Where am I going wrong?!!!

CutFade

CutFade

New member
So I have been making beats for about 4 years on and off. Now I just recently started recording music for friends and these are the tracks I've done with them so far. I can't seem to get my mixes really remotely enjoyable! i can go over my process if needed, but I was hoping you might be able to listen to a sample and give me some advice on where I am going wrong.




'
 
After listening through a bit, I'd say the biggest issue is the vocals (the beats are dope). The sound quality is poor and distorting/clipping at times (are you recording with the mic/interface turned up too loud?). Try having the rapper double up on certain rhyming phrases to give it extra presence. Once you get good clear takes, compress the vocals fairly heavily. After that, go easy on the limiting (or whatever it is you're doing) on the master track. It's good to have an "in your face" mix but if it's overdone the dynamics end up being squashed and distorted.

In conclusion... make the vocals extra punchy, but keep the mix smooth.

Hope that's helpful and not off target of what you're asking for...
 
Thanks for replying man it means a lot. i do think I have the vocals clipping a bit on the preamp. The master track I have izotope and I'm slamming it pretty hard. The vocals I have compressed as much as I could get them and EQ'd the mids around 500 down and upped the highs around 10K. I'll try and see what I can do about the clipping and add some overdubs for emphasis. I'll definitely tone down the in your face master slam too.
 
Nice, I've been running Izotope as well. What I've been doing whenever I use one of their presets for the master track is to look at the maximizer. I've been turning down each individual track and making sure the that the negative numbers (whenever it compresses a spike in the volume) stay around 2-3 and don't go much over 5 when all tracks are playing. This way it's smoothing out the mix but not overdoing it. The way it processes audio makes it really easy to slam the master too hard on accident. Hope that helps a bit.
 
Get a better microphone. It sounds like you're using a computer mic. Once you have a better mic then just follow the advice given above.
 
Here is my suggestions in order of importance:


1. Track with nothing clipping or even in the yellow area. Make sure the highest peak is still well in green. (Don't worry about volume here).
2. Get Vocal Align (I am going to explain why in a second).
3. Get melodyne or autotune (more to come).
3. Make 6 stem tracks for your left right and center channel from all your recorded tracks. (again I'll explain why in a second)

Now to sum it up all together:

What you will do is you will do six good takes (I suggest more than that honestly but most people are lazy and not professional vocalists) and comp those six vocals (comping is the process of slicing together the best takes.) You should be down from those six tracks to about your favorite 3 tracks. Now you will vocal align those 3 awesome tracks so they perfectly align and are in perfect phase! By doing this you double the amplitude of your vocals effectively everytime you comp them and align them without clipping them. Next slightly tune your vocals with autotune or vocal align with out making them sound like t-pain (if you have a pro singer these should be very small adjustments). Then to add a modern feel to your vocals buss out all your vocals and run vocal align on extreme settings to make them only be in scale, then gradually bring up the levels until they mix properly without adding too much fx. With time you'll figure out how much is too much and how little is not enough.

Finally make sure you pan your vocals left, center, and right so you have very large lush vocals, and mix them so they sit on the mix without overbearing the mix.


I hope this helps.

Lu
 
Last edited:
Ahh man.. okay, this is a situation where your mixing skills are holding you back. In this case, i think severely.

Vocals are distorted, yet also sound crushed.
The tracks sound like you only care about how hard the kicks hit and how crispy the snares are.

Question: What is your monitoring like? What monitors? What interface? What does your room look like?

Quickest fix:
Send the stems to somebody who can mix for you, on a good set of monitors


Slower fix, but you'll learn more:
1) Download a good VU meter. I use Klanghelm. It's fantastic.
2) Set the VU meter to K-14 at the max. I like mixing around K-18. This is to prevent you from mixing your sounds too hot, which it sounds like you're doing.

Your vocals sound very compressed. I don't knwo if you compressed or not.. but relative to the beat, they are distorting.. are not clear.. and are a distraction. When I listen.. I am focused on the lack of quality of the vocals. That should not be the case... you have a good input chain. I should only b accepting that the vocals are there, and focusing on the overall track's sound. So... do this:

3) Start with absolutely no plugins at all, and all faders set to 0. Now.. with the VU meters, make sure no individual sound goes over 0. AT ALL. Keep every single part of your beat under 0 at all costs. Even if it means turning down other things more than you are comfortable with. Don't worry about overall volume right now, you'll fix that later. Just get your levels correct. Your song will appreciate it. Also put a VU meter on the master fader and do not let anything go over 0. If it is, even a little bit... lower everything with trim or gain. Don't touch the DAW fader yet. That's the last thing to do. If your song is too low... just turn your monitors up more.
4) Now... don't add any compressors to your song anywhere. And I mean... AT ALL. You are not ready for that yet. And I don't mean that as a diss. I just mean that to reset your brain so you can attack this mixing thing the correct way.
5) Use only EQ, gain/trim adjustments, and hard panning L and R. With the EQ.. first, think about what's most important to you... it's probably the kick drum and the vocals. These are the two things you don't want other instruments competing with. So... take the EQ, set a high Q, and raise the gain all the way to the top and do a sweep of both the kick drum and the voice, so you can find the sweet spot. Once you find the sweet spot (the spot you just really like about that waveform) note it. Now.. set those EQ back to 0. Then.. add EQs to other parts of the beat and begin subtracting some of those competing frequencies out. This will make your kick and voice some relatively more clean, b/c nothing is competing with it anymore. EQ is your best friend.
6) Solo out the voice... and pay attention to any small parts of it that may be way too loud. You can usually eyeball this. Lower those parts by chopping it up and physically lowering the waveform. THIS is your compression lol. Be happy with it for now. Understand that the fact you are not using a compressor is not what will hold you back at all.
7) Pan the less-important sounds out of the way of your vocals. Things like strings, horns, hi-hats... move those away from the voice. Try to keep it even. This creates space.
8) Reverb is not necessary.. but it helps. Add a reverb to a bus... and then slowly add just a touch of reverb to a few things like voice, snare, etc. Use the same reverb instance on everything. One instance only... for now. Don't add multiple reverb instances. You'll mess up :)


