I appreciate the update. Now we can get a better picture of what tools you used, to get to the sound that you did.burakxmusic wrote: ↑Sun Sep 22, 2024 00:30 CEST@Mister Fox Thank you for your valuable feedback. I updated the song and explained why I mixed it this way. I hope you will support and support my innovative approach.see original post.
Kind regards
I invested a couple of hours to take a closer look (listen) and tried to maybe adjust some things for a better listening experience. Yes, your new mix is "more centered". But the way you've used phase shifted, chorused and micro-delayed signals, might not work on a lot of playback mediums. My impression is, that you tried a somewhat "made for headphones" mix (or ear-buds mix, rather -- which explains the focus on upper mids and sometimes really excessive bass). This is my only explanation as to what is happening. A lot of the instruments are pushed "behind" your head, rather than spaced "around" (for example the "blip" synth in the last third of the song)
Here is an attempted "fix", based upon the m4a file you provided (you didn't provide a WAV file this time). A lot of work went into this, so let me please go a bit more into detail as well...
phpBB [media]
File download, WAV file, 32MB, 44/24
To follow along, I also provided screenshots.
(please note: the screenshots will be auto deleted in about 2 months)
The first thing I noticed with your new mix, is that it's very low on signal this time. It's been at -31 LUFS ILk (loudness), and the maximum signal signal strength has been -12,4 dBTP. Before I started working, I boosted the signal by at least +9dB so that I have more reasonable threshold settings, and the analyzers I've used could have a better resolution as well.
The following screenshot shows a comparison between the original mix (boosted by +9dB), and the readouts after my applied "edits". This is basically a frequency specific correlation meter. It shows which frequencies drift too far into "negative values" (out of phase), which can result in phase cancellation, or other issues on playback mediums. Although I can't say what would happen on a surround setup - since I don't have one with discrete speaker setup like that on hand.
As you can clearly see, the stereo field is too wide, the focus is "in the back" (phase inverted). So the first tool I used, was to fix the lowend (which made the song more mono compatible, therefore translate better on most playback devices) by "monofying" it up until 95Hz. I also "narrowed" the stereo field from 100% to 85%. With the follow-up edits I did, the stereo width balanced itself out again.
Next plugin in series was a band-limited saturator that focuses on the midrange. This in turn tamed some overall harshness. I then follow up with a Linear Phase EQ, reducing the sub frequencies by about -1,5dB, shaping the vocals a bit at about 5,76kHz (mid only), and reducing some very harsh kick clicks at about 15kHz (sides only). The EQ is also automated at 2,85kHz with a narrow Q to reduce the harshness of the vocals in the mid channel (this filter is only used during the middle section of the song -- the vocals were very nasal).
I'm then running 2 dynamic EQ's in series. The first one only processes the mid signal, focusing on 1,55kHz, boosting about +3dB. However, I'm using a special feature of MDynamicEQ, which is adding harmonics to this particular frequency band. This balances out the +3dB boost (making it less noticeable) and basically "ripples" over the next couple of octaves, reducing the harshness of the mid frequencies. The second dynamic EQ runs in stereo mode and only responds to the overly loud kick drum sub frequencies, reducing it by about -2dB every time it hits. This is adding some motion, while also taking care of the excessive lowend.
Just for experimenting purposes, I then ran the signal through a wide-band compressor. I wanted to know if a gluing effect would add to this production. It didn't, but a side-effect was taming the still aggressive kick clicks. The compressor is only reducing about -1,2dB of gain every time the kick hits.
Sidenote: one thing I noticed with your new mix, is that the initial kick drum of the "low frequency" section hit is not there -- this was not due to me "editing". Maybe this is another point for a possible adjustment, to have a better transition.
Finally, I follow up with a more "broad"/subtle sound shaping EQ, once more in Linear Phase mode. About 90% of the mixes I work on might need a slight dip in the 650Hz range (mid only), which is what I did. I tried to add some "fairy dust" with a boost of 14kHz (mid only) while also attempting to bring out the masked hi-hats (this sadly didn't work in the long run). I then added back some "side" content by subtly boosting with a high-shelf at 6,7kHz. And finally, for experimental purposes, I wanted to know what would happen if I boost the sides around 3,3kHz... usually this introduces a harsh and nasal feeling. In this case however, it brought out the percussive work as contrast to the saturated mid channel.
Here is the full signal chain and the settings I've used.
The final steps in this signal chain were another gain boost of about +8dB, and a limiter array to not only bring the signal to the right loudness, but to also (mostly only) tame rogue peaks. I could have pushed it to -14 LUFS ILk (which is the recommended maximum for the Songwriting Competition), but I opted to stay with -16 LUFS ILk... basically only "clipping away" peaks.
Please note:
This is my interpretation how I would try to make your edit louder and usable on more devices. Pulling this edit into mono (mono compatibility check) does translate better (instruments are not as lost), and running the mix through sound design tools to simulate other playback devices (e.g. AudioThing Speaker - testing: smartphone, earbuts, old TV, boombox) also results in a better listening experience. You can actually hear a more balanced kick drum, have a "hint" of bass (due to the saturation), and the vocals are less harsh.
Should you get back to your mix agian... I would really dial back on the chorusing and phasing effects, maybe even make the vocals less harsh (give them back a touch more low-end frequencies). Bar minimum, fix the mono compatibility (still important to this day). And maybe push the mix into a better loudness area between -18 LUFS ILk and -14 LUFS ILk.
You've created a very interesting song - this is just what I'd do to elevate it.
Also... please remember to check the filename (see filename template), and while a submission in m4a isn't frowned upon... the higher the mix quality, the better the final analysis.
Once more, welcome to the community.