The Song Fight Bulletin Board

Mostess asked some people to remix his POE tune. I did, and here's what I heard and did.

The Original:
http://www.songfight.org/music/pieces_o ... ss_poe.mp3


The Remix:
http://www.sonofsupercar.org/mostess-poe-01.mp3

Mister Mostess is a dude who likes his tracks. This song had about 25 or so. Everything was recorded in bits and parts, and you could definately tell that he had his ear on a "sound" at the end. When I listened to it and reviewed it, I called it a mushy pile of mushy poo, or something. The original sounds great in the opening stanzas and quiet bits, but totally overloads with bass and whuff in the choruses. The part that really killed me was the big cathartic bit with all the backing vocals. I thought they were really harsh and overbearing, especially with everything else going on.

The other thing that really bugged me in the original was the distorted guitars. They didn't see to add much, and they were just kind of hanging out playing mosquito.

I think, also, in my original review, I said something about how when I like a Bruce Cockburn-ish tune, it's usually because of all the other stuff going on, not just the killer guitar bit. This song definately has an awesome set of acoustic riffs and electric interplay.

So I'm not pimping my mix as anything awesome. I just wanted to illustrate how to clear and clean things up and give the song some space, to try to reveal what was already there and not overload the listener.

The sources have about 10 tracks of guitar, 3 of bass, 5 or 6 vocals, and a handful of percussion.

The first thing I did was to make a bunch of group tracks so that I could compress like instruments together. Then I killed one of the bass tracks because it was just too wuffy and weird.

I ended up with group tracks for basses, percussion, vocals, electric guitars, and acoustic guitars. Each of those got a light compressor added.

Then I went back and figured out which instruments I wanted to be the base (and bass) of the song. I ended up keeping only the tupperware bass drum and a little of the acoustics and bass guitar. Everything else got everything bassy kicked right out. The guitars and vox lost everything below around 400 hz. The bass guitars, maybe 110 or so.

After that, I spent a while soloing groups and submixing each instrument group, trying to find good pan points for the guitars. I also killed the mosquito guitars entirely.

On the shrill harmony vox, I notched them at around 2100 hz and tried to mix them at the lowest level where they still sounded rightish in the mix.

I compressed the crap out of the tupperware bass and really cranked it up. The claps sounded great, so I just panned them and made sure they were plenty loud.

After that, it was just a matter of levelling everything out, and.. done!

Listening back today, I think the acoustics suffered a little. I probably should have not EQ'd them until everything else came in with them. There is a little bit of pluck distortion, too, which is an artifact of overcompression, I think. It could be in the source, but I don't remember hearing it.

I really only mixed this while listening around the 1:10+ mark, because that's the point I was trying to make, and I do think it shines when all the tracks are going. Being able to heard all the percussive elements ~1:50 just changes that whole bit. (And that's what I meant about the non-guitar parts really making yuppiefolk rock ass.)

This is a GREAT song and I really enjoyed mixing it.

Man, nice job blue. (And Mostess, of course! Great song.) Your mix adds some definition that I didn't realize was missing from the original.

I'm curious: Did you stereo-enhance the intro acoustic guitar? Or just run it through a verb? I hear phasing or something that I didn't notice in the original

I want to thank blue for his time, effort, and skill; not just in remixing (which is awesome and humbling) but in describing and publicizing the process. I will study all this and try to recreate it all on my own system.

I'm definitely still a novice to digital recording. The 4-track was easily saturated with sound and it did a great job mushing stuff together so that mixing was breezy and quick. But the computer makes errant sounds stick out unless I work to smooth them out. And a big part of my mixing ear is dedicated to worrying that things don't sound mashed-togethery, 4-tracky enough. My big strategy has been to record lots of stuff so my ear can't keep it all straight---I've been trying to get away from that since it's a pretty stupid (and time/disk-consuming) way to emulate something I could recreate simply by dragging the 4-track out of the basement. I have so much to learn, it's amazing to me that my skills are so fixed by my medium: I used to think digital recording was just another method, but it seems to be its own world.

If anyone wants to try their hand at this, the tracks are still on my server space (http://www.umich.edu/~rbrent/hm/poeremix.zip is a 147Mb zip of a bunch of stereo 16-bit PCM WAV files all starting at zero-time and lasting about 3 minutes). I also want to thank blue for not publishing that (I was worried the sysadmin might grumble, but no complaints yet so I'll push it farther!) The file will disapear at some point for that reason.

I find mixing harder than writing, performing, and recording. I imagine good writing, performing, and recording makes mixing easier, so perhaps I should focus more on those skills. But if better mixers than I can turn this pile of tracks into something coherent and pretty, then there's more reasong for me to work on my mixing skills, too.

