Mixing flawed tracks

[austin]

What do you all do when you're faced with mixing tracks that are deeply flawed sonically (excessive noise, digital clipping, bad edits, generally poor sounds, etc) but the band is unable or unwilling to retrack anything?

Do you clinically approach all the fix-it work first before even worrying about putting on your creative mixer's hat? Or do you jump in and try to get it feeling musical first, and then start addressing technical problems once they're in better context?

And from a psychological standpoint ? How do you get excited about mixing something that feels more like a salvage operation?

Any sage advice? Or funny stories?

[A.David.MacKinnon]

Been there many times. Every situation is a little different but for me I usually spend a day (or afternoon) importing the tracks into Pro Tools, figuring out what's going on and how I'm going to route everything through the board. Lots of self produced records suffer from crazy high track counts so for me half of the job is just figuing out what stays, what gets sub-mixed and what isn't needed at all.
If there are multiple mics on the same part I'll make some general decisions about what I'm going to use and then make the unused tracks inactive (and hidden). I'll often sub-mix things like Bass DI and amp tracks to free up channels. If one part is spread across multiple tracks (like a vocal with the verses on one track and the chorus on another) I'll bounce them down to a single track. I'll also clean up anything I notice at this stage. That usually means fixing bad edits, adjusting weak or too hot signals, cut and pasting to fix any clips, etc, etc. I also apply high pass filter on anything and everything that needs it. I swear high passing is a must when you're mixing records that have less than ideal tracking.

Once that's stuff is all sorted I'll start working song by song. There are lots of fixes here too but once the mix is going it's easy to tell which problems add character and which are just problems.

In the end it's rarely as bad as it seems when you first get the tracks.

[dubold]

And from a psychological standpoint ? How do you get excited about mixing something that feels more like a salvage operation?

think of it this way: if all the tracks sounded great and were at the right levels, you'd just bring the faders up and it'd be done! Anyone can do that. Whereas a salvage operation takes skill, imagination, talent, creativity, and a pinch of audio voodoo. You really have to know your craft to pull that kind of stuff off.

What do you all do when you're faced with mixing tracks that are deeply flawed sonically (excessive noise, digital clipping, bad edits, generally poor sounds, etc) but the band is unable or unwilling to retrack anything?

Unless band members have died, "unable" really just means "unwilling". they might not be able to retrack the stuff exactly the same way, but it could be done.

Having struggled with cleanup operations in the past, I find it breaks down this way: re-recording means more work for the band, but makes it less difficult to fix the mix. If you're going to have to struggle to correct it, it's easier for the band but the end product might be below your sonic standards.

So: if it's going to make your life easier, offer the band a deal: it'll cost X dollars per hour for me to fix this in the mix, or Y dollars to re-record it. (where Y = X minus whatever you think is a reasonable discount for headache prevention)

A lot of this depends on your relationship with the band, though. Are these people you know already? do they trust you when you say "there are some technical issues with these tracks"? If you show them bits that you think are problematic, do they hear what you're talking about?

Ultimately, the band is your client. You're trying to give them what they want, first and foremost. If you're thinking "but what they want sounds terrible, and if other people associate my name with that work..." then maybe this project is not a good partnership.

If there are other compelling reasons for continuing to work with the band - they're good friends of yours; they're good friends with a lot of other bands; they're incredibly famous; they're holding your family hostage until the mixes are complete... those end goals become the motivation until you get the music fixed to the point where you can be excited about THAT.

[vvv]

No set answer, of course, but generally I give it a faders-up listen, then I try to clean up the the tracks, submix the obvious, fix punches, levels and other technical glitches, etc.

Then, if it's really daunting, I set up the lighting and the booze, mebbe a snack, and go to it with a general mix, and then break it down by parts, ex. the verses then the chorus. I try to leave the coolest stuff and ear-candy for last; if it's the most fun, then it's to be looked forward to.

If it's really bad stuff, I try to mebbe work on something else in between.

I'm not commercial, I do a lot for friends and a lot of collabs, etc., so I run into other amateur's stuff all the time to mix from my li'l bedio.

[farview]

First, I double my fee. That's one of the ways to get excited about a salvage operation. But even at that, by the time I'm done rescuing the tracks, I have a hard time switching gears and getting into creative mix mode. Especially when I have other projects scheduled and the salvage project takes twice as long as I've allotted for it.

Then I see if I can turn anything that is suffering from 'general poor sound' into a feature of the mix. Make it seem like an effect instead of a screw up.

Most bad edits can be fixed or possibly won't really be heard in the mix by anyone who doesn't know they are there.

Sometimes you can get rid of clipping with a high pass filter, since it makes a square wave. A square wave is DC at the top and bottom, so a high pass filter sometimes smooths it out enough to deal with.

[austin]

Thanks everyone -- some good insights here!

