Loudness problems - newbie question

H

hreece

New member
This is a complete newbie question so I am going to post it here.

During my mixing, I am applying high pass filters to my leads and low pass filters to my bass. I am using a EQ on my drum machine instrument so that it is only displaying the vital parts in the signal which contain the kick, snare and hihat. Then a bass instrument, about 100-640hz. I try to make it in the order of kick, bass, synths, snare, synth and then hihat, so that each has a certain area and none are overlapping too much. Basically I aim to chop off all the unwanted and unused signal noise.

Once all my levels seem normal and I think I am going to have a reasonably-looking waveform, I mixdown my song and muddle with the EQ and levels in a wav editor. (Yes, this is my idea of 'mastering'.) Here I try to use as much leeway I can with my audio space to improve the loudness, by boosting the sound with a compressor and controlling it with a brickwall limiter.

...and my audio is still coming out very quiet! The sound is fine, but the loudness is very very low, even compared with other novice/amateur songs. Where am I going wrong?
 
I've just started messin with gettin my stuff louder too and have heard that the guys with loud mixes are doing multiple passes on their limiters.

But I'm not exactly sure. There's a few guys who have some exceptional recordings. Maybe they'll chime in here. ;)
 
hreece said:
I am using a EQ on my drum machine instrument so that it is only displaying the vital parts in the signal which contain the kick, snare and hihat.
Click to expand...
I'm not sure what you mean by that. What do you mean by "displaying"??? Are you using your eyes or your ears???

I try to make it in the order of kick, bass, synths, snare, synth and then hihat, so that each has a certain area and none are overlapping too much. Basically I aim to chop off all the unwanted and unused signal noise.
Click to expand...
Not sure you want to do that. Just because an instument seems to be in a certain frequency range, that doesn't mean you should be chopping off all other frequencies.

Once all my levels seem normal and I think I am going to have a reasonably-looking waveform,
Click to expand...
Again, are you using your eyes or your ears???
I mixdown my song and muddle with the EQ and levels in a wav editor. (Yes, this is my idea of 'mastering'.)
Click to expand...
If you mean you're muddling with EQ and levels of individual tracks, then you're not mastering, you're still mixing.

...and my audio is still coming out very quiet! The sound is fine, but the loudness is very very low, even compared with other novice/amateur songs. Where am I going wrong?
Click to expand...
That's hard to say. It could be way back at the tracking stage. How hot do you record your individual tracks? How loud do you push your faders up during mixdown? How loud do you try to get your mix when you're mixing down? You shouldn't be worrying about volume during any of those stages. No need to record "as hot as possible without peaking". Leave lots of headroom.

Commercial CD's are loud because of the limiting that's applied to them in the mastering stage, which is after the mixing stage.
 
I've got an entire rant about why you shouldn't master for loudness, but I'll hold it for now.

My guess is that your EQ at the mixing stage is killing your loudness. Our ears pick up midrange better than anything else. 70 db at 600 hz is going to sound much, much louder than 70 db at 50 or 12,000 hz even though it isn't. If you carve up each instrument, I'm guessing you are losing a ton of that midrange. Even though the wav "looks" full (because it is), it doesn't sound full (because the room is taken up by the wrong stuff).

Lay off the EQ a bit. And don't use any graphic displays. Just ears.
 
Hi guys, thanks for replying,

What I'm talking about is my approach to the production sequence. After I mixdown the song on my DAW, my next step to use a compressor on the mixdown, with the intention of simply compressing the signal - not using it to increase the db level much - and then using a limiter to do the final job of levelling it up. I know that there is a) no 'right' such way to do things, but that b) members such as Chibi are grinding their teeth at what I'm doing. This is what I'm really trying to get at.

70 db at 600 hz is going to sound much, much louder than 70 db at 50 or 12,000 hz even though it isn't. If you carve up each instrument, I'm guessing you are losing a ton of that midrange.
Click to expand...
I thought that for instruments to be effective when it comes to mixing that they shouldn't be severely conflicted with other instruments in the similar frequency. Are you saying to include as much within the 600hz range as possible before compressing?

As an example I am working with a synth that I have chopped off everything outside of 1-16khz. This is to let the bass and kick breath a little. If I don't do this, I completely lose my kick drum when I come to the chorus. When I remove the lower signals of the synth I can hear my bass drum again. I don't get how you are saying my mixdown is full when I have removed as much as possible from the signal?
 
If you're tracking at a peak of -24db then maybe it will always be a bit quiet...so what tracking levels are you getting? When I started out a few years ago of course I had low volume results too...but the problem was that I just wasn't recording loud enough in general. Anyway...it's a big topic and limiters and all that don't have to be used as basically loudness makers. If you have tracked all the parts at good levels then the combination of all your tracks could give you decent levels right away. Then, in the end you can use a limiter to up the volume by a few db or something. Anyway...use things how you want. I suppose there are some people who whack a limiter on every track and then on the master bus. I dunno really...these are just my general thoughts. My stuff is still quiet...so I haven't perfected it.
 
