Need Advice on Mixing Punk Rock/Hard Rock Tracks

anaesthesiac

New member
I've currently recorded a set of 6 tracks for a demo release but i'm having HUGE dramas taming them in the mix department. The more i play with the EQ/Compression the more the sounds seem to be fighting each other.

I really only want a tight drum and bass backbone [that won't blow out on any stereo] with the guitar and vox to sit over the top - nothing fancy.

I'm using Cubase 5 with T-Racks 3 EQ & Compression and a Lexicon Reverb, that's all. I was using Waves & Sonnox Oxford but found them even more confusing to use.

There are two examples of the same song at the link listed at the bottom of this message, the initial mix and the current mix [mix 3]. Any help/advice/suggestions would be HUGELY appreciated.

Cheers.

http://forum.recordingreview.com/f76/having-major-trouble-taming-set-songs-31334/#post205808
 
Apologies, i didn't realise you needed to be a member to get them. I've uploaded to my own server.

Mix 1:

Mix 3:


This should work perfectly. Keep in mind the music style is meant to sound rough and raw, but i've got overblown and ugly instead :spank:
 
My impression is that there is way too much happening in the 1-4Khz area which is where human hearing is most sensitive, and as a result your tune is hard to listen to for any length of time. I couldn't hear any bass at all. The guitars and vocals are fighting dirtier than Mike Tyson in there. If you're boosting with EQ, please stop. Instead try and cut the guitars away from where the vocals are to give them room. Then look to bulk up your low end a little. If you overcompress the drums you'll rob them of power. Also don't apply effects to individual tracks in isolation - mix the tune as a whole
 
Mix 3 is a little better than mix 1. I think you need to make a broad 1.5k boost on the vox, and pair it with a narrow-ish 1.5k cut in the guitars.

Is there only one rhythm guitar? No doubling? Maybe try panning it further to the side. It sounds a little too center panned. Try cutting a little around 500 and 100 - 125 to make some room for the tone and the rumble of the bass, respectively.

Back to the vox, give them a broad boost around 6k, and then a high shelf starting at 12k. That should give them a bit more presence.
 
Thanks a lot for taking the time to reply, i really appreciate it.

GREG_L
: All links should work fine mate, i've tried them and had others test them with no problems thus far.

FadeToMuffins: There are two stereo rhythm guitars panned 40% left & 40% right, i also have one fill guitar with a phase/chorus 10% left and one palm muted rhythm 10% right - 4 stereo tracks of guitar. The bass is mono recorded straight from a SansAmp Bass Driver DI and is centered. I'm going to sound like an idiot here but what do you mean by "broad" and "narrow" boosts/cuts? And same again for "high shelf"? I apologise but i've only ever mixed by ear and know sweet f**k all as far as these terms and there logistics are concerned.

Bulls Hit: Nice nickname. Everything is EQ'd seperately then the entire thing is mixed as a whole. The bass on the first version is hard to hear but in the mix 3 version it's fairly predominant on all of my sound sources [KRK Rokit 8 near fields, Sennheiser HD280 phones, Home Stereo, Car Stereo]. The guitars themselves aren't really EQ'd a lot, just a bit of removal regarding higher frequencies. It seems the only way i can get the drums to stand out and have that "punch" is compression, what am i doing wrong? If i don't compress them they get absorbed by the song and fail to have the impact i want. Basically i've got the drums and bass compressed 1.10:1 [i think] with the input -2db and the output at around +5db [again, i think] using the Opto Compressor in T-Racks 3.

I appreciate your suggestions, i'll start messing with it tomorrow before work and try and implement some of the suggestions [if i understand them]

cheers.
 
Oy!

It's like the tracks and mix are so rolled-off, EQ carved and messed with, that I can't even correct the spectrum imbalance...it's like there's no bass signal to boost, almost. There's hope for the work, but it's gonna take some individual track EQ fixing, and some frequency spectrum referencing and profiling to get you where you would want it to be.
 
I liked mix 3 better than mix 1. Mix 1 has almost no low end. I can hear a bass on 3.

I guess I like the vocal effects. For the style I guess they work.

I think the biggest thing holding the mix back is the guitar. Someone mentioned lots of 1-4K energy. I'd agree and say that the problem lies more in the upper end of that range. The guitar is very thin and crispy. It needs to be fat and growly.

To me, I'd work on tracking better guitar sounds.

Performances were good. Good energy to them.
 