At this point, you should notice you have a much cleaner song so far. That's because you are getting rid of problems. Your mindset should be "I just recorded a few things... now I need to figure out what's wrong with what I just recorded. Once I get rid of the things I don't like.. it will leave the things I do like, and it will finally sound good". Clean is best. When you get good.. you can think about dirtying things up. Now.. time to give you your loudness back.

9) Make sure your master output's VU meter is still not going over 0. This is important.
10) Add a basic compressor. Get some light compression going. Medium attack, medium release, maybe 3-4 db's of compression at the most. You're closing up the dynamic range.. ie> making loud things quieter. You can give it a little gain. It's probably okay to go over 0 on the VU meter now, since your song is now under control
11) Add a limiter and pull down the threshold until you like what you hear.
12) Spot check with popular songs at this point. You'll probably find a few things you'd like to get a do over on. But... don't expect to sound anywhere as good as that popular song until you get good. That level of mixing really requires a lot of experience. Just try to get your loudness in the ballpark for now.


Hope that helps, at least a little bit. I don't consider myself a pro yet... i'd say i'm semi-pro with a good upside. I compiled informatio non mixing for a long time from guys on this site, so keep your eyes open and your mind clear. Understand that you don't know a lot about mixing and accept it... that helps you become a student. Being a student is the most important thing i've done so far. Once I started diong what I told you above... my mixes started getting better and better. Now I feel I can be proud enough to hand someone music and not be embarrased about my product. You will get there... you can only get better, can't get worse.

---------- Post added at 10:41 AM ---------- Previous post was at 10:33 AM ----------

I haven't used Vocal Align.... but that doesn't sound like a good idea here IMHO.

One good vocal waveform is more than enough to work with. You don't need to sound like Eminem with stacked layers of vocals. Those days are over in HipHop. Just one good clean vocal is great.. well, that's my own personal preference.

Also.. the panned L and panned R idea is ehhhh... unnecessary for lead vocals. Lead vocals sound great when they are clean and down the middle. Spread out vocals can sound amateur IMO. Again... that could just be my taste. Theoretically.. if they were the exact same vocal panned L and panned R, then they actually cancel out and are just down the middle. And doubling up the same vocal just to get some more gain is a really bad technique for somebody knew to try to do. There just isn't any major benefit there and it leaves a lot of room for error, where somebody experienced wouldn't even know how to fix it.

Keep it as pure as possible... the result will be a clean sounding song.

---------- Post added at 10:49 AM ---------- Previous post was at 10:41 AM ----------

Lastly... I mix all of my songs with the audio waveforms. I don't mix straight from the VST instrument anymore.. i prefer to bounce it to audio and then mix. It just helps me work. it's easier to see what's going on... and audio is easier to chop and manipulate than midi is.

Not sure if you're making your own beats also but... if so, that's may help.

And also.. if you're recording over someone else's beat... PLEASE make sure they are giving you the stems of the beat, so you can adjust each element of the beat one by one. If all you have is one file render for the entire beat..... nothing I said above is even possible, and your mix will suffer. You won't be able to do the things you need to do to fix it properly.
 
Thanks for the replies guys. I'm going to read over this a few times once I'm home from work and I'll reply back then.
 
A lot of good info in here. Just to add my two cents:

1. As many of stated, never let anything clip. Mix-DOWN is called mix-down for a reason. Keep it in the green
2. I like to apply a stage of EQ before Compressing sounds, then apply a second 'final' EQ.
-What I do is put up 2 individual Waves Q1 Mono fx on EACH track(Normally, if you hold Ctrl and drag the FX to another FX bin, it will copy the FX and it's settings. Saves time)
-On the first Q, solo the instrument and set the filter to High-Pass. Most sounds don't have much presence below 100, so just move the node until the sound is affected. Roll back to just where the instrument/vocal sounds the same without the Filter. Do the inverse for bass lines and kicks. Low-Pass. Start all the way up at 16kh and move the node left until the sound is affected. Scale back just a bit.
-With the second Q, cut about 3db from anywhere between 400 - 800. This frequency range can sound quite boxy, nothing too important occupies this area. I usually go with a 3 - 3.5db cut at 600, with the q rounded to 400 and 800 respectively.
This just generally cleans up the track. Kinda opens it up a bit and allows it to breathe before most the work is even started.

From there I'll adjust levels of everything, pan things, make tweaks in the arrangement if need be, etc.

Now compressions and setting up buses. I'm not too big on the compression for loudness, but more for controlling specific frequencies. Some bass frequencies are quite louder then others, compressors are great for keeping them all in line. Or to make a snare crack, compressors are great.....but Vocals, I would splice individual loud parts and adjust their volume down before I put a compression on it. Usually something simple as the Boost plugin from Cakewalk with a very subtle setting is enough to sit your vocal correctly. Too much makes the vocal about as enjoyable to listen to as a baby crying. Keep it natural in my opinion.

I'll have more later.....I'm about at my mental limit for the time being. Hope what's there helps though!

-D
 
You must log in or register to reply here.
Back
Top