Topic: EQ'ing a matrix: best approach?

[willyjbrown]

So I've got a great soundboard tape and a great tape from my mics for a friend's band's recent club show. I EQ'd my tape immediately. I just recently got my hands on the soundboard tape, and it sounds really good. I'm looking for the best approach to putting a good sounding matrix together. Should I EQ the soundboard tape first? In my opinion it needs a little more low end and a little less high end. Or should I just throw it together with my recording, and EQ the final mix after I get the right blend and balance of the two? All advice/suggestions welcome! If it helps, I'm working with Adobe Audition 2.0. Thanks!

[ShawnF]

Why not try putting the two together first before doing any additional EQ?  Your aud tape will likely supply the additional low end, and should soften up the highs (though since you already EQ'd that source, maybe not).  But if you don't like it, go back and EQ the sbd.  Maybe Copy to New a few minutes from each source (from the same spot in the music, of course), and then matrix those with and w/o the various EQ possibilities and see what works best for that, then go back and apply it to the entire file.

[BayTaynt3d]

I usually take this approach, but it depends from case to case. First, sync up the sources and fix drift, and watch out for phase problems and delay b/w the SBD and AUD sources. Then, get to know each source's strengths and weaknesses by soloing each individually. Next, get a rough mix between the two, doesn't have to be final, but close. Then, if either source has a know issue (like boomy overbearing bass or tiring highs), start fiddling with EQ on just that source while listening to it both soloed and mixed, but pay particular attention to how it is affecting the MIXed version. Same goes for the other source. I try only to fix the most offending problems of each source by EQ'ing that source directly, but while paying particular attention to how you are affecting the mixed-down version. Then, while doing as little as absolutely needed that way, I may or may not EQ the mix on the master bus itself. But be very careful, if you don't have good monitoring, you are likely to f it up more than help. But always, always, always keep an eye on how the mix-down sounds. For example, if the SBD is lacking in bass and the AUD has too much bass, the mixdown might be perfect without any EQ at all. If you didn't pay attention to the mix as you EQ'd each source directly, you might have ended up cutting bass on the AUD and boosting bass on the SBD for no reason at all. However, you could still try that to see what sounds better. There is no right or wrong here, but my main point is to always think about how you are affecting the final mix.

[wbrisette]

My approach may not work because you may not have all the individual tracks. However, here is how I do it.

First, I use a multi-channel DAW, so again, this may not work for you. But...

I mute all but two sets of mics, listen to them, eq them. Usually this is a very slight thing. Nothing too heavy handed. In fact, you usually want to remove something rather than add to it. For this I use Wave's Q10 or Renaissance 6 EQ plug-in. Then I do the same thing for the next set of mics, soundboard, etc. I have some sub-groups I create, one for mics, and one for the sound board. Those then get fed into the main output. Any limiting and eq I try to put on individual channels. That tends to work better than trying to change things later in the chain.

I then, unmute all channels and start brining the levels up. As I do this I listen for what makes the best mix. There is no magic bullet, I've seen people say 80% this, or 75% that... But in most cases we're really not working in percentages, we're working on levels, dB to be exact. And each recording is different, so it always varies.

I find that if I EQ the channels correctly, I never have to eq the mix, just the individual channels.

Wayne

[easy jim]

I usually take an opposite approach, at least when dealing with a post-matrix of 3 - 6 tracks (from a mono SBD + stereo pair to a 6 track stage pair + room pair + stereo SBD).  First off, I more often than not like and generally use some light compression (~ 2:1 ratio with the threshold set somewhere between the average peak levels and the highest peaks) in all of my mix-downs from sources with more than 2 tracks.  I always try and make sure, however, that I preserve the dynamic range of the show as much as possible and use compression conservatively.

After getting some mixing lessons from a few studio engineer/live sound engineer friends, I now use a shortcut for EQing the final mix.  By dialing in compression first, the effect on the final mix is that the frequencies which go over the threshold are most affected.  Thus, out of control low, mid or high end stuff of particluar frequencies tend to be balanced somewhat by applying the compression.  Then I listen to the resulting mix to see if it sounds properly balanced.  If it is not balanced and the adjustments I want to make are slight, I will EQ on the master fader/bus post-compression.  Only when things are totally out of whack will I start EQing the individual sources/tracks.

For any full mutitrack recordings (above 6 - 8 tracks), however, I will work on individual tracks separately as necessary.

