Recently Chris ‘Von Pimpenstein’ Carter was commissioned to mix a dance record for Australian recording artist Max Leon. Chris was nice enough to talk about the process and give some insight into his approach on mixing this record. To get some background on how Chris thinks as an engineer, you may want to check his interview, Psychology of a Mix Engineer: Chris Carter.
If you would like to contact Chris please visit his website.
Upon asking him how he landed an Australian gig, Chris said that almost half of his work now comes from overseas.
“I do a lot of international work. I think it all started when I did a record that did really well in Italy and from there other people starting contacting me.”
With the rise of the Internet, mixing engineers have become more accessible to a large number of artists around the world. But that’s not enough to secure the gig. As an engineer you need something more to instill confidence in your prospective client.
“I’ve gotten used to working with people from other countries so I’m good at dealing with issues that tend to come up, particularly with language barriers which I’m good at managing. I think because of that people tend to feel comfortable working with me and that’s why I get so much international work.”
Title: Todo Para Ti
Artist: Max Leon
Producer: Sebastian Ivanov
Label: 60 Seconds Music
Mixing: Chris Carter
When the management contacted Chris for mixing, they were pretty specific about the sound they were looking for.
“They actually provided a lot of detail about what they were looking for. They sent me a reference track with a chorus that they wanted their track to sound similar to.”
However, in this situation it wasn’t fully possible to achieve the sound without making production changes. This is why sometimes it pays off to contract a skilled engineer – especially one with production knowledge – because they can go beyond just the sonic objectives to help achieve the overall vision of the project.
“The only problem was that the record wasn’t produced the way they wanted it to sound. When I listened to the reference, the chorus was a big vocal stack and from the recorded parts that I actually had, it was like a doubled loop. I had to tell them it wasn’t going to sound like that and they understood.”
With some guidance from Chris the management decided to re think the production.
“I told them from a producers stand point, what really needed to be done in order to get it to sound the way they wanted it to. So they came to the conclusion that it was better off to just start the record from scratch. So they hired Sebastian Ivanov to redo the record.
The lyrics, the melody and some of the music ideas stayed the same but the rest of it was completely changed. When I got the updated record I had to start that completely from scratch and had to re do everything. I actually have a co-production credit, but it is largely related to some logistical issues and guidance I provided during pre-production – the creative production was all Sebastian.”
Chris has a few things in his mixing process that he is very particular about and they are implemented on every mix that he takes on.
“First, you will notice that I use the channel trims A LOT. I’m very careful about gain structure. It makes a big difference at the end of the day. I don’t care what computer scientists say about floating point and unlimited headroom; it makes a difference that is clearly audible no matter which DAW you are using (not to mention essential when interfacing with outboard gear!). As a result, you will also notice that the bulk of my faders are hovering in the ballpark of -0dB. Another thing you will notice is that I pretty much only pan hard left, hard right and straight up the middle. This is a trend I do on pretty much all records, except ones that specifically need an all-over-the-place approach (e.g. recreating a live soundstage, etc.). I think I only have one effect return that is panned something other than true LCR.”
When asked about any obstacles when mixing this record, here’s what Chris had to say,
“The biggest challenge was that there were a lot of vocal stacks and they were all doing something different while playing at the same time. I had to make all that stuff meld together and make sure that each part was audible in the way it was meant to be. Its one thing to have a stack where all the vocals are singing the same thing but when you got 4 different things going on it becomes a lot more challenging. That was by far the most difficult thing in the mix.”
Chris says that he prefers to receive tracked out sessions without any effects but the stems for this session came with everything already printed.
“99 times out of 100 this will bug me, but Sebastian did a really good job with them and erred on the side of conservative. He has a surprisingly good ear, so it didn’t cause the kinds of problems it usually does. The piano track caused a little bit of an issue, but I was able to get the dry version of it and blend it in parallel with the processed one to get the right amount of effect necessary.”
