1. Didja accidentally blow through the whole, "We're using our real names" thing on registration? No problem, just send me (Mike) a Conversation message and I'll get you sorted, by which I mean hammered-into-obedient-line because I'm SO about having a lot of individuality-destroying, oppressive shit all over my forum.
    Dismiss Notice
  2. You're only as good as the harshest criticism you're willing to hear.
    Dismiss Notice

Phantom Menace Droid Invasion Mixing Feedback

Discussion in 'Tips, Tricks & Talk' started by Matthias Calis, Nov 10, 2019.

?

Which mix sounds the best to you?

  1. Mix 1

    50.0%
  2. Mix 2

    0 vote(s)
    0.0%
  3. Mix 3

    25.0%
  4. They all suck

    25.0%
  1. #1 Matthias Calis, Nov 10, 2019
    Last edited: Nov 11, 2019
    Hello friends,

    Can I lend your ears for a moment? I've set out to improve my mixing skills and mocked up 4 bars from the Droid Invasion cue from Star Wars the Phantom Menace. Below I've listed 3 different mixes, all of these are "in the box". Which one of these is the most aesthetically pleasing to your ears? Is there anything in any of these mixes that you feel needs to be fixed?

    Mix 1:


    Mix 2:


    Mix 3:


    I've been hearing these 4 bars so much now over the past few days that I don't think I'm capable of making a good objective judgement here, hence the question.
     
    Martin Hoffmann likes this.
  2. #2 Doug Gibson, Nov 10, 2019
    Last edited: Nov 10, 2019
    Sorry. I'm the asshole.

    First, I really would not have never chosen the words "They all suck". That won't help you learn. Be nice to yourself.
    You trying to simulate something every single player in that band spent decades working on too.

    I would have simply said they all sound like samples. 3 is by far my least favorite. For 1 or 2 it's tricky as somethings I like in one are missing in the other, and vice versa.


    My first and biggest question was: Where is my bass? The is the part at @1:25 in the recording right? I need a big fat bottom on that. Sound like just timp. on yours.

    Really.....explore the space Gene. Fella's I'm telling you.....you're going to want that.. Gotta have it baby.

    Remember, I put my pants on one leg at a time just like everyone else. Just when these pants are on they make gold records.
     
  3. On the one in a million chance, my last paragraph was not obvious

     
    Martin Hoffmann likes this.
  4. I think I'm the one in a million then because I wouldn't have had a clue what your last sentence was about without the video!

    You're right, the basses and celli have all but vanished in the mix. I also just realized I completely forgot to put in the piano part! Oops...

    For reference here's the original:

     
  5. My first instinct was to say I like 3 the most, but I think it's really the wrong question to ask and not an ideal way to present the material for conveniant A/B testing. Imho the question should be: "here is the reftrack, which version sounds the closest?" and then do A/B comparisons in one uploaded file where we hear the ref playing first, then switching to your mockup while progressing through the piece and going to the reftrack again. That's how I'd personally set it up in my DAW too: make sure you don't just hear your own piece over and over again, let part of the reference also play over and over again so that your ears can stay anchored to the goal and you don't lose sight of the target.

    I also find it helpful to route A/B comparison of reftrack and mockup playing in sync to SPAN, to visualize differences in the frequency spectrum.

    I can relate, I've been there. I also have experienced tweaking stuff for a whole day, going through half a dozen different versions and realizing in the end that they all are worse than what I had at the start of the day. Don't worry about it, I assume it's quite normal. Take a break of a couple of days if you feel it might help. No reason to rush this imho.


    I would recommend this: continue by mocking up a different part of the piece, where individual sections/instruments are playing more isolated, like around the 2:00 minute mark of the reference, focus on getting room and placement right first with A/B testing with closed eyes and focusing on how the "position" of the instruments "jumps" from one to the other, and tweak till you no longer perceive any movement when A switches to B or you can't seem to get any closer. That might also be the right moment to reach for EQs and try to match the frequency range of individual sections/instruments, as that contributes a lot to the perceived position of them. Then go back to the part you have already mocked up, and focus on balancing volume between the different sections. Then check SPAN (or another spectrum analysis plugin) to see where your mix and the reftrack still differ in the overall spectrum and try to get yours closer.