In my current situation, these guys are friends, so I'm happy to do whatever's needed to make it work. But coming in to my place to retrack isn't an option because we don't live in the same city (and the band members themselves are scattered among a few locales). They do understand the problems now that I've pointed them out, but have made clear that they want me to just "do my best" with what we've got.

So that's what I'm up to. First discovery: running the clipping guitar tracks through a speaker sim in Amplitube is smoothing them out pretty well!

[Marwood Williams]

Sounds like you have your work cut out for you. I always try to find something useable or salvageable that I build on.

From a "psychological" standpoint, one way to motivate yourself is the actual challenge of being able to make something decent out of a mess. It's quite an accomplishment if you can pull it off.

[Nick Sevilla]

I do the following:

I asses the raw tracks with whomever is "in charge" before doing any work.

This I call the"Estimate" time. typically within a couple of hours of listening to each track and its component parts, I can get a rough idea of the options.

Then, I give them two options :

BEFORE MIXING ANYTHING:

1.- Re-record anything that they deemed unacceptable. We make a to re-record list, and I can give them an estimate on that part.

2.- Edit what is there, without re-recording anything. I again give an estimate on this. Usually, this takes longer, and therefore costs more $$$, but you would be surprised at how many artists opt for this option. You'd think they're made of money.

Either of these carries the same hourly fee.

Once this is done, then mixing can begin.

MIXING STAGE:

My usual fee applies here, since I don't charge a mix by the hour.

[bronsonmestizo]

I swear high passing is a must when you're mixing records that have less than ideal tracking.

Square low end is almost impossible to balance, because there are NO dynamics. If you shave off the lows of individual tracks at different frequencies, it can help open things up.

And low passing... an extreme slope high cut can take out a lot of noise that's created due to digital clipping or bad audio file conversion. This can help you accentuate the fundamental tones.

Also, it may sound counterintuitive, but you can distort and reamp digitally distorted elements to modify the characteristics of the distortion. Almost like "Blurring" a pixel-y digital photo to make it look "Old." Distant mic the reampped signal, and the room will further smear the edgy "chchchchch." EQ the signals on their way to the amp to avoid exacerbating Hi/ Lo end problems.

[bronsonmestizo]

I swear high passing is a must when you're mixing records that have less than ideal tracking.

Square low end is almost impossible to balance, because there are NO dynamics. If you shave off the lows of individual tracks at different frequencies, it can help open things up.

And low passing... an extreme slope high cut can take out a lot of noise that's created due to digital clipping or bad audio file conversion. This can help you accentuate the fundamental tones.

Also, it may sound counterintuitive, but you can distort and reamp digitally distorted elements to modify the characteristics of the distortion. Almost like "Blurring" a pixel-y digital photo to make it look "Old." Distant mic the reampped signal, and the room will further smear the edgy "chchchchch." EQ the signals on their way to the amp to avoid exacerbating Hi/ Lo end problems.

[cgarges]

I'm dealing with this right now on a live mix I have to do next week. The band and I have already had the necessary conversations about it, so I feel okay talking about it here, even though it's not finished.

The first issue was that the recording was done with stage mics only. No audience/room/ambient mics. The second issue was that there was no drum overhead miking, so the cymbals are only represented by extremely minimal bleed in the other drums mics (impressive isolation there, I must say) and the extremely large amount of bleed in the vocal mics caused by the completely pummeling compression printed on those tracks (issue number three). When the band gets quiet (or loud), the vocal tracks remain at exactly the same volume, so this is going to be a fader ride-heavy mix.

Issue number four was that the guest guitarist on a couple of numbers was using a laptop amp simulator plug-in thing that the soundguy just took as a DI and right off the bat, the signal was printed right into digital clipping. Then, the guitar player spent the next few minutes of the tune adjusting his volume (at the guitar, I assume) so that I've got an entire song with wildly varying levels on this third guitar.

I had a similar issue with the bass player, who kept tweaking his volume the entire time and all I have for the bass recording is a miked amp signal.

The biggest issue is that one of the two primary guitar players (of course, the guy taking many of the lead lines and solos) kept having amp problems that involved a completely obnoxious and unavoidable cracking static that cut out entirely much of the time.

So here's how I dealt with each of these:

The drums sounded pretty good with the exception of some basic phase issues and the lack of any sort of ambience and overhead representation. Fortunately, a smart friend/fan of the band recorded the show out in the house with a portable stereo recorder. This dude made the right decision to record the audio as WAV files and not start by recording them as MP3s. Hot damn. Dude sends me the complete show (one continuous pair of files), which I upload to my system. There's a bit of drift between this recorder and the multi-track, so every few songs, I chop the ambient recording and line it back up with the close mics. Works like a charm, makes EVERYTHING sound better, and now I've got lots of options to play with AND a reference of what it sounded like out in the house. Plus, it now sounds like the sold-out show that it was.

