Compare mixes with equivocate. an experiment

so im playing around with equivocate and having fun experimenting. who knows if ill learn something or its a waste of time.

In any case I laid down a ‘song’ with no vox yet. I just did a really rough mix to get the levels more or less balanced. No real eq going on except maybe for bass.

ok, then i used equivocate to individually match the bass, the drums, and the guitars to a song from STP. I didnt try to match the whole song yet…just the individual elements. I threw on some compression and limiting and printed that

So I take that printed wav file and go back into reaper. Now I use equivocate to match that wav file to a different STP song and then to a chevelle song and I render those

here are the 2 results. I have my opinions (but Ive only listened a few brief times) but I want to see what others think. Not really a bash per se, just a fun experiment

very short snippet

original song final referenced to STP

original song referenced to Chevelle

how do they sound?

Cool experiment! I’m at work at the moment, so I can’t listen on decent speakers - I’ll take a listen when I get home tonight… BTW, what are the two songs you used as references?

well dont expect much quality wise lol. its really just a quick thrown together thing

IIRC first i took the individual drums, bass, and guitar parts from Sex type thing and match up my individual parts

Then I sort of forget what I did but I think i took ‘where the river flows’ and matched the whole song to that and then also ran it thru compression and a waves l2 limiter and printed it to a wav

then i took that wav and matched it to chevelle ‘face to the floor’ and printed an mp3 going thru much reduced compression blah blah and also printed the ‘river’ wav to an mp3 so the 2 tracks would be consistent leveled etc

of course they are probably just over compressed white noise but it is what it is

to my ears the guitars were pretty unnatural sounding individually BUT who knows, they may work in a mix

in any case its unnerving how radically different my recorded stuff sounds compared to a matched ref. Its weird to see it add 12 db of highs while cutting 12 db somewhere else. makes me feel like my ear is 24 db off. That being said, i dont “trust” equivocate as far as i can throw it but its still a cool tool

these are only 15 second snippets btw lol

The Chevelle referenced mix sounds better… but then again the Chevelle mix sounds better (IMO) and much more “contemporary” than the STP mix. It’s understandable, because the STP mix was mastered in the era where mixes had to work on vinyl as well as CD, so the sub-low end is very subdued compared to the Chevelle mix, but I digress…

What is common to both of your mixes is that the kick drum appears to be missing a ton of low end compared to your references. The drums in general lack low end “weight” - the snare too, but to a much lesser degree. Your mixes also sound slightly strident in the high end, but I think this is down to an overall lack of lows and low mids, rather than an excess of highs compared to the references.

Interesting…I’ve personally never tried this “reverse engineering” approach to mixing, so I wouldn’t be quick to dismiss it out of hand… perhaps you should persist a little longer with it. I can hear that you are definitely getting a similar general “signature” of each reference track here, but there is something that isn’t working as it should…

Just thinking about it further, possibly the biggest factor here is that (as I understand it) you are dealing pretty much strictly with EQ as the difference-making tool…

EQ is basically a volume control for specific frequencies, but don’t forget there is a time-domain factor at work here. How long a sound extends in time has a big impact on how it “feels” in a mix. High end sounds are very immediate, but low end waveforms take more time to develop. This is a critical factor with drums.

For example, even if your kick drum has the same frequency response the solod kick in your reference, it’s low end is primarily defined by how long the kick drum waveform extends after it’s initial transient hit. If a kick drum is compressed in such a way as to not allow the tail of the waveform to fully develop, then theoretically, while a it might have appear to have the same frequency response as the reference drum, in the context of a mix, the quickly-disappearing tail of the waveform would be swamped by the other low end elements of the mix (like the bass), causing the kick to appear to be only just the high end “tick tick” (which is what I’m hearing in both these mixes).

The same principle applies to snares and toms. That might be the issue here - just a thought!

2 Likes

yeah, this is why i say i dont really trust equivocate. it seems to want to peg 3-4 freq bands in a row either +12 or -12. Im pretty sure you arent supposed to use an mp3 as a ref but of course I do it anyway

so it takes my guitars and adds +12db from like 8k to 12k etc. yeah, no wonder they sound strident. and my kick was way heavier before ref matching lol. Im wondering if maybe my mids are generally sort of wack and then it tries to compensate by making the highs and lows wack also

i also wonder how overall volume affects the matching process

this would be a 10 minute mix of the song WITHOUT any equivocate referencing and without any reference track referencing. (also without any real tinkering with drums or guitars). The crunch guitars are mid boosted more than I normally would as a result of me trying to affect the equivocate curve (trying to get equivocate to boost mids LESS lol)

no automation of course or even worrying about crossfading stuff

this would represent something like the natural state of my ears…which I admit aint too good

in any case im pretty sure my methodology in the usage of equivocate (and in general) is off

original song

rhythm section only…how does the bass playing etc sound?

like I said, these arent really mixed at all, this is just showing more or less what was there before equivocate did it massive eq moves