Okay, first and foremost... no need to be at each other's throats. We're all here to learn from each other, and improve on our skill sets.
Your video examples, albeit fine, do sadly not cover the topics that I brought up. Let me please try to explain a bit better what I've meant, and back that up with examples of possible tools/settings to try.
Topic: pumping
It is a perceived effect. You start with a sampled rain, and then there is this sudden jump in volume with the piano, that also feels heavily processed on top of that (I am not familiar with the FL Studio instruments, so I can only go by "what I hear"). Then the drums set in, and due to their core sound (they are already fairly processed to begin with), this gives the impression of things being pushed too hard - or "pumping" rather.
I don't know how to address this, other than maybe a transition effect, or maybe less compression on the piano (at least in the beginning, of that is possible). But this was my initial impression. There is also no reference point in terms of beat / rhythm in the beginning. And this in turn throws you off.
Topic: "too mid-range focused"
You can show me the most perfect frequency plot / curve on any FFT analyzer, and yet my ears still tell me that certain frequencies stick out and make this mix sound "honky"/"nasal". Things that can easily be fixed by tweaking some frequencies. A very quick and dirty example that you could try, is with any standard full parametric (node) EQ and the following settings:
Band A (Bell/Peak): 636 Hz (too much in the 630 Hz to 670 Hz range can often quickly result in a "honky mids" feeling), default Q (bandwidth), -2,2 dB
Band B (Bell/Peak): 1,86 kHz (I went by gut feeling here), narrow Q (I used a bandwidth of 4 from-0,5-to-10), -2,7dB (this is a "resonant frequency" that stood out to me)
Band C (Bell/Peak): 5,65 kHz (this is the "upper mid" range, where the snare hits and hi-hats are quite aggressive), default Q, -1,6dB
Band D (High Shelf): 3,5kHz, wide Q, +2,5dB (I would actually use a parallel EQ for this, as it would work more subtle for "lifting things", a "series EQ" is more aggressive/noticeable)
Try it, bypass the EQ ever so often, then you clearly hear what I mean by the "mid-range heavy" focus. With reducing these frequencies on the sum (full mix), I created a (to me) more enjoyable listening experience. My assumption is, that this build up mostly comes from the drums, and the brass sections, also the piano (playing lower notes). In fact, while experimenting, I even used SlickEQ GE to "tilt" the frequencies just a tad (center point 640Hz, +0,25dB), and also narrowed the stereo field with bx_control (Mono Maker up to 160Hz, Stereo Width reduced to 85%).
However - this is not a manual (as in: you have to do this) - it is just what I would do to elevate your mix even further.
Topic: mismatching rooms (reverb)
I had to download MT Powerdrumkit 2 to understand what is actually happening under the hood (I don't know every tool out there). The reason why I brought up
"your instruments feel like being in different rooms", is because "ready to go" drum kits like EZ Drummer and co, usually come with a certain room sound "printed" - and you have barely any, if no control over that. This is also the case for MT Powerdrumkit. There is no room mics, the "room impression" actually comes from the snare, which features a very gated reverb (no close microphone drum kit setup sounds like that!).
And while this is a common trick during mixing (e.g. give the snare an additional plate reverb to add to the "stadium rock" type sound, make the kit larger than life), in this case it's actually not good (in my humble opinion and all that). The reverb feels artificial, too "mid range heavy". It clashes with the rest of your used instruments. In fact, the later section of your song feels like you're using multiple different "rooms" (reverbs) to set your instruments apart. The reverbs are also very bright. This doesn't give the listener a sense of cohesion.
This is where I would make adjustments:
First, I'd address the drum kit. Make the kick a bit less "clicky" and more "mellow" by looking at the frequencies 50Hz (bell, default Q, dropping by about -2dB to reduce the "boominess"), then 325Hz to bring out the "body" of the kick (bell, default Q, +4dB to even +6dB is a good start), and finally around 5,4kHz to reduce the "click" (bell, default Q, I tried reducing by -2,5dB).