There's not much to do about the compression on the vocal tracks. The SOUND is good (if not extremely boomy), but not anything I can't handle. I just have to be prepared to do LOTS of fader riding. A big issue is going to be bringing the vocal mics down a bit when the guys aren't singing because the bleed that happens when the compression lets up is playing havoc with my drums. That stuff is also causing issues with the little bit of pitch correction I'm doing on a few of the vocals to tighten things up. But overall, there's not too much issue with the picth correction in relation to the ambient mics, which is a nice surprise.

I spent a good bit of time making some volume adjustments to the bass and the guest guitar player's tracks, just to get them to a consistent level with which to work. I also tried to do a little bit of cut-and-paste work to replace the digital clipping segments on the guest guitar track.

I wasn't happy with the sound of any of the recorded guitars, so I spent a good bit of time reamping the "healty" amp with a beefier sound that I'll combine with the originally recorded tone. The combination sounds good and taking the run-through time to do this allowed me to fix a few small mistakes that won't look goofy when this stuff is synced to video later on. After that, I reamped the guest guitarist's stuff, which made a HUGE difference.

I spent some time setting up a reamp sceanrio for the bass and got a tone that I liked, but the reamped version had a clarity that exposed some timing issues and made the music feel worse. I didn't have a real problem with the bass sound to begin with, but thought I'd experiment with reamping it. After a few minutes of listening to it, I decided it wasn't worth the time.

So now, I'm down to dealing with this broken guitar amp. I set up some stuff to reamp that track as well and was able to make a bit of improvement to the sound. As it stands, I might do one more pass of reamping that track with a different sound, just to see if the combination of reamped tracks might mask some of the crackly static. Like the other guitar track, I made some edits to this guitar as I was doing the first reamping pass, but it became apparent that some of this was going to be just plain unuseable.

A call to the bandleader was in order and I explained the situation. He was completely understanding and made the decision to mix the entire thing, with the messed-up guitar, and then make the decision about what to use. Some of this stuff is fine, as the broken guitar amp made completely erratic appearances and isn't a problem on some songs at all, but there's enough of the show where it's a distraction for it to be disappointing. Still, having a conversation with this guy made the whole thing a lot easier.

So, I think that's what it boils down to, as usual: communication. My years of experience, coupled with a few lucky breaks and help from people close-by provided enough opportunity to make most of this this stuff passable and some of it downright GREAT. Being able to communicate these problems clearly, while giving the band their available options made the outcome easier to deal with. I'm still in the organizational stages of this project, but I think we're going to have more than enough great material to make a cool project happen with it and I think they're gonn abe really excited about what we get out of it.

Chris Garges
Charlotte, NC

[cgarges]

EQ the signals on their way to the amp to avoid exacerbating Hi/ Lo end problems.

This is exactly what I did with the aforementioned broken amp issue and it helped in a really big way. We got a much larger percentage of useable guitar out of those recordings that way.

Chris Garges
Charlotte, NC

[Nick Sevilla]

Someoneonce told me:
EMBRACE the defects.
Sometimes you will not be able to make a recording into a generally "acceptable" representation of the music. For whatever reason, the actual recording will not be what you expected.
To me, this does not mean that I have to jump through crazy hoops, especially if I did not make the questionable recording. Instead I try to be creative and figure out what can be reasonably done to make an acceptable mix.
Cheers

[donny]

I've come up with a few philosophies dealing with this (mainly with my own screwed up recordings!):

1. There is always a creative solution to a creative problem! (helps a lot to keep this in mind)
2. Your brain may no longer be the boss!
3. Don't be a chicken-$hit; Do it rough and do it $hitty! (stolen from Gabe Roth)

keeping these philosphies in mind, you have three options:

1. fix the flaw(s)
2. hide the flaw(s)
3. emphasize the flaw(s)

Since I work with tape only, I have some special limitations.

I will give one story:

We had a Halloween party and made a recording that contained a great performance of a new song.

When I went back to the tracks, the drum preamp was cutting in and out (for milliseconds at a time) every few seconds. There was also an obsene amount of bad sounding distortion on a bunch of elements. The vocal was bad and bleeding into everything. the guitarist and bass player screwed up the entire last chorus. the thing was a complete mess. but the performance had a magic quality to it.

First thing we tried was to re-record the entire thing. it just wasn't happening. People started bickering. I kept telling them to play the way we did that night, and they couldn't even remember what they did. the substances we had ingested were of course long gone.

So i remembered 'there is always a creative solution to a creative problem' and went about salvaging the original tape.

here's what I did:

* killed a bunch of highs on the drums & mixed them low
* added acoustic guitar, toy piano & tambourine & sort of played them to the drum rhythms
* overdubbed some flute & violin
* re-did the vocals
* mixed the 2nd chorus twice and spliced the 2nd one in place of the screwed up third
* mixed the track so hot that none of the faders were even moving on the tape deck or the mixer

track came out really good !

[vvv]

Well, I don't know about anyone else, but I'm learning tons here!