Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Into my Dreams - AZMix
#1
[The file quality I normally use was too big to upload. Went with a somewhat lower quality instead. Should still be acceptable.]

So here's my take. Mixed and mastered. Smile

It can really be tough to work on a long multi-track like this, especially one that has lots of parts coming and going. With the length and the changes, I found it really difficult to get a decent picture of what they mix ought to be. In some ways, it really is five separate mixes. I found that I got the guitars or bass balanced and sounding good on one section only to realize that I had upset another. That left me to doing a lot of my basic level adjustments via fader automation. By far, I think that was the biggest challenge, just the sheer scope.

Vocals... Okay, these were... a good deal of work. I usually try to avoid doing much of a high pass on female vocals because they thin out really quickly. That left me trying to figure out how to keep them from being masked by the instruments. Also, I spent a fair amount of time in Melodyne working on these. Were I the recording engineer, I would have gone with a dry, solo playback on these. Helps to better expose problematic areas. There are a few points where a redo might have been the better option. Between the doubles and the chorus, there were also a good number of timing issues. When working with a singer, my observation from the mixes I have done is that it is worth spending time with them to work on enunciation and attack/release. If these vary too much, especially on a double track and more especially at the front of a note, it can be very hard to get them to sound unified when you mix. (eg, "ssssshoould you, wwwwwon't you" vs "should you, wonnn't youuuu-hhhhhh".) Had to nudge a lot of things around and adjust formants and attack duration. These variations in performance also tend to lead to things such as inconsistent drift (like one pass falling off the end of a long note or lifting up at the start of another while the other is steady or even rising a tad in anticipation) and that can make it hard for two parts to sound in proper tune. Over correct and you get Cher instead of Adele. Undecided Most of what I did I think was successful here.

Frequency separation was a challenge, as it always is when multiple synth parts are involved. Not really happy with the job I did there but at the least, if I muted something, I noticed that it went away.

For the guitars, I chose to use blended mics. Some got two, some got three, some got all four. Mostly it was mic #2 and #3. I then bussed each guitar part to its own aux fader and put the compressors and etc over there. Largely, I didn't touch the faders on these tracks once I had the rough mix set. (Largest count of aux faders I've ever had!)

Next problem was getting good bass tone and consistent low end. I tried to over drive it a bit to generate some mid tones but didn't like the sound. In the end, I put it through an amp plugin, compressed it, then ran it through lowender which is always a dodgy proposition. Great plugin, but small changes make a BIG difference. I admit that I never really got this settled until I went to the mastering stage and found I had to stick a multi-band compressor on the main bus specifically to try to tame low and low-mids.

Lastly was the drum overheads. The mics seemed to be positioned higher than I would have preferred. That meant MUCH room tone and the sound didn't fit where I saw this mix wanting to go. For me, with a rock or metal mix the overheads are really cymbal mics and I otherwise build my kit from the bottom up. To get enough cymbals and hat here, I had to catch more of the room sound than I wanted and couldn't really find a way to dry out the sound so I could put my own reverb on the kit. In the end, I think I finally found a good compromise but it wasn't easy. Also struggled with the kick.

Anyway, after about 25 hours, here's my version. Wish I had Mike's level of experience. This tune has good potential but it's above my skill to get it there.

Oh, and a reminder to save early and save often! My DAW locked up my PC at least four times while I was working on this. Oddly, it never just crashes, it always takes the entire system with it as it cries out in pain... Sad


.mp3    intomydreams-mix-20170715-256k.mp3 --  (Download: 32.24 MB)


.mp3    intomydreams-mix-20170717-256k.mp3 --  (Download: 32.24 MB)


.mp3    intomydreams-mix-20170720-256k.mp3 --  (Download: 32.24 MB)


.mp3    intomydreams-mix-20170806-256k.mp3 --  (Download: 32.24 MB)


Old West Audio
Reply
#2
I know the feeling My son a more power full pc for full HD video editing so I swapped and using an old laptop with 2 gig of ram and have to freeze all tracks but still struggling with over 120 tracks with cpu hungry plugins and set reaper to 5 min auto save lol.
Love your mix sounds great on my monitors ,the time you have spent has payed off your mix sounds well balanced some minor things that pop out to my ears ,at 3.32 love the vocal effect ,the synth could come up a touch.
Also at 10.10 the synth could come up just a touch.
Maybe roll off some loose bottom end off the bass and kick below 40 hz as my room is shaking lol , a touch more punch in the 90hz area and taming some resonance on the kick around 170-180hz area could tighten up the bottom end and a db on the snare around 3-7.2k for some air would sound awesome at least on my monitors anyway ?

Great Job Big Grin

Please Help Mike Keep This Awesome Educational Site Alive And Become A patron !
https://www.patreon.com/CambridgeMT/posts

Reply
#3
Hi! Agree with thedon about some bass problems. Maybe it's the synthesized sub octave? Other than that the mix sounds very good. Yeah, it's definitely the bass causing problems. I'd suggest reworking the bass sound.

