Issues With Mastering

[BrandonS]

I know this is a great place to ask about Mastering.

So I've been lounging around here over the years. I learned about EQ, which helped open up my mixes a bit. I've figured out how to scan the wave to discover the frequency used by the sound, and carve out a space for the vocals... I stopped using EQ afterwords on the finished song wave to add treble or whatnot, and rather just focus on the song sounding natural before-hand without changing the output too much.

So in short, I am getting the perfect mixes. Unfortunately, the mastering I am attempting on these doesn't do them justice. I try to get them loud enough, and they will crackle, or distort. I know people will say "you don't need them to be that loud" but I would like them to be normal, at least. I should be able to make the volume consistent with a recording from the 90s, at least, as far as volume goes. If I open up an old Cryptopsy and play it at the same time as my stuff, the Cryptopsy will overpower my stuff with its volume.

What I've basically tried so far is hard limiting. I'll set the maximum at 0db, and then boost the 4 or 5 safe decibals to get it up to that point. I can't go TOO far, because then it will crackle or distort. That limiting basically gets rid of the high peaks. I know I've asked about similar things before, but I've never really got this far with such a good mix before.

What exactly do I need to do to get the volume without crackles and such? Is it compression? I'll attach pictures of what my options are in Adobe Audition 1.5, and hopefully you guys can help me out... I don't know anything about the options in Dynamic Processing, and the lookahead time and such of hard limiting doesn't make sense to me either.

(Here is the Hard Limiting window)

(This is the graphic mode of Dynamics Processing)

(The Traditional mode of Dynamics Processing)



Sorry about the drained appearance of the color. All I've got is MS Paint... Screen captured, and saved as a jpg.

Yet again, I appreciate all your help!

 

[Torniojaws]

To get the volume up, you cut off the peaks and compress already in the original tracks (each if needed). And for a quick and dirty way, just setup several compressors in the master chain and then only use the limiter as a brickwall with level at -0.2 dB or so and threshold at 0.0 dB (ie. no boosting). What you do is boost the volume before the limiter and only keep the limiter there to keep the mix from clipping.

 

[Deaf Ear]

Getting volume out of a mix usually takes a lot of careful attention to detail in the HP filtering on each track. I like to get rid of anything that's not doing something for me, then start zoning in on woofy areas. Then there's some drum limiting BEFORE the 2bus. Then some limiting after a compressor on the way out. That's the very short of it all. Do you have a clip to post of this song? I'm sure that would help more than anything.

 

[BrandonS]

I don't really have a recent clip you could listen to, but there are a few mixes up on the MySpace page... I usually try to get a decent compression on things that seem to be peaked, like bass performances or vocals, but my guitar rhythms usually record at a consistent volume throughout (due to a Line6 Toneport with compression), and the drums are consistent as well.... There isn't really a lot of peaks sticking around, but should I try to compress that more? And the keyboards too? They're not as consistent because it records depending on how hard you hit the keys.

So you mean to put on the limiter basically to limit it while I use a strong compression to boost it up? I had been using the hard limiter to boost it by about 6db, and then limit it to -1db to cut off some peaks, then normalize it back to 100% volume - 0db. But that resulted in some crackles. I'll try to see what I can do with that stuff.

No idea how to HP filter - that's macroeconomics, but furthermore, math - something I was never good at. I didn't even know this specific thing existed in music until I read your post. And I'm not sure what you mean by 2bus, it must be a technique that I haven't used before... If you guys need me to post a li'l somethin-somethin that you can import into your audio programs to inspect it, let me know and I'll try to get up a 192kbps MP3, but the reason I don't have any at the moment is because I'm on dialup. Thanks!

 

[ptgeek911]

HP (high pass filtering) is basically taking out all the LOW frequencies that are making an instrument "woofy". A guitar track should be high passed anywhere from 80hz to 100hz. Those frequencies are used for kick and bass. Guitars should also be low passed (taking out all the high frequencies that are make a lot of noise usually from 12k on up). That is just an example of filtering.


Anyways.

When I use a limiter, I always have it at 0.9 or 1db. NEVER 0. Then I take the threshold down until it sounds loud, but not distorted or hard limited. I also use a compressor plug BEFORE the limiter. Usually at a 2:1 ratio, letting it reduce anywhere from 2-4db depending on the songs. Sometimes I will use a multiband compressor or the LinEQ to control the overall frequencies.

