spectrum analysis

[Qnnilson]

hello

how can i use an spectrum analyser (like blue cat's free freqanalyst) to eq my tracks? i mean... what should i have in mind when eq'ing while looking at the spectrum... more bass/less mid/less treble?

thanks in advance

[halfadder]

hello
how can i use an spectrum analyser (like blue cat's free freqanalyst) to eq my tracks? i mean... what should i have in mind when eq'ing while looking at the spectrum... more bass/less mid/less treble?

the spectrum analyzer is there to answer just these questions for you. if the low end is to high then cut the low end. mids too low then boost them, if that is the sound you want.

[Tarekith]

There's no right way to EQ, that's like asking what colors you should use in your next painting. Depends on you, and what you're trying to achieve. If you don't know WHY you're EQing, then don't do it.

[hambone1]

I EQ with my ears, not my eyes. It's audio.

Now when I'm doing video, I DO tend to use my eyes a lot...

[Tarekith]

Spectrum analysis can be a good tool if you're trying to narrow in on a problematic frequency though, so it's worth knowing about. But yeah, EQ is a tool you use when you have a specific purpose for it in mind, not something you just arbitrarily do.

[halfadder]

yeah that is fine and all but there are certain things you cannot hear like sub 60hz. when i am making a bassline i will notice that it sounds fine and deep but it looks empty sub 60, so i bring in another osc/synth/something pitched down -12 -24 -36 or something to get the sub bass... doesn't make too much difference on my studio monitors but it is definitely noticeable on a pa.

[halfadder]

also spectrum analyzers can help with a mix, for me atleast. if a certain area of freqs appears to be less prominent than others then it gives me a clue as to what i might need to bring up, i bring up the track and it sounds 100x better. so it can give you little clues like that. it is definitely not the end all be all and for a final mix i always avoid eye contact with the monitor or close my and just let my hands and ears do the work.

[Qnnilson]

thanks for your answers

is it a good idea to grab a track from a cd (commercial cd), run a spectrum analyzer, and try to imitate the frequency range? or at least take it as a reference for my own tracks?

[halfadder]

thanks for your answers

is it a good idea to grab a track from a cd (commercial cd), run a spectrum analyzer, and try to imitate the frequency range? or at least take it as a reference for my own tracks?

well no. imitation is not good in this situation because eqing should be specific to what your track needs.

however i have heard a track or two and wondered what it would look like under the scope...

[ChiDJ]

thanks for your answers

is it a good idea to grab a track from a cd (commercial cd), run a spectrum analyzer, and try to imitate the frequency range? or at least take it as a reference for my own tracks?

I would say YEs!

Imitation is a great tool.

1) In your genre, find a song you love and write a track with the exact same arrangement, (from mix in -> Breakdown -> mix out. You will learn a lot.

2) Spectral, (FFT) analysis is invaluable. And matching the EQ of your tracks to the finished tracks you want to sound like is an awesome tool. How else do you find the holes in your mix?

I always A<-->B, (visually and aurally) my tracks to finished similar tracks. Why wouldn't I? If I'm going to play my material out, it has to stand up to commercial releases. After I get my levels in the neighborhood, I'll burn a CD and mix it in and out of songs that I would play in my set. Then, I listen. Is my song as loud?, Is it bright enough? too bright? does the track breathe? Overly compressed?, Are the vocals clear?....

Eyes and ears....they both get you there.


Tod

[Tarekith]

thanks for your answers

is it a good idea to grab a track from a cd (commercial cd), run a spectrum analyzer, and try to imitate the frequency range? or at least take it as a reference for my own tracks?

I'd say no, what you see in the commercial track will likely look entirely different from your tune unless you used instruments in the same frequency range and at the same volume. If you EQ's your tune to match the one you like, it might look the same, but it could sound awful.

[sebovzeoueb]

I use spectrum analysis to find where the main frequencies of a sound are, in the aim of cutting these out to make it cleaner, especially for low and high pass filtering.

[Dibubba]

I'd say no, what you see in the commercial track will likely look entirely different from your tune unless you used instruments in the same frequency range and at the same volume. If you EQ's your tune to match the one you like, it might look the same, but it could sound awful.

Terekith is right, here. Keep in mind that if you do spectrum analysis on a commercial track, you're seeing the finished (mastered) results.