Tracks 1-7 [Reference Material] – “I’ve got a David Guetta track, a Rihanna Remix, a dance remix of one of my records, The original rough version of the song and then the producers rough mix of the new version of the song.”
Track 8-43 [Drums and Music]
Track 44-63 [Groups] - “There’s a lot of groups on this record because there’s a bundle of vocals.”
Track 64-100 [Effects Tracks] - “The reason why there is so many effects tracks is because There was very little sharing of reverbs and delays.”
Track 101-190 [Vocals] - “Not all 90 tracks are being used, it’s more like 86. A few of them are turned off because I had outboard on them which got printed back.”
Chris isn’t a stranger to high track counts; in fact it’s the norm for most pop records that he receives.
“The high track counts are what keep me working. There’s a lot of guys that can mix a rap record and 3 vocal tracks but then you hand them a straight pop record with a billion tracks and they don’t know what the fuck to do – So I get the job [laughs].”
This 808 kick only plays a few times in the song. It’s subtle and just adds some emphasis to the main kick at key points in the song. I added the T-Racks Black 76 plugin because I needed more body and sustain out of it so I that I could just tuck it underneath everything and ring a little bit more without poking out. I’m taking out about 5 dB of gain reduction.
This kick goes (along wth the main kick and hand claps) out to a Soundcraft desk and then to an Aphex 104c2 (settings are Drive 12:00, Mix 12:00, Tune 10:30, Mix 12:00, normal harmonics). Normally, I would have just boosted 35 Hz on it, but since the main kick was going out to the Aphex, in order to maintain absolute phase I sent this 808 out as well and just skipped EQing it. In fact, the EQ was left on the insert; I just wound up not actually using the EQ.
I used the Kjaerhus Classic Compressor to expand the dynamics, specifically, to accentuate the high end attack portion. The attack time is about 5ms and does the same kind of action as a dbx160x/xt. There’s a little boost at 35 Hz for some extra lows, but since this kick ultimately goes to the buss with the Aphex104c2 on it, I didn’t need to add too much down there. The boost at 10 kHz helps bring out the attack.
It’s important to note that the compressor is acting almost more like an EQ than a compressor, so it’s the combination of the compressor and EQ (and the Aphex) that makes the high end stick out on the kick. This is what cuts through the mix and makes the kick sound like it’s bangin’ you upside the head. The fader plug-in is there to trim my fader automation without actually touching the automation itself.
This snare comes in only in the choruses for some thickening – it’s this super rich gated snare with plenty of lows. I think I ran this through an outboard compressor I have that I use for tone and printed it back. A little low end boost that is incredibly subtle, then a dB of reduction on the T-Racks Black 76 plug because it wasn’t quite there yet.
This is the melodic bassline in the chorus. I cranked 35hz and because the bassline was a little light to start with I needed more so I then added a little boost at 30Hz. The bass pulses opposite the kick and Sebastian printed it with the sidechain pulsing, but it wasn’t enough. Better that he didn’t go far enough rather than too far! So I copied the main kick, gated it and used it to sidechain this bassline some more and some other instruments as well. I just used the Cubase compressor for the sidechaining to get the pulsing.
This is a synth playing the same thing as the bassline, but a couple octaves up, so it’snot really a bassline; it’s a lead synth. It was printed as a stereo track, but only on the left side (right side had nothing). So I used Prefix to get it into the middle (because I needed the panpots to be L/R). Then I set up a Haas delay to make it big and stereo then added a bunch of lows and highs – the highs really bring out the dirt.
“The vocal tracks are all Max with a few additional vocals by Sebastian, just tucked in there for a little additional oomph. I don’t know the signal that was chain that was used while tracking the vocals, but it sounded like there wasn’t any compression. I mean they were recorded well but no special processing as far as I know. I think they came to me without any EQ or compression; it was just straight mic.”