    I'm still pretty much a noob with this myself, but I would assume this should get you at least closer to what you want, and if I'm wrong I hope some of the pros here will correct me!


    P.S.: In case you don't use something comparable yet, I use the free plugin "panagement" for subtle stereo widening or narrowing all the time when I try to match a reftrack. I think it's a very useful tool.
     
    Matthias Calis likes this.
  6. FYI, I listened to this version



    FYI: the youtube video adds on about 8 seconds timing. It's about 1:26-ish in this clip
     
    Martin Hoffmann likes this.
  7. #7 Alexander Schiborr, Nov 11, 2019
    Last edited: Nov 11, 2019
    Hej mate,

    Good exercises and transcription. Sure getting close to the original..is some sort of pointless because there are too many nuances and factors which you are fighting against. But anyways I guess we are not here to clonewars the recording as this is anyways a waste of time rather than trying to get a bit of the "sound impression". I think one aspect that you need to add is a bit more of space and air and that your version simply sounds a bit too small in terms of spacious appeal. So rather than cloning the original I think the best is to improve your version based on the source you have and accenting the good things in it?

    I dont know if I improved your example at all but I loaded both your 3 versions and the original in my daw and just tried a bit to mess around with a few parameters to catch a tiny bit more of what I think the mockups needs? Shit or hit? I am very limited time right now unfortunately...so I apologize in advance messing around and interfering here..

    Here is my take based on your mp3 mix trying to get closer of what I think the original sounds like.

     
  8. I think it's a clear improvement!
     
  9. Thanks everyone for the responses so far.

    @Doug Gibson After hearing Alexander's version I really can't argue with your vote in the poll. Not that I felt the need to the first time around (I knew it wasn't going to be spectacular upfront), but now I'm feeling like voting likewise!

    @Martin Hoffmann I might've made my original goal insufficiently clear. Essentially there are two goals: first and foremost I wanted to challenge myself to push for realism as much as possible. This is part of a subgoal which is to improve my in-the-box template sound. My goal wasn't necessarily to re-create the recordings, I'd be fine with it sounding like maybe it's a different room, but the primary focus was on trying to make the most out of the libraries I have. A/B'ing could help a lot with this admittedly, and I've taken note of all the advice you graciously offered. The second goal is much more generic. I know mixing is one of my (many) weak spots and I'd like to get better at it. Admittedly, A/B'ing would've helped here too. So, my takeaway is that perhaps I should try to get as close to the real recording as I possibly can because that makes all of this much easier to quantify and measure.

    @Alexander Schiborr Wow, I didn't expect anyone to do what you did. That's great! It's absolutely not shit, it's in fact infinitely better than what I put up and honestly? I'm seriously questioning just what I was hearing because if I go back from your version to one of mine... it's almost like mine are in mono. I will also readily admit it does kind of baffle me because there were a bunch of widening and panning tools involved in my own mixes and I really thought I had it sounding wide and open...

    Inevitably, your version makes me immensely curious just what you did exactly? I'm familiar with a few tricks to make a wider sound, like doing mid-side EQ'ing, generally using stereo enhancers, making source sounds narrower and using more extreme panning, etc.... but clearly I'm either doing it completely wrong, or I'm being too cautious with any of the settings (every time I put on a widener I hear a voice yell MONO COMPATIBILITY in my head). Could you shed some light on what you did? Your version makes what I spent most of my time on (getting the performance right with a whole slew of midi tricks) come more to the fore. I'd imagine it still sounds sample-y to the trained ear.... but I think your version could fool most? Or is it too stiff, do you think?
     
  10. Personally I think this brutal method of A/B comparisons and trying to achieve the impossible is the way to go, because it gives you clear goals, and a relatively clear feedback on whether you are getting closer or not with each step that you take. Those are key ingredients to learning any skill. You'll never achieve the perfect copy with samples, but I think there is a lot of value in trying again and again with different reference tracks. After you've gotten as close as you can, the next step would be to focus back on just the template, without reference, and evaluate how the next round of changes could better highlight the strengths of your libraries and your writing, and how to possibly mitigate the weaknesses of the libraries a bit. I think at that point all the practice you've had with mixing towards reference tracks will pay off in giving you better understanding of what each step will do to the sound and you'll likely feel more confident in your choices.

