, usually there are peaks and valleys in the ideal curve. But with practise you can use a parametric EQ to fine tune these peaks and dips to make the curve smoother. I have found that vocals respond well to the ideal curve, down to about 200Hz, below that they should roll off to reduce popping sounds. Other instruments do not necessarily follow this curve at all, but with practise you can study different recordings to see where the energy is in each instrument. An electric guitar curve will vary according to the type of tone but for a bright sound like instrumentals and other modern guitar sounds the curve should basically be fairly flat from 80 to 500 and be a few dBs higher than the ideal curve between 2-3KHz. It does not really matter if any deviation from the ideal curve exists, as long as the curve is fairly smooth. I have noticed on many Shadows tunes that the main energy of the guitar is around 1-2KHz and the energy at lower frequencies is rolling off at the lower end (this is like the setting 2 on the EF86 channel of your amp).
For example, if Voxengo Span shows you that there is a consistently narrow sharp peak at 630Hz, you can set the parametric to 630Hz, set the width according to how wide this undesirable peak appears, then reduce the level of that frequency by the number of dBs that there is too much. When you run the track again, you should see an improvement.
An electric guitar has a narrower range than the full human audio range, from the bass E string of about 80Hz up to the frequency range of the guitar speaker. Most guitar speakers can reproduce up to about 5-6KHz but even at that extent the power of those frequencies is already dropping.
EQ can compensate up to a point the sound of a guitar but it usually isn't possible to make one guitar sound like another. This needs to be done with the playing, the guitar and amp and the microphone position.
EQ means equalization, which really means to correct a sound so that it sounds good with other accompanying sounds.
The aim is to get a curve that is fairly close to the ideal curve for a whole mix, but this curve can be arrived at by combining several different curves, one for each instrument or vocal or other sound. Aiming for this will help to learn how different instruments can be made to have their own frequency range and not interfere too much with the ranges of other instruments. A cleaner sounding recording should be the result.
Regards,
Charlie

[betowelch]

Hi Charlie:

I want to back to this thread as now I already had installed the Voxengo Span plugin and I can study the spectrum of a sound and with it to decide where I need to boost or where to cut.

We need to give a smaller focus to this discussion as we are talking in just only EQ for a clean guitar sound. Nothing about others instruments.....

This is something complicated matter but as I'm starting to use this Voxengo plugin I'm not very sure if I'm well deciding where it will more recommended to alter the frequency level.

Let's take a recent example that you work on it and sent it to me (TFYL)

www.4shared.com/file/188888975/af95905/TFYL_spectrum_before.html

According your recommendation (assigned in white points) it will needed some cuts in a region between 400 - 600 Hz and others cuts in 850 Hz and also in 2.3 KHz. OK about that it clearly that these points are higher that the others.

But also we have some lower points (that I'm indicating in yellow points) in frequencies about 700 Hz, 1KHz , 1.5 KHz, 1.7KHz and also in 2.1 KHz and 2.8 KHz. These points don't will to need give it some boost on it?
The rule it's just only cut the higher frequencies and left the lowest points without any alteration?

I had been reading something about ....and I confess to you that my brain it's burning each time I try to learn something more about it............In one article, I could read that it's better just only cut than boost....In others I read that you must to alter the spectrum so left it more balanced to be more proximity to the ideal spectrum (flat), so you will need not only cut but also to boost some frequencies......

There are some practical rules?

I'm thanking in advance some recommendations about it.....

Cheers from Brazil!
Roberto

[Charlie Hall]

Hi Roberto,
In general, it is good to cut frequencies that are too much (deciding how much is too much is another matter).
For a single instrument, it is not such a good idea to boost the low points. It is OK to try to do this with a whole mix where there are several instruments at the same time.
Regards,
Charlie

[betowelch]

Hi Charlie:

Ok. Thanks for your reply. I will avoid to boost lower points. So I must concentracting in cutting the peaks points.....

Cheers from Brazil!
Roberto