Learn about mixing with pink noise to get some balance. You can even shape your pink noise sample to the spectrum of one of your favourite tracks and use that as a reference.   

Some reading about this here:
http://www.soundonsound.com/sos/dec14/articles/pink-noise.htm

And a video with Tom Demac explaining how to use it:
https://www.sonicacademy.com/courses/tom-demac-volume-1

About the delays and reverbs, mostly using them when producing for creative effects or stereo widening. So doesn't really matter as long as it sounds good    

I just move the faders until it sounds right based on the current focus of the track (basses are loud when they're important, just like vocals). If you ask me why any of the levels are at any of their values, the only response I can give is that "it sounds good." I've never really followed a textbook or looked at an analyzer (hold Vengeance's Stereofield analyzer, but even then I just like how it looks). I solo out the four frequency bands individually in Ozone to check how they sound, drop Utility on the master to check mono, and low/high pass just enough to cut away the frequencies I don't need without audibly damaging the sound.

Also, I've started doing general effects (basically reverbs and delays) on my sends. Two or three reverbs mix up a lot less mud than like eight (which is what I used to do). It's also easier on my cpu. And the only things I ever use a compressor for are really just changing drum transients and sidechaining. Compressors are really good at making organic sounds more consistent in regards to overall volume level. But if you're like me and 90% of your sounds are programmed in and don't really exceed the level you set them at, it's kind of redundant.

There are no "correct" ways to do anything. Just ways that work. You can crochet a sweater, or you can knit a sweater. They're going to do the same thing. What matters is that you like the end result.

I bring all my faders down to -INF. I really recommend mixing your kick & bass first. They are the backbone of your entire track; getting those two things squared away first is really important.

After that, I mix in order of next importance. Typically I work my way up the "frequency chain". If an element contains lower frequencies than another element, say a piano vs. a high hat, than I'll mix the piano first.

Good luck bro!

I bring all my faders down to -INF. I really recommend mixing your kick & bass first. They are the backbone of your entire track; getting those two things squared away first is really important. After that, I mix in order of next importance.

I work really similarly. I often add my EQ, effects, and pan tracks during the arrangement/composition process and adjust levels appropriately either in the VST or with a utility. When the time comes for me to mix down I usually just double check all of my track have an EQ, there's sidechaining where appropriate, etc. Then I'll take everything to -Inf and typically pull up the drums and bass first, as mentioned. The only time something else will take precedence is if it's a very vocal-heavy song, in which case I will bring that up first. Then I'll mix the other elements by importance and intended prominence in the track. I'll also mix in order from the loudest section to the quietest/most sparse (i.e. drop -> intro).

Quote from: Lydian on May 10, 2016, 02:55:47 pm

I forgot. Were you the one that goes to school for audio engineering?

Yeah! The workflow I use was mentioned by three separate teachers I've had so far, though I haven't had any dedicated mixing classes yet. I like ErikF's idea of working up the frequency spectrum, since the highest frequency sounds are going to be the ones that require the least amount of level to be heard.

I'm also a huge advocate of working in cycles of analysis and action using "fix sheets" - a physical piece of paper that you keep with you when you're working (so it applies to every step of the process): Before doing anything, listen to the project start to finish. Write down every single thing that you think you need to fix, remove, add, or otherwise change. Once you've gotten to the end of the song, start fixing problems on the list top to bottom. If you notice new errors, add 'em to the bottom of the list. Start broad, and work your way in. By the time that list is including things like "change the decay time on the reverb by .3 seconds" you've reached the point of diminishing returns and should call it complete.

I've always liked to start from the lowest frequency spectrum as well. The only issue I'm beginning to see with it is that I've seen people recommend to start with the most important element of the mix which actually isn't always kick and bass.

I've used fix sheets in the last track I finished. Found them immensely helpful. Makes things a lot faster. I may try to incorporate that new workflow though. I really like the idea of mixing the way your teachers describe. The only reason I wouldn't build the mix from bottom to top though is because I usually work in groups so I like to build out the entire drum bus first before bringing in bass and leads.

Then again I have yet to be satisfied with my mixes so It may be time to try something new. Since I mix in groups I've always tried to get them sounding good in solo before mixing them with the other groups. This may be limiting me since it's generally advised to not mix in solo.