    What I also would recommend when you mix with reftracks, is putting some plugins on the master that let you compare certain aspects of yours and the reftrack in isolation. like e.g. soloing the side-channel in a mid-side EQ and listening just to how that one sounds A/B, and which elements you hear there in the reftrack vs in yours. Do that for different frequency spectrums, mono, mid, side, whatever you can think of. I've easily spent more time with stuff like that this year than with composing I fear. Frankly it just comes easier to me than composing because goal and feedback are so much clearer.
    With some practice, I'm sure your mixes and templates will get a lot better too!
     
    Manuel Cervera likes this.
  11. Hey - you ripped that off of Mike! :p
     
  12. Sorry, you have it backwards. Although I would agree that none of his mixes "suck", telling someone to be nice to yourself is the exact opposite of be hard on yourself. THAT will not help him learn.

    Want to improve your skillset? Then be hard on yourself.
     
  13. Matthias - as I stated above, I don't think any of them suck. To me, 1 is best (and by no means bad), 3 is decent, and 2 is far from good.

    The first thing to understand (and I am stating the obvious here, but not everyone understands this) is that a good mix starts with good orchestration. I wouldn't say that a properly orchestrated piece "mixes itself", but from the getgo, you are going to be a lot closer than if you have something that is not properly orchestrated. In fact, with the latter scenario, it's possible that you may never get a good mix.

    I take it you did this by ear? It sounds like the piece is not 100% properly orchestrated. It may be close, but if you are - as Alexander made reference to earlier - missing a lot of nuances, and if the chord voicings aren't completely correct, and you just don't have everything down pat, you are already at a disadvantage when it comes to the mixing portion of the exercise.

    So, what to do? The best you can. I will be back later with some more thoughts...
     
  14. My wife just told me that this is not correct and you say: "It is a bit pointless" or "it is pointless" but not some sort of pointless. Man, my English some sort of seems to suck. lol, It is some sort of pointless.
     
  15. #15 Jeff Hayat, Dec 15, 2019
    Last edited: Dec 15, 2019
    I had another listen, and compared your v1 to the orig.

    Here's a short comparison snippet, level matched.



    See how the orig is fuller and rounder, esp in the lows? I am having a hard time discerning whether that's an orchestration thing, or a mixing thing, b/c I don't know if in your piece, certain elements are missing, or just mixed to low.

    Sounds like you have tympani, but that they are not full enough and lack some low end. Same for the basses, and possibly celli. The snare in the orig is a little snappier (you can try using a higher vel. layer if avail, and/or adding some high end, in the 5k range for starters), and the orig has horns panned to the left a good deal, and tho it sounds like you have horns there as well, they lack definition. Also, try using a shorter artic for some of the trumpet notes. If you are using a stac artic and the notes are still not short enough, try Kontakt's TM mode, with settings of:

    speed: 133, tr size: 2.5, transient copy, legato, large

    - as a good starting point.

    BTW - you aren't that far off.

    Cheers.
     
  16. No doubt I like v1 the best. It sounds the most natural and most pleasing to my ears (at least on my studio headphones). The v2 sounds a bit over EQ'ed and v3 somehow sounds like extreme-stereo!?

    I think v1 does need some more "room" though. The reverb is pretty long, but the instruments (esp brass) still sound a bit too close. (My opinion) Might want to cut off some more of the high end. Also, have you tried Mike's stereo-delay trick for panning? Would give it some more width.
     
    Matthias Calis likes this.
  17. I actually posted a 4th mix on my soundcloud quite some time ago but never shared it in here. It's not necessarily closer to the recording (especially the trumpets are not nearly as rounded as the recording) and I'm still missing the piano in the bottom, but on the whole I think this is the best mix so far. All other mixes sound somewhat muffled in comparison, at least to my ears.

     

Share This Page