spectrum analyzer for mixing / mastering

Pretty much completely useless for mixing, all you see is freq vs SPL and not what is contributing to it!

Yeah, I've got a tip - EARS, NOT EYES!

This comes from one of the foremost cowbell utilization experts on the planet

during mixing/tracking stage spectrum analyser could be useful but not essential but mastering without a good analyser is like trying to walk around the house while blindfolded.
Sure, you can find your way around if you get used to the blindfold but why not just remove it in the first place? Makes things easier for you.

Yesterday I gave a spectrum analyzer a try to help me Master a song . I am no Mastering engineer at all , its like black magic to me . Anyways I think it turned out pretty good . Maybe you guys with better systems and ears can give a listen ?

I copied the EQ curve of the Nirvana song In Bloom . Not exactly the same , but the overall frequency relationships . I looked at 20hz to around 200hz , then 200hz to 700 hz , then 700hz to 3K..........and looked at the rms difference to kinda figure out the curve .

So if you have time , compare my song (Ramones Cover) with In Bloom .

www.dropbox.com/s/fx66039y0ctm0d4/Beat%20on%20the%20smat%2016bit.wav?dl=0

Sibilance is also a problem for me , I'm always fighting it . Let me know if theres too much sibilance .


Theres one word in the first chorus "back" I'm gonna fix it sounds weird .

Well, I guess that's OK as long as it doesn't become a crutch, but really you should work more on being able to determine that stuff by ear.

What I generally do when looking for that sort of problem on something is identify the approximate range and set up a parametric eq on the channel set with a fairly narrow Q. They I put a hell of a boost on that frequency band and pan it up and down while listening to the source (track, whatever). I ignore the numbers. Numbers are a distraction (and they're seldom perfectly accurate unless you have really good gear, especially if you have pots, not switches.) You do that, you get to the point where the problem is, it'll jump right out and slap you in the face. At that point you stop panning your frequency because you've found it, dial back the gain on that band to flat, and start cutting and maybe adjusting the Q control if it's continuously variable until it sounds right. It's actually more accurate than a spectrum analyzer because analyzers are generally set up in 1/3 octave bands (which is what, about 4 notes? And most of those notes will register to some degree on two adjacent bands.) and this gives you pinpoint frequency accuracy, and it's acrtually faster once you get used to it. The spectrum anaylzer is useful to give the pahdsoosah's girlfriend something pretty to look at.

And make sure to listen to your results out of solo. A tone that sounds great soloed very often sounds like hell in context, for one reason or another. Might want to fine tune your cut settings (amount and sometimes Q) in context.

I sometimes use a similar technique to determine boost frequencies, too, but bossts usually don't seem to me to be so surgical in nature, for the most part. You'd probably find this useful on your cymbal boosts (at a more moderate Q), although I generally prefer to deal with that with mic placement, cymbal placement, or if possible, with choice of cymbals. Also a few judicious (and always diplomatic) words with the drummer can help a lot, especially if he's a pro.


Well, not really because you can't compensate for a room resonance with EQ for reasons that go beyond the scope of this post or thread. But I get your meaning.


<chuckle....> Once you figure out what to listen for you shouldn't need it!