Jump to content

How loud to make the drums?


Recommended Posts

  • Members

I have a different question that might be relevant. When I take one of my mixes and squish it through a limiter to get rms levels up a bit, it seems to kill the drums. The relationship in perceived volume between drums and say, guitar changes. I do understand why, it just makes me wonder if some of the big guns mix a bit drum heavy to compensate. I am sending some tracks to be mastered soon and hope the issue is with my "mastering" and not my mixing.

 

I am sending my drums in at 140db ;)

Link to comment
Share on other sites

  • Members

I don't think I could put a specific number on it. I've got some mixes that peak at -10 or -8 dB...and I've got others that peak at -18 or -16 dB. I think as soon as you get an answer that says, "Drums should peak at -11 dB," you've started down the wrong road.

 

Every mix is different, so I'd say do whatever sounds the best. Be sure to use reference material and listen to the balance there.

 

As far as levels before mastering, give your mastering engineer a call and ask him what he prefers. He should be able to direct you.

Link to comment
Share on other sites

  • Members

I have a different question that might be relevant. When I take one of my mixes and squish it through a limiter to get rms levels up a bit, it seems to kill the drums. The relationship in perceived volume between drums and say, guitar changes. I do understand why, it just makes me wonder if some of the big guns mix a bit drum heavy to compensate. I am sending some tracks to be mastered soon and hope the issue is with my "mastering" and not my mixing.


I am sending my drums in at 140db
;)

 

This is where multiband limiting comes in handy. If you're loosing kick or snare you can target those frequencies so that doesnt occur. I wouldnt say boosting something to have it more proninant later isnt how its done. Just having enough should surfice. The opposite is likely to occur. If something is too loud, its going to get limited.

 

My best guess from your description is theres too much compression on the tracks. It seems to be the #1 problem with many who make this complaint.

 

Second might be targeting the frequency responce of the instruments. If every instrument is targeted to its own frequency range, problems with masking wont occur and the instrument will continue to stand out after mastering.

 

Trying to make something sound "bigger" may be possible by widening the frequency of that instrument, not a volume war between it and its compeditor. It can also be done by narrowing the responce of the others nearby. What cant be done is to make them both equally wide in responcein the same frequency range or ones going to get masked by the other. This is where many beginners make the mistake of soloing an instrument, mixing it to sound big, then wondering why it dissapears in the mix when played with the other instruments, in fact, an instrument soloed bay sound terrible soloed, but in a mix it sits just right with no interference by other instruments.

 

Prudent use of reverb can help separate things in the same ranges too. It does come down to the mix though and what you're trying to get.

Link to comment
Share on other sites

  • Members

 

This is where multiband limiting comes in handy...

 

 

OTOH, the first thing that struck me is that the drums might have been mixed without any/sufficient compression, and all the transients are in the snare and kick. Therefore, when mashed with a limiter, all the previously spiky drum peaks get knocked back, but almost nothing else on the track sees much gain reduction.

Link to comment
Share on other sites

  • Members

 

How loud should the drums be in a standard rock song before it's sent to be mastered (in decibels)?

 

 

 

I don't know.... is it the sort of rock song that has loud drums or the sort of rock song that has softer drums?

 

 

Your mix should sound like a rock song before it goes to mastering. So ask yourself, "are my drums loud enough". This answer should change song to song.

 

 

Also, if you're writing a "standard rock song" you might have another problem entirely.

Link to comment
Share on other sites

  • Moderators

I learned a cool trick from Mick Jagger. Not that I was hanging out with Mick or anything, but in a rare, music recording oriented interview with him he said (paraphrased), "Turn the track all the way down so you can't hear it. Just a barely audible peep. You should only hear the vocal and snare."

 

It works.

Link to comment
Share on other sites

  • Moderators

I'm currently having an album mastered by a very experienced mastering engineer and he noted that I had normalized some of the tracks, asked for the non-normalized version on one of them.

 

This piqued my interest, so I said, why this one and not those?

 

He said, "This track has a lot of dynamics on it, it would be easier to get it through my signal chain if it was, say, 3dB down from 0. But if you don't have the original, non-normalized mix, just be aware that I'll start out by taking it down by 10dB or so."

 

I asked him, "Do you think that matters to the sound? I trust your judgement but the scientist in me thinks that a 24 bit recording that had been normalized up 10dB then dropped back down the same amount would be indistinguishable from the original. We're talking about creating 1-2 dithered bits at the bottom on the pull up, then ditching them on the drop, aren't we? Is there even a remote chance that's audible?"

 

He replied, "I'm a purist. I think I can hear that in the high frequencies."

 

I told him, "How about I give you the 32bit float version," thinking to trick him.

 

He said, "That'll be fine."

 

:idk:

 

Terry D.

Link to comment
Share on other sites

  • Members

As a generic non technical reference, I bring drums and bass in @ -4db, guitars and horns and keys @-2db and the background vocals @ -1 db and lead vocals @ 0.db In reality, the snare peaks @ about -2db but the kit overall is less;@ -4db. This is on mixed down material that's been mastered to my gear. Additional outboard mastering may be required and or desired according to wealth and taste...

Link to comment
Share on other sites

  • Members

 

I learned a cool trick from Mick Jagger. Not that I was hanging out with Mick or anything, but in a rare, music recording oriented interview with him he said (paraphrased), "Turn the track all the way down so you can't hear it. Just a barely audible peep. You should only hear the vocal and snare."


It works.

 

 

I've heard this too - not that I hang out with Mick either.

 

And the main thing to take from this advice, whether anyone thinks this is a good idea or not, is to LISTEN.

 

Stop looking at yer damn meters and LISTEN.

Link to comment
Share on other sites

  • Moderators

 

Terry, why did you normalize some tracks in the first place? That's generally considered a no-no.

 

 

I hear that repeated a lot, but so far no one has provided a credible answer as to why it's a bad idea. Normalizing a 32bit floating point file changes it in no significant way whatsoever. That's the point of using 32 bits.

 

I normalize them to slightly even out the balance between the mixes without compressing them, so I can listen to them more easily.

 

Terry D.

Link to comment
Share on other sites

  • Moderators

I can't comprehend this question. What about the mix?!
;)

 

Yeah, it's really two questions jumbled together.

 

(1) What level to make the tracks sent to master. DOES NOT MATTER (send 24 or 32 bit with no digital overs).

 

(2) How loud should the drums be in a mix? HOW LOUD DO YOU LIKE THEM?

 

Terry D.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...