Mostess, I'm not having any luck with that file address..says it ain't there no more...

Post edited. Try again...

Think it'll still be up in a few hours? I'd like to check out the tracks, too, just for fun.

Days, probably.

cool...downloading it now. I'm gonna work this up too. I really want to listen to what Blue did, I'm sure it's awesome, but I won't until I get through with it, don't want it to colour my muse.

Putting up source for remixing is cool. More people should do that.

Mostess wrote:If anyone wants to try their hand at this, the tracks are still on my server space (http://www.umich.edu/~rbrent/hm/poeremix.zip is a 147Mb zip of a bunch of stereo 16-bit PCM WAV files all starting at zero-time and lasting about 3 minutes). I also want to thank blue for not publishing that (I was worried the sysadmin might grumble, but no complaints yet so I'll push it farther!) The file will disapear at some point for that reason.

a quick mail to spud or jb will solve your music-sharing-related bandwidth problems.

Here is the Bell Green mix:

http://www.yogarup.com/music/poebgmix.mp3

And the mixlog:

13/10

Have to set up the technical side of the mix first and plan what I'm going to do so that my workflow is not inturrupted. I don't want to have to deal with any technical details when I'm in my creative mode. Things to set up:

Set up folder and import tracks
Sample rate convert ('cos my software won't do 22050)
Label all the tracks
Set up effects busses and signal processing
Set up group busses and label them
Plan when to do the actual mixing

I don't even want to listen to the song before all of the above has been set up because listening to the song is really the beginning of the mixing process.

Next, scrub through each of the individual tracks to check for noise, clicks and pops etc. Maybe normalise if necessary. And of course make sure that I'm ready to mix! If I'm tired at the end of the day, that's not a good time. Ok, donkey work done.

17/10

Well, donkey work still isn't done as a friend came to stay over the weekend and we finally recorded one of his songs. So today after work will get all the file management done and set up all the busses. Take your time, don't start listening yet. It's worth putting in the preparation time.

Line trim adjustment. Not really, i don't have an analogue console, but I just go through each individual track in isolation pushing up the fader just so it's into the yellow, but only just. Then mute it moving on to the next one. As well as getting a basic working level for each track/instrument, it also gives me the chance to audition each one very briefly.

Panning. Just playing with the pan pots with groups of instruments. With the three acoustics two worked well together and enhanced each other, so the other one had to go. One of the claps had thumps and thud sounds, so that had to go as well. Better to listen as you play with the pan pot rather than look at the pan pot. Finger pointing to the moon and all that. And so on to the end.

EQing. Cutting and boosting to get the sound I want. Bass low has a different sound so out it goes. Hard to choose between bass 1 high and bass 2, so I'm going for an eq'd bass 1, but bass 2 will come in now and then. Some more thudding and thumping sounds on the lead vocal track, may be due to foot tapping or hitting the mic stand. Most of the tracks have been well recorded so eqing actually makes a difference. If on the other hand they had not been so well recorded it's more difficult to cut or boost frequencies that aren't really there.

Compression. Just to enhance the sound where it's needed. And again very subtly so as not to squash the life out of it. The compressor has to move with the rhythm and dynamics of the song. This song doesn't need anything to be really squashed.

Basic/static balance. When I listened to all the individual tracks the one that really stuck out for me was the "tupperware". I don't really know, but I just really liked it. So that was my reference from which I balanced the other tracks. This song has a lot of swing to it and it's shaky and groovy and the tupperware track really had all of it. So that's where I started.

Reverb and delay effects. The original recording have a fair amount of room prescence in there so I wanted to keep that and really just try and enhance that. So this is purely for effect rather than to make it sound natural. Acoustics, lead vox, tupperware and soda can. Oh and the bottle. Nothing else goes down the effects bus.

The fader moves. The lead vox moves the most and some of the other things just to say "hi here I am". Other than that it stays fairly close to the initial balance. Before the fader moves I have trimmed each of the regions so that it improves clarity and reduces noise. I couldn't really be bothered by automating mutes, although I usually do that.


It's a great song and a great performance. It swings and grooves and I just wanted to try and bring that out. Didn't bother with group busses in the end either as after taking out 3 tracks ac guit 1, bass 1 low and clap 3, 21 tracks I found to be fairly manageable. And this is only one mix. If I did it again it could be quite different perhaps with more dynamics or effects. Perhaps even stripped down to guy and guitar thing. And this is how it works in the "real world": the musicians rarely mix their own works. The recording engineer probably won't mix it either.

Well, hope you enjoy what I've done and feel free to ask any questions.

Bell Green wrote:And this is how it works in the "real world": the musicians rarely mix their own works. The recording engineer probably won't mix it either.