Greg_L: I'll upload the tracks to a different server for you, not sure why you can't get to them.

jeffmaher: The individual tracks are pretty messed with, i kept getting a bottom end blowout on the home & car stereo in playback so basically in the main mix EQ i rolled the end down. Now to add some more quotes to the idiot pile, how do i go about this "frequency spectrum refrencing"? And what am i looking for...

TripleM: Thanks, the guitars actually have a bit of growl to them in their initial recorded state, but i found it hard to get it to sit well with the bass and drums so i rolled off ALL of the body and kept only the high-mids and highs.

I'm just about to open Cubase and start messing with some of your suggestions, i think i'm going to strip each Insert channel and just start again. I'll post up a new version of it before i go to work [if it works!].

Thanks again to all.
 
I have played with it for a couple hours, i think i'm getting somewhere but i feel i've got a lot more to work out. On my monitors it sounds a bit too mid-heavy, it's odd but in cubase it sounds 100% different than when i mix it down. The clarity in Cubase is great, down to mp3 it sucks all the life out of it and sounds different.

I panned the two main rhythm guitars 100% left & right, cut a section of the guitar highs to give the vocals some room, took away the EQ on the drums but kept the compression, with the bass i used a tiny bit of EQ to roll some frequencies off in the highs and used compression.

Should i be user a compressor on the Main Mix channel? I've done that and i'm not sure whether it works in my favour or to my detriment. If i shouldn't be using one, what should i be looking to add to the main output mixer channel?

here's the current mix: mix 4

url:


Thanks again.
 
Last edited:
It sounds a lot better. Panning the guitars wider really opened up the mix. I think the snare needs to be a little louder now. The vocals still need that high-shelf in order to stand out. A High shelf is literally what it's name implies. You pick a frequency (let's say for this song, 12k) and everything above it gets boosted. For this song I would give a 12K hi-shelf of about 3 - 5 db's. I would also give a nice and wide boost at 6k. By wide, I mean that the range of frequencies boosted is very wide. Most EQ's have a "Q" setting or something similar that decides the width of the particular boost/cut you are going to make. I'm not familiar with the plugins you are using, since I use waves plugins, but perhaps if you show me which EQ software you are using I could help you out.

Also, the drums would benefit from some parallel bus compression. Load up all your drums (minus the overheads and kick) and send them to one track. Put a compressor on it with a fast attack (10 ms) and slow release (75 - 100ms) and a fairly steep ratio (8:1) What this will do is boost the transients and body of your drums without boosting the attack, effectively allowing the drums to be both louder and fuller without upsetting the balance of your mix.
 
FadeToMuffins: Thanks mate, all this advice is very helpful. I'm running late for work at the moment but i will read through again when i get home and follow your suggestions. I use the inbuilt EQ with Cubase 5 [in the Inserts area, not the EQ panel] - i think it's called GE-10, but i do have Waves and T-Racks, when i opened them up though they kinda confused me, i'm used to Bar style graphic EQ's and i had no idea what to do with all those knobs.

I'll get back to this thread in a few hours, and thanks again.
 
Referencing:

Find some pro-recorded material and play it in Windows Media Player with the bar graph visual showing....it's a spectrum...low to high frequency, in the general range of human hearing. [or you can load the stereo file onto your DAW, and play the track through a parametric EQ or other display that will show you a more detailed graphic picture of the frequency spectrum].

Look at the picture as the work plays. Some para eq functions let you freeze the line a dozen or so times, to give an 'average' line. That's what a good recording should generally 'look' like. And the ear hears it as 'normal' generally.

The rule of thumb is that above 500Hz, the average frequency spectrum should decline to between -6db to -9db as it approaches 20kHz. The former is what is known as the '3db rolloff'; the latter, the '6db rolloff'. The former is an indicator of a 'hot' recording...the latter, an industry standard.

Below 500Hz, the transients [momentary energy bursts] should top off ...not average....along the same line, extended leftward above the 0db line...as the average of bass range is difficult to average, where drums are included. Just an easier way to analyze the visual.

Your mastered recordings should look the same way, in large part, to sound 'correct'. A pre-mastered mixdown should run parallel to the average line -3 to -6db down.

If you have some organic experience recording, you'll find that the freq specs of your mixed tracks will find their spec in that ballpark....and you have not much work to do to fix the soup.


Now, your last mix:

On Windows Media, I can see the bar graph as it plays. Right off the bat, I see weak, muddy bottom end: the graph is way low, and their's not much in the way of transient spikes when the bass drum or snare hits. No punch in the kit.