This my opinion, and some might think it's a weird approach, but I'm happy with my mixes.

 

[ptgeek911]

Oh, also. There is a thread around here about using parallel compression for drums. It helps tremendously. Basically you bus all your drums to an Aux track and compress it there as well on each individual track. And then also have and Aux track with the same input without a compressor. So basically you will have drums that are clean, and ones that are smashed to shit with a compressor. Then you blend to taste.

 

[BrandonS]

Sounds cool, but surely you do raise the volume back to 0db at some point? All the songs I've imported show up as a flat line.. Except for the new Therion, which had a bunch of dips in it... and that CD is a bit quieter than my other CDs... I bet Sneap didn't master it.

With my guitars, I had basically been dipping an area for the vocals wherever it made the vox more present, and I had left the high and low areas flat. What Q would you recommend for using to pass the frequencies you listed? For the Vox pass, I had been using a Q of 1 with just a little dip.

I'm going to print out all these tips and see how it goes. The ratios didn't make much sense to me when I was trying that yesterday, it seemed like I could put anything in there and it wouldn't make any difference... So after filtering, I should basically tell it to compress between 2 and 4 db to try it, and then boost it to about 0.9db you said? And then take chiny chunks off in limiting to make it sound loud but not distorted or hard limited, aight... I will let you know how it goes when I return. I'm almost late for Music Theory (which teaches nothing of recording methods) so it'll be a while.

 

[Torniojaws]

The ratio doesn't do anything if you don't set the threshold. Threshold sets the minimum level before the compressor "turns on", and ratio sets how much it compresses; for example, if you set a 2:1 ratio, it takes 2 dB increase above the threshold level to increase the actual level by 1 dB. On a 4:1 ratio, it takes the original signal a 4 dB increase above threshold to actually increase by 1 dB. Attack is the delay before the compressor "answers the phone" after it starts ringing.

 

[Deaf Ear]

I listened to some of the mixes up on the myspace page, and I will say that High pass filtering will help you a TON. There's a lot of low end masking and rumble going on. Do this just for practice, and I promise it will help.

Find an eq plugin in your arsenal that has a HP filter on it. You may have to click on the curve to make it do this. It will make the graphic interface look like there's a ski slope on the left side. Look at your instruction manuals if you don't get how to make an EQ do a HP.

Now, strap it to every single channel you have open, and I do mean everything. On each instrument, slide the filter as high as it can go without ruining what's important to the sound. I usually find that this is somewhere between 50 and 100 for Gtrs, about 100 for vocals, around 200 for the snare (depending on the drum) about 120 for rack toms, 50 or so for the floor tom. The kick drum I would roll off below it's fundamental. This is usually between 50 and 70. I like to leave the bass wide open, or have a gentle filter set at about 30 to just tuck it away a little.

Take a few hours and do this with a mix just to practice what HP filtering will do for you. All that low end energy ( most of which you can't even really perceive ) is eating up your headroom and will make your mixes very quiet indeed. This is step one, when you get that figured out and have a good understanding on how it works, and what it does to your mix, we'll move on to compression. Having all that low end in there is going to make your compressor pump and act funny all day long. You'll never get good compression results til you cut the hell out of that low end.

May the force be with you, and let me know how it goes.

 

[BrandonS]

Wow, thanks a ton... That will definitely help. I had no idea my problems were being caused by a lack of filtering.

I found Scientific Filters in Audition, and it's got choices for high pass, low pass, band pass, and band stop... I've got it selected to high pass and I set it at 600Hz on my "Impaled northern moonforest" program test mp3, and it works really good! Uhmm but it works the opposite of what I thought it would. High pass makes it only high, low pass makes it only low and removes the highs. I thought passing meant that high pass took the highs off, and vice versa.. No matter, it won't be hard to remember. I'll see how this goes, and I'll be sure not to remove TOO much passings.

 

[Deaf Ear]

Passing an entire mix at 600Hz is NOT the way to go, haha, but you're on the right track. Yes, they are named accordingly, a High pass filter allows the highs to "pass" through, and a low pass allows the lows to "pass" through. The point of filtering is to just edge out the frequencies that aren't important to the instrument you're filtering. If you're doing it correctly, it won't be that obvious on each instrument, but the mix will clear up dramatically.