I can see why. Both your and blue's mixes are stunning to me: I never would have chosen to bring things out so clearly. I'm obviously trying to hide the things I'm embarassed about when I mix. Not a good idea, apparently.

Funny, too, that you both took out the buzzy electric guitars (Bell Green, you really mixed them almost inaudibly low: good idea). I recorded them specifically to try to saturate the overall sound, like sand to fill the wasted spaces in a glass of rocks. You both used reverb for that instead.

This is awesome.

I don't think you need to fill up any sound, dude. You could probably just do 1 bass, acoustic, perc, and maybe 2 vocal tracks and every song would still sound great. I bet your songs would sound great with organ as well.

Thanks, Blue, for initiating this madness. Yes, it's very real-world.

There's a Glen Ballard quote in "The Art of Digital Music" that gets into whether artists should hand the mixer/engineer/producer their whole set of takes and whatnot. He feels the artist should already have his tracks in order, and have self-edited them so that the mixer can focus on the creative process of mixing, not arranging.

glen ballard is a wise man. plus, he makes good music!

i'd also add that learning the production process (arranging, mixing & mastering) as much as possible adds to one's own knowledge about what's possible, what's realistic in terms of timeframes, cost, etc.

i know stuff about cars but i still take it to the shop for stuff i don't want to deal with, or stuff that someone could do better than me. makes more sense than just handing it off to them with out having some idea of what they are going to do.

i doubt i'll ever be a top notch producer, but the more i do it, the better i get i think. and i enjoy learning about it.

knowledge is power!

roymond wrote:He feels the artist should already have his tracks in order, and have self-edited them so that the mixer can focus on the creative process of mixing, not arranging.

that's probably a good way to get the same mix from many different mixers, but it seems like the opposite of the art of mixing. for instance, i don't normalize anything any more, because i have a sneaking suspicion that the amount of interpolation required to make very quiet tracks very loud, especially as implemented by base DAW software, goes a long way towards making things sound digital and crappy. so now my mixes are very quiet right up until mastering, when i at least have more headroom occupied by mixed tracks. i like plenty of room at the start and end of each track to mess with, and i hate getting stuff that's already been panned.

there were plenty of mistakes in this tune that M could have edited himself, and i left all of them except for one rogue bass fill, which i copied over with the second bassline. but after the song was all assembled, i felt like, even though the timing was a little disconcerting in places, everything added up to an even cooler pirate feel. YAR.

to me, a big part of the art of mixing is finding the cool bits that should have been left out and introducing them as elements into the song. sos' tracks are loaded with mistakes or cool bits that got the ctrl-c ctrl-v treatement dozens of times. i always wants to give the artist back more than think they put in, if i can.

but Mostess' tracks are not typical. this was an extremely well recorded and arranged tune from the outset. (mosquito guitars notwithstanding )

deshead wrote:I'm curious: Did you stereo-enhance the intro acoustic guitar? Or just run it through a verb? I hear phasing or something that I didn't notice in the original

uuuuuhm, i don't have the mix in front of me, but i hear what you're saying. i think maybe M didn't turn his second acoustic track on until he started singing, where i had them both going from the outset. if you listen to his, you can hear the acoustic start right, move left when he starts singing, and then you can hear the doubled sound during that first verse.

Mostess, you should do a sample CD of handclaps and tupperware percussion. What mic are you using on those?

blue wrote:

deshead wrote:I'm curious: Did you stereo-enhance the intro acoustic guitar? Or just run it through a verb? I hear phasing or something that I didn't notice in the original

uuuuuhm, i don't have the mix in front of me, but i hear what you're saying. i think maybe M didn't turn his second acoustic track on until he started singing, where i had them both going from the outset. if you listen to his, you can hear the acoustic start right, move left when he starts singing, and then you can hear the doubled sound during that first verse.

Mostess, you should do a sample CD of handclaps and tupperware percussion. What mic are you using on those?

Apologies to blue: all the WAV files I gave were stereo: panned the way I had them in my mix, with the exception of the three acoustic guitar tracks which were recorded with two mics, one L and one R.

For my mix, I cut two of the acoustic guitar tracks for the first verse and supplemented the remaining one with a heavily compressed copy of itself so it didn't sound so thin. The compressed bit dropped out at the start of the 1st chorus, when the other guitars join in. I'll bet the phasing or whatnot has something to do with my double-micing setup.

I mic everything with my favorite: a NADY PCM-950 condensor. The guitars I double mic---the NADY and a TASCAM PE-50 condensor. The latter is a hilarious mic. I've had it since 1989. It takes a AA battery instead of phantom power. It has a switch to toggle between two response-curves (one labelled "voice" and one labelled "music": as if the two are mutually exclusive); when its battery is low, it has some pretty interesting properties, I used to use it as a cheap compressor in my starving 4-track days.