To tame the "predetermined room" that the drum kit sits in, I would then use a transient designer on the snare drum. I actually tried various gates first (including Kilohearts Gate and Slate Virtual Mix Rack with "Gate:Drums". which also has a "DeBleed" function). However, I got the sound that I wanted with a simple transient designer. In my case, I used old
Sleepy Time Records Transient, default timing, reduced the attack by -10% (no modifier) and the Sustain by -50% (with the x2 modifier on). Suddenly way less "printed room", therefore giving me way more control over your overall room placement with dedicated reverbs! The rest is down to proper kit piece levels, EQ and compression.
Speaking of:
I would pick two "rooms" at maximum. One "global" one (moderately big and warm), and one "medium" one for various more "subtle placements". You can even have an additional "long one" just for the snare again... but the focus should be "warm", not "bright". Maybe even look at the "Abbey Road Reverb Trick" (if you don't know that one, basically: the signal is being filtered with a Low Cut at 500Hz and High Cut at 10kHz, sometimes even with a bell-EQ dip at 1,5kHz before hitting the reverb, which was also "dulled"/high frequency cut to taste - this reduced excessive frequency build-up and worked better in a mix). Then place your instruments as you seem fit (the more reverb, the further in the back things feel, of course).
The idea here is "one coherent room". Like with a Salsa / Big-Band.
And finally - the topic: Loudness
I am absolutely not arguing here. I wrote pages upon pages on various communities, white papers, I regularly talk about this on the Discord servers I'm on. The platforms I mentioned do indeed lower the loudness, but they also don't remodel the lost transients from over-processing.
Looking at your mix:
I could only download the MP3. Your named the file -14LUFS, but analysis showed that it's actually -15,9LUFS ILk and the maximum peak does not exceed -6,4dBTP. So I don't know what happened here... either SoundCloud "messed" with the file even further, or you uploaded it this way. Maybe double check that.
If I assume that the original max peak was -1dBTP, then the original LUFS was something around -10,4 LUFS ILk, and things have been pulled down to -15,9 LUFS ILk. Due to the nature of the processing (limiting), this removed a possible 5dB to 6dB of healthy transients - making things sound less dynamic and actually more quiet (once "just pulled down").
There is really no argument around that - the following might be an old video, but it brings up the problem of the topic "just master loud" quite nicely.
Youtube:
Music Streaming Services: Bring Peace to the Loudness War by Matt Mayfield Music (once again), from 2017.
I
do not give anything about the so called "debunking videos"/"I know better than thou" comments from whatever direction, and that
"the ideal loudness one should master at is -8 LUFS - even <insert high class audio engineer> does that!". This is so untrue, it's not even funny! And this stance goes against
all releases prior to the year 2004 (where the insanity to "push for loud" really kicked off without any sense of limits). This also goes against format limits (especially tape and vinyl, even MP3/OGG/M4A has limits! Especially max peak - peak reconstruction during decoding can result in drastic ISP clipping incidents). In fact, more and more platforms now straight up reject mixes that go "over" certain given parameter limits. (especially Dolby Atmos mixes, which is currently -18 LUFS ILk absolute max, and the majority of streaming platforms really pushes Dolby Atmos as of late).
Fact is: transients = more impact, less squashing = less ear fatigue, better listening experience.
(some prime examples are: Metallica - Death Magnetic initial CD release versus Guitar Hero DLC - which is over 7dB more quiet, or Carpenter Brut - Turbo Killer global release vs official music video, which is 4dB more quiet!)
Fact is also: the Mix(ing) Challenge has loudness limits for various games. In case of the Songwriting Competition, that merely only results in a loss of bonus points and not in a "tagged disqualified" scenario. It is your prerogative which route you decide to go.
Mixes can still be -14 LUFS and both "loud" and "punchy as all hell" - the early CD era (1990s) and the revival of vinyl proves that.
Please understand, I will stand my ground on this.
I hope this extensive bit of feedback helps, and you might be able to address the one or another thing with your mix by the end of the game.
Have fun!
And I really hope to hear more entries from other participants as well!
BEFORE WE GET STARTED:
SONGWRITING COMPETITION THREAD INDEX