The midrange is way low....and that's where the power of a recording is. It's usually a problem having too much of it there, but on your piece, you've maybe over-compensated, and carved out the energy from the thing. The 4K range is hot: that buzzing thing. Then the 'air' frequencies roll off drastically...and you need those to let the HF componants of instruments complete the sonic realism of their broadcast sound.

So I started twiddling the EQ sliders of your record to visually normalize the freq spec on WMP:

31Hz +9
62Hz +4
125Hz +6
250Hz +11
500Hz +10
1K +8
2K -6
4K -6
8K +1
16K +9

That puts the freq spec visually in the ballbark of most pro records. It sounds a lot better, too. But it also throws the sounds of individual instruments out of whack because the fixes are so radically large.

But with this rough norming, you can start to hear things like the swimming ambiance on the lower drum kit filling all the space in the bottom end, etc. It lets you hear things that are killing the mix.

The way to begin fixing the recording is looking at the freq specs of each individual track, and EQ carving so as to assign roughly indivualized chunks of range ..that sound good. Mix them, and tweak to place the overall average freq spec in that reference area. You know what instruments will give you a bump here or there. Experience is the only way to get the knack in the bones. [1/2 rule: EQ carve a track to taste...then apply HALF the medicine. Overcompensation is a real hassle, early in mixing trials. Still a headache for me.]

And I think part of the problem you're having is that you're completing the mix including the vocal. Maybe?? Try mixing the band first, if you are. Get that right, then add the vox. Vox generally sound heavily in the 300 to 700Hz mids. There's a nice notch in your 'band mix' where a vocxal could feature itself. But your vox is buzzing in higher ranges. A good plan might be to kill off some of the 4K bump in the vox, and use that space lower to sit it in??

So try messing with the WMP EQ sliders. A good tool for roughing out EQ plans.

Hope this helps.

And , yeah...some EQ functions really suck. The best I've used is Ozone Izatope's Para EQ. Very organic. Some of my Cakewalk plugins and other EQ tools are just whack....never use them...confusing...unfriendly.

ps..I know some might feel that using a freq spec average to assay a work is not 'artistic'. But considering it, then deviating to produce hotter, colder, pumpinger, or anything new and 'artistic' in the sound, is a lot like figuring out where you are before you glide to where you want to be. Flying blind is not necessarily productive with subjective, tired ears.
 
Last edited:
Mix 4 is an improvement. Guitars sound better, but I still think you'd be better off recording better source sounds and tweaking these. But one benefit of doing it the way you're doing it is that you're learning what an EQ does and you're training your ear. So there's certainly some benefit.

I agree with the comment about the vox lacking some high end. They sound distant. Don't know if that was a tracking issue or a mix issue.

The drums are a bit weak. This genre needs really kicking drums. The snare sounds really slappy and weak.

Still a ways to go, but I noticed real improvement between mixes 1 and 4.
 
jeffmaher: what can i say... wow, thanks. That's a massive amount of information, i'm going to have to read it thoroughly and go through it bit by bit. I don't care about the artistic process, as such, in mixing, i'm primarily a musician [a goofy one, but one nonetheless], i play all this crap myself and just want it to sound listenable when it's all together. I appreciate the info and will get back to you how i go with it.

TripleM: i agree it's beneficial, but it's seriously draining, nevertheless i feel i'm getting a better grasp [slowly]. I've decided to re-record the guitar, bass and drums. It's going to be a MASSIVE pain in the arse but i've laid down a guide track today with new drums, new bass and two new rhythm guitars [an EH Big Muff on the left, a Boss Metal Zone on the right: all run through a Fender Frontman 212R].

FadeToMuffins: I'll be trying your parallel bus compression as well, if only i didn't have to work!

The LATEST mix is here: MIX5

Vocals need redoing, as there seems to be some weird time lapse later on in the song. But i'll spend this week re-recording everything. I think it sounds better, but nowhere near what i want it to be.

Thanks all for your advice so far...
 
Last edited:
Referencing:

Find some pro-recorded material and play it in Windows Media Player with the bar graph visual showing....it's a spectrum...low to high frequency, in the general range of human hearing. [or you can load the stereo file onto your DAW, and play the track through a parametric EQ or other display that will show you a more detailed graphic picture of the frequency spectrum].

Look at the picture as the work plays. Some para eq functions let you freeze the line a dozen or so times, to give an 'average' line. That's what a good recording should generally 'look' like. And the ear hears it as 'normal' generally.