Can't complain about computer power. I bought a basic $1300 pc from components last summer, installed hackintosh, and haven't had any performance issues. My session in it's current state at buffer size of 64 samples (2.4 ms output latency) takes around 10-20% per CPU core and disk usage some times shows some activity. Mastering chain loads 50% of one CPU core, as one audio object (channel) is processed in one core, and the load can't be spread across cores unless you create a chain of buses feeding the next one and spread the plugins evenly between buses in the chain. Too lazy for that. ^_^
Reply
#4
Okay, uploaded a second version. This should be a big improvement over the last one where the bass is concerned. Problem here is that I did that mix later in the day and wasn't really thinking. I typically want less bass than most people so I have my playback setup set to tamp it down but SONAR doesn't go through that so I was trying to bring it up to more typical levels and overdid it.

Anyway, I adjusted several things.

1) Reworked the bass and got something more to my liking.
2) Finally took the time to get drumagog working again (the upgrade from SONAR X3 broke it) and did a 50% replacement of the kick with a DW52. Didn't even have to adjust the tuning! I just couldn't tighten up the low end of the kick and end up with a sound I was happy with.
3) Tweaked the snare drum sound a tad, mostly with regards to the sustain reverb I have inline with it.
4) Caught my "overhead" compressor behaving badly and smoothed that out.
5) Added in a reverb on the gong hit since it ends kind of abruptly and automated changing the blend from entirely source to entirely verb. Helped a bunch.
6) Oh, dulled that synth a tad and raised it up. I wanted it subdued but perhaps overdid it.
7) Reduced the "instruments" bus compression (API 2500) a dB or so.
8) In fixing the bass, adjusted the multiband compressor that I stuck ahead of my limiter because it wasn't needed quite as much and now functions more as an EQ which means less gain reduction on the L2 limiter that follows it.

Probably random other things I've forgotten.
Old West Audio
Reply
#5
Hello,

I listened to your second mix on my Yamaha HS7 monitors.

First thing that hit me was that the mix seems to have a lot of action on the upper mid freqs ,which makes it a little "harsh" to the ears.
Vocals sound very narrow and nasal.
I like the drum sound.Sounds like metallica.
In general the mix is good.I believe that if you woek the mid freqs thing out ,it will be even better!

Simon
Reply
#6
You know, I'm sitting here working out a few more things and half thinking that I've made this mix too dynamic. Problem is that at this stage, that can be hard to repair.
Old West Audio
Reply
#7
Oh wow, I finally sat and listened to this mix on headphones and couldn't keep 'em on very long, because of all 2-4kHz stuff.

This thing need major surgery. :/

I'll upload a new version in a couple of days if I can keep the mix from falling apart in the interim. (Really stressed lately from work and it makes it hard to concentrate.)
Old West Audio
Reply
#8
Hi AZ,

Another useful strategy for those in power hindered environments is to change the bit depth from 24 to 32bit-float. The CPU will be doing the maths in it's native format which saves it a whole bundle of hassles. But don't upsample from 44.1kHz to a larger one as it's only going to reveal some learning opportunities!

Before you do anything, I recommend downmixing to mono and play it over one speaker only so you avoid comb filtering and see if you'd like to re-visit some of your decisions made in the stereo domain. A mix that works well in mono will sound better than one mixed entirely in stereo. Especially this one. Stereo can give you false assumptions of goodness. Forget not that a room is simply a mono speaker at the end of the day (and a very mischievous one at that!), and not many folk sit in the sweet spot anyway or have ideal speaker placements. Look for balance/level/loudness disparities which are overly zealous. Also check if your vision of depth and ambiance remains reasonably intact in mono. And listen for spectral congestion and masking issues. If this lot doesn't take your mind off work, you've had it

I like your willingness to explore beyond the recordings to get some synergy going and help the listener with sticking in for the long haul. I'd say, just be careful that you aren't pushing the envelope too much and losing consistency within the concept. It could end up sounding just a little off the beaten track in places, like I've hit the track skip button. There were times I felt you were using the tangent button; it's a song, rather than an EP. Try to stay grounded within your vision. Reading one of your posts reminds me that mood and mindset needs to be in the right place before we touch anything. Take a walk, clear the head, read a book, listen to some music. Relax, then mix your butt off by getting into the Zen.

Good luck, by the way.
Reply
#9
In a mix like this, trying to keep separation and isolation is always a challenge. It's trying to be such a grand affair that sounds can't help but try to sit on top of each other. As you say, congestion is one other key concern, trying to keep some frequency bands from becoming just a cacophony of sound instead of music. These are two areas I've been working on. Another evening and I might have it straightened out. My working version already sounds a ton better but there's still room for improvement.
Old West Audio
Reply
#10
I think the mute button will be your friend here too. In today's bottomless hard drives and unlimited budget time-wise, with no studio costs to bear, over production is more likely than under production. Every idea is a good one, and a keeper. On it's own it might be, but all that good stuff has to work in a mix, and there's only so much space between two speakers to do it in. Our brains can only focus so much anyway no matter how much is thrown at them. Bands are not well known for their impartiality. Actually, not many mixing engineers are either, perhaps it's the fear of upset.

If this helps, perhaps keep a focus on Pareto, and don't fret over the 80 percent of the mix that is only contributing 20 percent of the needs of the proj.

Reply