I'll typically drop my spectrum analyzer into the Master when I'm mixing, to see the overall "big picture", but I don't necessarily use it as a guide for EQ'ing or mix balances.

Trust your ears. Here some tips based on my own experiences:

1. If what you hear on a PA is radically-different than what you hear in your studio, that's an indicator that it might be time to spend some money on new monitors and/or a sub-woofer.

2. Buy a cheap MP3 player, and load your test mixes onto it. Then, plug that MP3 player into your car, your boom boxes, your friends' cars, their boomboxes, your girlfriend (naw, wait - that's a different plug)...

3. Get a really, really good set of headphones. I use AKG K-240's, closed-cup with dual drivers. They have good bass response, and I mixed on them for years before buying a couple of sets of Near Field Monitors.

4. If you're concerned about "muddiness" prior to or during mixing, route all of the tracks you suspect may be causing the mud into a Send, and drop the spectrum analyzer into the Send's effects rack. Then, selectively mute the source tracks until you see what's causing the mud. Ableton makes this really, really easy to do - easier than any other method I've ever used.

5. Don't ever mix a song on the same day you completed tracking it. Your ears are fatigued by then, and judgement can go awry. Make a "test mix" if you need to (so you can gloat the next day), but do something in the mix that will DEMAND that you come back later to re-do it. I usually put a "beep" in the middle... that way, I'm forced to come back with fresh ears and re-mix it.

6. Buy a copy of T-Racks. It's wonderful, fabulous, tremendous! It will save you DECADES of mixing/mastering angst!

Not sure if this will help... but use what works for you!

Peace.

[ChiDJ]

Tarekith,

I understand your reasoning, maybe our goals are different.

When I "pre-master" and mixdown. My goal is to get my mixes to sound like the songs I play. How can this be done without reference?

If the commercial release is brighter, then I must boost some mid / hi frequency in my mix to bring it closer. I hear it and, with the spectral analyzer, I can see it. Of course I'm not going to match the frequency response exactly, (as your point well illustrates). However, "seeing" the specific frequncies that I'm missing speeds up my mixdown and teaches my ears to become more freqeuncy specific. ie "that sounds like it needs to be boosted between 300 and 500 Hz"...Turn on the FFT, "yep! there's a dip at 425 Hz"....

I just don't know why you would discourage this practice. IMHO, it's an incredible tool.

Tod

[Tarekith]

Tarekith,

I understand your reasoning, maybe our goals are different.

When I "pre-master" and mixdown. My goal is to get my mixes to sound like the songs I play. How can this be done without reference?

If the commercial release is brighter, then I must boost some mid / hi frequency in my mix to bring it closer. I hear it and, with the spectral analyzer, I can see it. Of course I'm not going to match the frequency response exactly, (as your point well illustrates). However, "seeing" the specific frequncies that I'm missing speeds up my mixdown and teaches my ears to become more freqeuncy specific. ie "that sounds like it needs to be boosted between 300 and 500 Hz"...Turn on the FFT, "yep! there's a dip at 425 Hz"....

I just don't know why you would discourage this practice. IMHO, it's an incredible tool.

Tod

Well, first our goals are definitely different, as I could care less if my tune sounds like a commercial release or some other record in the same genre. I'm far more concerned with making it match the sound in my head, and defining my own "sound" in the mix and pre-master phase.

Second, you have mroe experience than the original poster, you work in a studio, you know how to correlate what you see, with what you're trying to achieve. No offense to the original poster, but I got the impression that he was not that experienced and was trying to use a spectrum analyzer to visually EQ his mixes.

It's certainly great tool to use during mixing and mastering, I use it in mastering myself. But I just wanted to make the point that EQ your mix so that it LOOKS like another song, will not tell you anything. You'd have to have identical mixdowns of the same instruments in order to even get close doing it that way.

For troubleshooting problematic frequencies as I mentioned (as did you), it's a great tool. But it's not intended to be used the way the OP was asking. There's even software plug ins now that do exactly this, copy the EQ curve from one song and apply it to another. But if you've ever used them, you know that it rarely every makes your song sound any better.

EQ is one of those things that is so vastly different for each song, and each use within a song, that I don't think you can just blindly look at song and know what to do with your own song.

Besides, too many people copy others, there's nothing wrong with mixing and mastering a song in a different way from everyone else.