The rule of thumb is that above 500Hz, the average frequency spectrum should decline to between -6db to -9db as it approaches 20kHz. The former is what is known as the '3db rolloff'; the latter, the '6db rolloff'. The former is an indicator of a 'hot' recording...the latter, an industry standard.

Below 500Hz, the transients [momentary energy bursts] should top off ...not average....along the same line, extended leftward above the 0db line...as the average of bass range is difficult to average, where drums are included. Just an easier way to analyze the visual.

Your mastered recordings should look the same way, in large part, to sound 'correct'. A pre-mastered mixdown should run parallel to the average line -3 to -6db down.

If you have some organic experience recording, you'll find that the freq specs of your mixed tracks will find their spec in that ballpark....and you have not much work to do to fix the soup.


Now, your last mix:

On Windows Media, I can see the bar graph as it plays. Right off the bat, I see weak, muddy bottom end: the graph is way low, and their's not much in the way of transient spikes when the bass drum or snare hits. No punch in the kit.

The midrange is way low....and that's where the power of a recording is. It's usually a problem having too much of it there, but on your piece, you've maybe over-compensated, and carved out the energy from the thing. The 4K range is hot: that buzzing thing. Then the 'air' frequencies roll off drastically...and you need those to let the HF componants of instruments complete the sonic realism of their broadcast sound.

So I started twiddling the EQ sliders of your record to visually normalize the freq spec on WMP:

31Hz +9
62Hz +4
125Hz +6
250Hz +11
500Hz +10
1K +8
2K -6
4K -6
8K +1
16K +9

That puts the freq spec visually in the ballbark of most pro records. It sounds a lot better, too. But it also throws the sounds of individual instruments out of whack because the fixes are so radically large.

But with this rough norming, you can start to hear things like the swimming ambiance on the lower drum kit filling all the space in the bottom end, etc. It lets you hear things that are killing the mix.

The way to begin fixing the recording is looking at the freq specs of each individual track, and EQ carving so as to assign roughly indivualized chunks of range ..that sound good. Mix them, and tweak to place the overall average freq spec in that reference area. You know what instruments will give you a bump here or there. Experience is the only way to get the knack in the bones. [1/2 rule: EQ carve a track to taste...then apply HALF the medicine. Overcompensation is a real hassle, early in mixing trials. Still a headache for me.]

And I think part of the problem you're having is that you're completing the mix including the vocal. Maybe?? Try mixing the band first, if you are. Get that right, then add the vox. Vox generally sound heavily in the 300 to 700Hz mids. There's a nice notch in your 'band mix' where a vocxal could feature itself. But your vox is buzzing in higher ranges. A good plan might be to kill off some of the 4K bump in the vox, and use that space lower to sit it in??

So try messing with the WMP EQ sliders. A good tool for roughing out EQ plans.

Hope this helps.

And , yeah...some EQ functions really suck. The best I've used is Ozone Izatope's Para EQ. Very organic. Some of my Cakewalk plugins and other EQ tools are just whack....never use them...confusing...unfriendly.

ps..I know some might feel that using a freq spec average to assay a work is not 'artistic'. But considering it, then deviating to produce hotter, colder, pumpinger, or anything new and 'artistic' in the sound, is a lot like figuring out where you are before you glide to where you want to be. Flying blind is not necessarily productive with subjective, tired ears.

Hot damn, that is some useful information. I've never thought to analyze my mix that way. usually, when two instruments are fighting I slap a frequency analyzer on each and try to tame each of them into their respective corners, but I've never considered putting it on the whole mix like that. Good stuff.
 
Sorry mate, but the mix sounds totally flattened to me, devoid of any dynamics. Pity, it's good music. The phone effect on the whole vocal line is tedious and the singer might be singing in chinese for all I know. The snare sounds like a trash can in a garage, the kick almost inexistent when all the other stuff is raging. This just doesn't work, so I suggest returning to the drawing board.
 
I still think you'd be better off recording better source sounds

That's the best advice you've recieved in this thread, and it only took 2 seconds to read.:eek:

You're spending hours re-mixing, re-tweaking, re-EQ'ing. You might not be aware of how important the tracking stage is. Your un-EQ'd, no-effects mix should sound pretty close to the finished product, as far as the sound of each instrument is concerned. If it doesn't, then all the tweaking in the world won't help you.
 
Last edited:
Back
Top