Spectrum Analyzer

[andrushkiwt]

First, I'm happy with the way my latest track came out... it'll be up in the clinic shortly, and i'll link to it here later once I can get on SC (during lunch).

However, I did try putting a spectrum analyzer on it and I was curious about what it was showing me. This is NOT an EQ, and my dynamic EQ shows a much more balanced frequency spectrum than this SA does. Am i missing the point with the SA? What is it telling differently than my dynamic EQ is? While my dynamic EQ graph is showing much, much more balance between the frequencies, this SA is showing a slope with the top end drifting lower. How should I interpret this as it relates to a SA? You can see (hear) for yourself in the track that the EQ balance is fairly distributed. I feel that I'm wanting to read this as an EQ, but its results are something else...

this is NOT a screenshot from MY project. It is a google search image, but it is exactly as mine shows on this track:

https://lh4.ggpht.com/NU5dWDzxSNFFldWp552Tal8b9AaM-Pa7LIcn7-0R7HPVXY6_VcGiRphuLI4KobGoZg=h900

actually, the thickness of the bars on the low end (left) increase on my SA. And the top end bars are very, very thin. mass-wise, if you will.

here's the track, though it's also in the clinic. let me know if i should remove this as i don't want to post it in two places. but this is the track in question that shows a SA like pictured above:

 

[Farview]

If one is showing you that the frequencies of your mix are balanced from low to high, and your mix isn't tearing your head off with high end, the scale must be weighed in some way.

It takes much more energy to get volume out of low frequencies, so it would stand to reason that a graph of something that sounds balanced would slope down from low to highs. There is a reason why a PA system would have 5000 watts going to the subs and 200 watts going to the tweeters.

Anyway, spectrum analysers were generally used to calibrate a system. You would pump pink noise, which is a sound with all frequencies at the same level, through the system and catch it with a measurement mic. Then you would adjust the eq until the SA read flat. (Actually you would have the lows pumped up by 6db to give the system some headroom)

However, this is an obsolete way of doing it, because the SA doesn't take the timing into account. Now, you would do the same thing with the pink noise and mic, but you would run it into a SMAART system to adjust the delays so the sound from all the cabinets is time aligned. Then you worry about eq.

All that is to say, looking at your mix through a spectrum analyzer doesn't show you anything useful, since it doesn't matter what sound looks like. They are pretty, though.

 

[andrushkiwt]

All that is to say, looking at your mix through a spectrum analyzer doesn't show you anything useful, since it doesn't matter what sound looks like. They are pretty, though.

ok good to hear. as well as everything else you wrote. thanks!

 

[ashcat_lt]

Yep. The EQs have some weighting slope applied to their graphs. I use Voxengo Span, and it has a parameter to adjust that. I generally set it at 3db/octave which is just about right for most mixes.

 

[andrushkiwt]

Yep. The EQs have some weighting slope applied to their graphs. I use Voxengo Span, and it has a parameter to adjust that. I generally set it at 3db/octave which is just about right for most mixes.

I just want to make sure i'm not missing something and that it doesn't mean to be read as some other real-time EQ. thanks! It looked like it was telling me i had way too much low end and nothing up top, though the mix certainly doesn't imply it has more than any other "rock tune" or whatever. just needed to figure out what i was looking at. thanks!

 

[Btyre2013]

that spectromiter is only useful when you can't find a certain problem frequency, that's how I use it mostly, don't forget your ears are the most important decider in the end.