A couple baseline things to keep in mind:
-always start mixing/EQing from the raw sources (if possible) vs. a pre-processed source
-make sure you have a good monitoring situation or it will be very hard to know if your mix will sound decent on other systems
-always listen to your draft mix on a couple different systems, particularly poor quality ones like a stock car stereo or a boombox, to make sure the mix translates well

[BayTaynt3d]

Just curious if you're talking multi-band compression or not? I too use a little bit of compression, often using Waves MultiMaximizer as my effective last step in "mastering" to go just a little bit beyond what normalizing would do and adding the 16-bit dither at the same time (b/c it is built-in to that plug). Curious what compression plug you might be using?

[easy jim]

I have been using the "Dynamics" plug-in in AudioDesk, MOTU's free version of Digital Performer that comes with their interfaces.  It does not offer multiband control, although I have always been curious to play with a multiband compression plug-in (I believe Digital Performer comes with one) or better yet an analog unit.

When I started mixing my matrix recordings in post, I was also generally using a little compression as my last step.  After following the suggestions from some more experienced folks (one of the Delta Nove guys who is a studio engineer/mixer and a close buddy who is a FOH engineer), however, I've found I am getting better results and having an easier time getting my mixes right doing my EQ (if necessary) post-compression.

My plug-in chain on the master fader usually looks like this:
-dynamics (light compression)
-para EQ (if necessary)
-trim (adj. gain to normalize)

sometimes I may add a soft knee limiter after EQ'ing if the SBD feed has those outrageous 10+ db spikes from the toms, or if there is a lot of proximate clapping on the AUD source and it is louder than the music.

[iriewsp]

I have found there is no one method.  I like a few methods of mixing two 2track sources together.    One method I like alot is to find out what the best features of each source are.  I find alot of the time with my mics that the mics have plenty of low end, more drums than the boards and less vocals.  I also find that the sbd recordings have less low end, more clean upper mid range around the vocals and guitars.  With these findings I decided to filter out everything below 800hz on the sbd source and everything above 1khz from the mic source..    I then went back and cut out some fequencies with a parametric eq around where both mixes crossed over...   Then I mixed the two sources... I then proceed to compress and limit the final mix.  Either via TCMasterX, SSL Quad compressor, or waves ultramaximizer L2.  Here is an example of something I did this way.   http://www.archive.org/details/jjj2006-07-29.flac16

Another method I like to do is similar to the above but not so harsh when it comes to EQing.  Instead of useing filters on the two different sources I use parametric eq's.  Same idea though, cutting out of one source, what you would keep in the other.  Then the same type of Comp and lilmiting.

I find when mixing anything be it multitracks or 2 sources or even a mid side recording before it is encoded you want to attack with eq on individual tracks before it hits the mix  thus allowing the tracks/sources to mix... 

[easy jim]

I have found there is no one method. 

Well said.  Especially when we are recording in all kinds of different venues, etc., there is no one tried-and-true method.

I've tried using the technique you describe of completely cutting out the low from one source and the high from the other.  I've only liked the results better that way, however, when the low end was completely out of sorts on the original captures (for instance room mics in REALLY boomy rooms).

For an example of something I recently mixed following the compression-first style I described, here is a recent Michael Franti & Spearhead show:
http://www.archive.org/details/mfs2007-02-21.matrix.flac16 or flac24
The mix position, which was just behind where my rig was set up, was ~ 50 - 60' back from the stage/PA, and the room was a bit (but not excessively) boomy.  The SBD feed had a nice balance, including decent levels for the bass.  I decoded the m/s pair ~ midway between mono and stereo.  The raw mix accentuated the boominess of the room and I believed I was going to need to use EQ in the final mix.  After applying the light compression, I decided to leave the EQ alone and think that the boominess was cut down enough post-compression for an adequately balanced mix.  Hey Now Now is a good track for checking with a lot of bass.

On the contrary, though, I may end up separately EQing the tracks and potentially cutting the low end out of at least one of the mic pairs I used for the same set-up style two nights later at 4th & B in San Diego.  4th & B is a cavernous, large, and very boomy room...especially back by the mix position which is off center kind of to the right and ~ 20' in front of a wall that reflects back a lot of the sound from the PA.  The sound is futher muddled because the stage and PA are set up on an angle that is not perpendicular to the orientation of the mix position or the reflective wall behind it.  The right side of the stage and right stack are ~ 60' from the mix position, while the left stack and left side of the stage are ~ 80 - 90' from the mix position.  The mic feeds, of course, give an excellent representation of the boominess and muddled room sound.  The SBD feed, on the other hand, has a ton of kick drum but very little bass.  I've gone through 2 draft mixes of Delta Nove's opening set from this night, and am still unsatisfied with the mix and prepared for more radical measures.