 

[Skaldir]

So in short, I am getting the perfect mixes. Unfortunately, the mastering I am attempting on these doesn't do them justice.

Why don't you upload a mix for us to listen to? I always wanted to hear "the perfect mix"

I usually do a lowcut at 40 HZ or something.
I'd like to know if you guys ever felt the need to do a highcut in mastering though. I don't know.... there is a very little amount of mixes I highcut at 19 or 20 khz. But in most cases I stay away from it.

 

[BrandonS]

I would post the perfect mix, except now my mixes need re-done because I noticed after applying these passing filters to each track, the song needs re-mixed.... It took a lot of the bass out of the guitars and keyboards, which is nice - it made the mix a bit thin, so I've got to raise the bass track, and lower the others a tad and get them all in shape again as a coherent mix... But perfection is in the eye of the beholder; I thought they sounded great (until I went to master) but you never know what a different person would think of it. In short: I'm biased.

I cut 100Hz and 12k on the guitar tracks, and I really couldn't hear that much of a difference on their own but when I played it back after applying all the passings, it sounded much more open... I also cut some bass on a cello, and keyboard strings.. that really opened up that section a good bit and it sounds better without being so overpowering in the low freq.

Unicorn, you're saying you 40Hz cut during the mastering stage to take some bass off of the songs? Sounds interesting...

 

[Zombietakeover]

I promise you, you dont have a perfect mix.....If that was the case this forum would be named after you......plus if u dont even know what a high pass filter is ...I cant see you having a perfect mix

I would post the perfect mix, except now my mixes need re-done because I noticed after applying these passing filters to each track, the song needs re-mixed.... It took a lot of the bass out of the guitars and keyboards, which is nice - it made the mix a bit thin, so I've got to raise the bass track, and lower the others a tad and get them all in shape again as a coherent mix... But perfection is in the eye of the beholder; I thought they sounded great (until I went to master) but you never know what a different person would think of it. In short: I'm biased.

I cut 100Hz and 12k on the guitar tracks, and I really couldn't hear that much of a difference on their own but when I played it back after applying all the passings, it sounded much more open... I also cut some bass on a cello, and keyboard strings.. that really opened up that section a good bit and it sounds better without being so overpowering in the low freq.

Unicorn, you're saying you 40Hz cut during the mastering stage to take some bass off of the songs? Sounds interesting...

 

[BrandonS]

The whole point of my spiel was to say that the mix was perfect to me, it sounded better than I thought it could possibly sound when I was done...

Now I'm in more pickles..

I did the passing, and remixed a song yet when I try to raise the volume, it crackles a lot. Here is the song before compression or anything else in 192kbps format... if you can figure out what it needs, please help me... I've been pulling all my hair out trying to figure out how to master these friggin' songs.

The guitars have a 100Hz - 12k pass...
The bass has I think a 90Hz low pass. And I turned it down a lot.
The pass for the keys and drums I'm not so sure about now, but they sounded right... Except the BD which was odd.

The part from 2:17 to 2:20 is an obvious give-away whenever I am checking for crackles, it is always stuffed with them for some reason after I've mastered. Stuffed!

Would the parametric equalizer on each track be interfering with the mix? Do I need to remove or redo them? Do I have to remove the reverb? nooo....

 

[Jaymz]

100hz - 12k high pass? Jesus I think I have mine at 48hz otherwise its as thin as fuck lol.

Whats the crack with those vocals? lol

 

[Zombietakeover]

lol.....yeah those vocals are something........to be honest dude if i was at your level which im not much higher im pretty new to this also......I wouldnt be worrying about mastering Right now....There are plenty of other issues i think you need to address......
I dont think mastering is going to save your ass as far as raising the volumes........i know mastering helps raise the volume at the end......but i think your volume issues lie in the tracking stage....either your recording stuff at low volumes or your using in proper gain staging......

 

[BrandonS]

Well I try to normalize the stuff I record to 30%, is that the problem? Pretty much everything gets normalized to 30% so it can be pretty consistent as far as the volume goes.

 

[Jaymz]

Ive never used normalizing.....mainly because I dont exactly get what it does....please englighten me!

 

[BrandonS]

Normalizing basically changes the volume. I use it in percent mode, which means it changes the volume to 30% of 100% of the possible volume, 100% being 0db. So 30% is a good bit under 0db I'd say.