These were triple tracked, but I used two and disabled the third. Although I frequently will treat a lead and it’s double differently, in this song I treaded them exactly the same. You don’t see much for plugins because it was mainly done outside the box. Both went to 1084s with HPF@70Hz, miniscule low cut at 110Hz shelf, tiny boost at 4.8kHz, and a tiny boost at 10kHz shelf. Then they both went to two sides of a Pro VLA (in dual mono mode) getting 5:1 compression with what passes for a fast attack and fast release on that box (3ms, 300ms or thereabouts….). I don’t remember how much gain reduction, but probably in the vicinity of 8 to 10 dB on the louder portions of the verses, maybe more. These were printed back to new tracks (I saved the old and wrote down my settings in case I needed to go back). The new tracks got some subtle de-essing and the fader plug-in was for trimming automation.
This is the first half of the stutter vocals in the breakdown (the second half is a vocal stack) which I had to edit myself. I used a couple snips of Max’s vocal from elsewhere in the song and chopped/copied/sliced/diced/flipped/reversed/etc. into what you see. It got so complicated that I had to render some of the edits just to keep it sorted out, so there are really twice as many edits as you see.
On the inserts I used some DeEssing followed by some basic compression which was really there just to level the thing off because all the edits made it a little jumpy – strictly dynamics control. Then I used some distortion from Devil-Loc, which is also a compressor. Without that first compressor, the level going into Devil-Loc would have been too inconsistent causing too much of a change in the amount of distortion than I wanted.
There’s a bunch of automation. Fader and pan for obvious reasons. Then there are several sends that turn on and off at various points and also have the send and/or returns panned and fader automated, etc. It’s so complicated it makes my head hurt and just trying to put it in words is something that makes me sick to my stomach.
The “stutter ghost” track below the “stutter” track has the fader at -inf, but has a pre-fader send to a super long 8th note delay with a huge plate reverb followed by ridiculous amounts of compression. All for the purpose of making sure the sound basically NEVER decays and then the ubiquitous fader plug to trim my fader automation. This creates a ghost-like sound of one word (what word? Hell if I can remember) that I could control and ride throughout that section with fader automation. The sound (via the return) starts a few seconds after the actual audio plays and slowly builds to a climax and tails off at the end right before the vocal stack stutter part (oh oh oowoohh) begins. It helps build tension for the listener.
Normally doing stutter edits isn’t really part of mixing, but this was an unusual situation.
This is the same as the All Vox buss, just with music. If I hit mute, I get the a cappella. Slate VCC here is linked to the Vox buss, so they share settings.
The 2buss feeds my SPDIF output. I have Slate VCC which has the meters essentially pinned during the chorus sections. Then I hit a HPF at 20Hz which is about 6dB/octave. It is basically inaudible, but it does some kind of monkey voodoo that I like.
Then I have The Glue doing maybe 2dB of compression. I don’t use the 2:1 ratio too much because it’s got such a soft knee, but it worked for this song. That’s why you see the threshold so high – because of the soft knee grabbing everything.
That’s followed by Ferric TDS. I’m using it strictly for the dynamics (the saturation is set to minimum, although even at that setting the needle was wiggling a little). I use the 12-string preset, which sounds crazy – I stumbled upon it by accident one day. Then I adjust the dynamics knob until I get what I want (in this case, about 2dB of reduction) and the saturation until I get what I want (in this case, almost nothing).
Last, the only actual hardware being used at mixdown is the Aphex box. Whenever convenient, I try and print back my outboard when I get to a point where I think I have a mix for approval. It makes it easier for me to switch between projects without having to recall those manual settings, or worse, recall them inaccurately.
The reason the Aphex isn’t printed is because it’s on a buss getting multiple tracks, so it’s not convenient to do. That said, I never change the settings on that box; I haven’t moved them in years. One comment about the Aphex that is important is I set it for +4 operation, but have it connected with unbalanced cables. It’s a little gain staging trick that turns it from a rather annoying box into a killer box with my “preset” setting on it.