This is the first song (besides "Prayer for Pancakes") recorded in our new house in Michigan. The recording computer is in a pretty large room (20'x22') with big rugs and some padded furniture at odd angles. The crown moulding is a very large, plaster concave curve all around the room and the most logical place to record in about 4' from a corner of the room. I was so psyched the first time I clapped my hands in the room: lucky to have such a good space. Glad it's not just me.

blue wrote:

roymond wrote:He feels the artist should already have his tracks in order, and have self-edited them so that the mixer can focus on the creative process of mixing, not arranging.

that's probably a good way to get the same mix from many different mixers, but it seems like the opposite of the art of mixing.

Yeah, I didn't mean applying treatments and having a final mix before passing it on. What you're describing isn't mixing as much as arranging from the raw pieces then mixing as you go. Which is great and all part of the fun. I think passing over complete tracks that can be dropped or panned or treated or compressed is where he was going. A little of both, but not the whole kitchen sink. Plus a bathroom one. But one really can't draw these lines and the interaction between all the players is what eventual creates the masterpiece.

This got sorta buried in the Fight thread:

Hostess Mostess - Pieces of Eight (Future Boy : Buried Treasure Mix)

Enjoy, mate.

I guess I should provide a bit more explanation, to go along with the previous two mixes in this thread.

I did a remix that involved giving the track a different feel. The only things I used from the original source were the vocals, two of the guitars, and samples of all the percussion. For the percussion (aside from the claps in the bridge) I sampled a single hit and then sequenced new rhythms in Fruity Loops.

All of the synths were also sequenced in Fruity Loops.

The lesson to learn from me is more about the recorded matertial itself than about how to mix it.

You didn't play with a click track:

This made it impossible for me to use the tracks as is and sequence electronic elements against them. My solution was more radical than most: I hand edited the guitars to get them to fit into a tempo grid of 152BPM. That means at almost every single eighth note I split the track and shuffled it into place. While this was very time consuming it did afford me the opportunity to put in lots of subtle glitch, which I wouldn't have done otherwise. I also had to adjust the vocals, but they were much easier to do.

You didn't tune your guitar with a tuner:

This means that when it came time to layer on synths, I had to fine tune them by ear so that they wouldn't sound out of tune with your guitar and voice. The guitar was about 30 cents sharp or so. This was annoying because I didn't know where the fine tune knobs were on the synths. However, I managed to find them.

Those were the two big issues. Once I got past those, it was a piece of cake to do what I liked. The quality of the recordings is better than you give yourself credit for, I think. The guitars, in particular, sound beautiful with no effects on them at all. I maximised them and reversed the stereo on one, that's it. The lead vocal was a little stuffy sounding but not awful. If you aren't using a pop guard you should start or simply go a little softer on your consonants. I edited out the harsher Ps and Cs.

It's an awesome song and I'm glad I had the opportunity to remix it. Very catchy.

seriously, you're insane.

Future Boy wrote:I hand edited the guitars to get them to fit into a tempo grid of 152BPM. That means at almost every single eighth note I split the track and shuffled it into place.

Hey! I spent hours putting those individually played samples of single acoustic guitar eighth notes in exactly the right non-metric times to sound like a sloppily-played guitar. To undo such a thing of beauty...

This is so fun. I can hardly type, I'm having too much fun listening to this.

I love what you did with the bridge: I can't figure out what notes those triangle-wavish high synths are playing. Is it the same as my "twang" guitars but 8va higher and with some strong harmonics?

The third "I don't wanna cry" in the second-to-last chorus is hilariously syncopated. And the ending vocal noodle is shifted a beat late which is also hilariously off-center. In general, it's incredible to hear my voice and guitar being played like puppets, moving against my will in ways I wouldn't have done spontaneously. I recommend the experience.

The breakdown last verse is growing on me. Sounded disjointed the first time around but I'm getting the flow after a couple. It's fast becoming my favorite part.

The resonating synth on the chorus is fascinating. You're a good player with a good ear for melody. The bass line on the verse fills in the holes very nicely.

Thank you for your time and efforts! I can't say I learned much about mixing from this, but it will get a lot of play at our house. I have no doubt the 2-year-old will prefer it to dad's old sound.

Glad you liked it!

The high synth in the bridge starts out playing the same notes as the twang guitar, I think. I didn't copy it completely on purpose. The first harmony note that comes in is a 6th lower and the second harmony (4th repeat of the four chord structure) comes in a 6th higher. So the chord is voiced C#AF# and then just descends more or less following the harmony of the strummed chords.