Monkey Allen said:
If you're tracking at a peak of -24db then maybe it will always be a bit quiet...so what tracking levels are you getting? When I started out a few years ago of course I had low volume results too...but the problem was that I just wasn't recording loud enough in general.
Click to expand...

This is backwards. Even if the end goal is screamin' loud, you don't have to push your tracking levels.
 
You need to mix to sound good to you and unfortunately there are no hard and fast rules
EQ wise you generally need to EQ the tracks in the mix (Not solo) so that they sit well to compliment each other and not necessarily to sound good individually
If there is mud you need to figure out where and cut it, If there isn't then maybe there is no reason to do a low frequency cut.
I use EQ to emphasise certain aspects of each instrument but not to try and make each instrument and voice live in an isolated bubble that never interacts with anything else. sometimes that might be necessary for a particular track or even desired for an effect but if you do it on every single track I would imagine your going to get quite a strange result.

Unfortunately the only answer I can really give you is get as much right during the arrangement and tracking as you can and then play/experiment with the EQ until it sounds right to you. Trust your ears to get the sound you want. After all it's your music.
Once you have that then getting it louder is actually the easy part.
 
Chibi - I just mean that -12db peak is louder than -24db peak for example...I remember recording stuff which peaked at -40db and I wondered why I wasn't getting a loud enough end result. You don't have to push your tracking levels, but you can't disregard them either right?
 
hreece said:
I thought that for instruments to be effective when it comes to mixing that they shouldn't be severely conflicted with other instruments in the similar frequency.
Click to expand...
This is a bit of a myth and greatly misunderstood.

Yeah, conflict is bad. But this applies to arrangement and tracking way more than mixing. If you have a guitar solo while the singer is singing, that is similar frequency conflict. Both are at about the same frequency, at the same time, requesting the same amount of attention.

The same situation where the guitar only plays a "response" in between the singer's words is probably not a conflict.

Conflict is also rarely an issue with instruments that are not the focus at the moment. A guitar strumming away at chords does not conflict with a singer even though they are the same frequency, at the same time. Likewise two guitars strumming the same chords do not conflict with each other, and they're literally doing the same thing at the same frequency at the same time. A guitar strumming away while a singer is singing that is so loud and full that it buries the singer and demands your attention over the singer...that is a conflict.

So when people say your bass guitar and bass drum shouldn't have frequency conflicts, that does not mean they can't overlap frequencies. It means you can't have a thunderous drum shouting "LOOK AT ME! LOOK AT ME!" and a thunderous bass guitar shouting "LOOK AT ME! LOOK AT ME!" at the same time. It is much more a game of directing the listener's attention than it is cutting up EQ. For example, your bass drum can be frenetic and busy causing the listener to focus on the speed and energy. In that case you probably shouldn't also expect the listener to focus on a chest-shaking explosion with every hit. So you record a drum with less low end. That frees up the bass guitar to really growl down low...as long as the bass line is simple with long held-out notes so it does not take attention away from the frenetic speed of the bass drum.

You could have a constant four-on-the-floor slamming bass drum with a twangy noodling bass.

You could have a groovin' booming bass drum with a groovin' booming bass guitar as long as they are doing the exact same thing and contributing to the same "focus of attention" instead of doing separate things each vying for their own attention.

etc.

Are you saying to include as much within the 600hz range as possible before compressing?
Click to expand...

No. I'm just saying to not cut frequencies in that range just for the sake of keeping instrument frequencies separate.
 
Last edited:
Yeah, I got more punch and clarity when I started tracking a little lower.
Still workin on gettin mine louder too tho.

So what about the multiple passes? :confused:

What I think I understood was to take the mixdown, strap a limiter with a little boost to it and do another mixdown. Take that one, limit with a little more boost etc...

??
 
Monkey Allen said:
Chibi - I just mean that -12db peak is louder than -24db peak for example...
Click to expand...

No. It is not. Neither of them are anything until you convert the numbers to analog and amplify the sound.

Even then you can't say the -12 peak is louder than the -24 peak. I could run the -24 peak through the PA in some arena and take your head off. I could run the -12 peak through the ringer on a cell phone and you could barely hear it.


If, and this is a big "if", the digital noise floor (the noise of your converters and all that digitizing crap) is above the recorded analog noise floor (the noise from your mics, the air moving in the room, the cars outside, the fan in your computer, your preamp) on both the -12 and -24 recording, then you can say the -12 recording has the potential to become louder before being swallowed by noise.

But I don't even think it's possible for the recorded analog noise floor to be that low in a home studio. Very unlikely that the digital noise floor is that high, too.

So if the digital noise floor is below the analog recorded noise floor (and it is), there is effectively no difference between the two. Both can be pushed up with faders to peak at 0. Both will have the recorded analog noise rise the same amount and end up at the same place. Both will have digital noise so far down it can't even be heard through the analog noise.


Peaking at -40... That's probably too low and you've pushed the analog noise below the digital noise. But it should be obvious to not track that low.
 
Last edited:
Yeah...I'm not a scientist...just make sure you record a decent signal, opening poster. It's a pretty simple message.

Rami...yeah I'm sure I used to get low acoustic tracks like that...
 
Monkey Allen said:
Rami...yeah I'm sure I used to get low acoustic tracks like that...
Click to expand...

Yeah, I was only questioning it in case it was a typo or something. I believe you, but I didn't think it was even possible to track that low.:eek:
 
Try recording and mixing with no eq. The best tracks I've done have no eq. I read somewhere (Fletcher at Mercenary I think) that eq is the work of Satan, and there's a lot of truth to that. So I always try to record so that I need no eq. I do end up using eq on every song I do, but very often stuff like the lead, if it's a vocal or sax or trumpet has zero eq, same with my drum overheads.

To me if you have to eq the shit out of everything then the sound sources need to be looked at... maybe use a different mic, change samples... always try to change the sound in the source, that's a recording basic.

It's like adding salt when you cook, you need to taste first and than add if needed, and not just as a matter of course add salt to everything.

Also, when I eq, I rarely boost anything, almost always subtract. Boosting, on most eq's, sucks to my ears.
 
Last edited:
In your wav editor, are you using the brickwall limiter only to control peaks after a compressor? Many people actually use a brickwall limiter to get their loudness. Naturally, this has sonic artifacts and can reduce fidelity, and is not necessarily the route you want to take, as it greatly reduces dynamics. However, if you’re trying to compete commercially in the loudness arena, it is an option open to you. To use a brickwall limiter in this way, set the output of the limiter to -0.1dB at most, and then crank the threshold down (or input up, depending on your limiter) until you get the loudness you want. Many limiters will have a reading of your RMS (average loudness) built right in, so you can see how your loudness compares with other recordings. It is possible to “prepare” your audio in certain ways to prevent the limiter from changing your sound too much, or sounding bad (like hi-passing your entire mix at about 30Hz for example).

Also, “carving-out” frequency niches for certain instruments can help those elements stand out, but you may be taking this idea too far, and cutting out too much spectral content. Keep EQ as subtle as you can (of course, sometimes you need a lot of EQ, sometimes you need none), and I normally only hi-pass and lo-pass at the extremes of the frequency spectrum.

Hope this helps,
Damian
 
hreece said:
Hi guys, thanks for replying,

What I'm talking about is my approach to the production sequence. After I mixdown the song on my DAW, my next step to use a compressor on the mixdown, with the intention of simply compressing the signal - not using it to increase the db level much - and then using a limiter to do the final job of levelling it up. I know that there is a) no 'right' such way to do things, but that b) members such as Chibi are grinding their teeth at what I'm doing. This is what I'm really trying to get at.


I thought that for instruments to be effective when it comes to mixing that they shouldn't be severely conflicted with other instruments in the similar frequency. Are you saying to include as much within the 600hz range as possible before compressing?

As an example I am working with a synth that I have chopped off everything outside of 1-16khz. This is to let the bass and kick breath a little. If I don't do this, I completely lose my kick drum when I come to the chorus. When I remove the lower signals of the synth I can hear my bass drum again. I don't get how you are saying my mixdown is full when I have removed as much as possible from the signal?
Click to expand...

Sounds like you are waaay over eqing things. The reason your kick is getting obliterated by the synth is because you've killed the definition in your kick by eliminating the frequencies that you think are outside it's influence. You can put alot of definition in a kick track simply by increasing the upper mids to gain some "snap". If you are using a drum machine, chances are the samples being used have been compressed and eq'd for a relatively optimal sound. By removing this information via subtractive eq, you are losing alot of tonal info.
While it is useful to remember where instruments sit in the tonal spectrum, it is not a hard and fast limit. Moving things around in the stereo field, in carefully incremental amounts, can free up alot of real estate in the middle of the mix..which is where things tend to mud up.
As for the volume component...I've never mixed for volume...ever. I do get my tracks as hot as possible and any gains there are to be made in volume, if I feel like I want any, are found in the mastering phase. For me and the material that I work with most, judicial usage of multi-band compression is beneficial.
 
You must log in or register to reply here.
Back
Top