Average Volume Question

Status
Not open for further replies.
4

4tracker

Guest
I'm working on a remix of some music my band recorded in the late 90s. The engineer had the final mix at -12 RMS and -.5 for clipping. My mix, when finished and even with a peak limited on it, has an RMS of -19.5. I'm actually find with that number because one reason I wanted to remix is that I wanted to lower the volume. But I am confused as to why I can't get as high an RMS as he did. Does this imply that I am mixing something dynamic, like the drums, louder than he did and that's giving me less headroom?
 
More likely you're not compressing the drums as heavily as he did, and you're probably not slamming the master comp/limiter as hard either.

19db crest factor is more than I would want for most styles of music most of the time. 11.5 is a bit squashed, but not exactly the loudest out there.
 
ashcat_lt said:
More likely you're not compressing the drums as heavily as he did.
Click to expand...

That's what I thought, thanks. Yeah I hated his mix so I'll go with -19ish and leave the drums more open sounding.

What I can't figure out is the right channel is closer to -20db. I've had this problem before on other mixes, and I'm not sure how to isolate what's causing it or how to correct it. Is there a method to evening out both l/r before exporting for mastering?
 
Every mix from modern rock has the RMS at -6db at least.

-12db is actually hard to find these days...
 
Unfortunately, different programs have.different calibrations for their respective numbers. I've got 4 sets of meters that don't agree with each other. One of them reads ridiculously high. It will be at -9rms when everything else is closers to -12rms.
 
RAMI said:
:eek: You're joking, right?
Click to expand...
Perhaps a slight exaggeration saying "every", but it's not that far off. Pull up some modern commercial mixes and analyze them. It can be pretty ugly.

I generally shoot for between -14 and -12, but it of course depends on what ends up sounding right for the song and if it's part of an album, then it has to make sense in context of the flow of the record.

What I used to do for "mastering" is pull it up in SoundForge and just look at the waveform. Normally you'll see that most of the peaks hit at a certain level, a couple hit maybe 3db above that and maybe one or two are a few db above that. If you peak normalize, that one or two are going to 0dbfs, but most of the actual peaks are only hitting at -6 or so, and the RMS for the track might be -18 or more. By clipping off that extra 6db, you're right about in the spot. It tends to be more transparent if you use a series of compressors to knock them down a bit more gracefully, but the point is made.

Nowadays I'm not going to SoundForge as often. I've gone to a master bus with RMS/lookahead compression and a soft clipper that pretty much just gives me appropriate crest factors when it sounds right.
 
I've never heard anything louder than -9db, and that's close to distorting. I'd love someone to point me to a -6 RMS song.
 
RAMI said:
I've never heard anything louder than -9db, and that's close to distorting. I'd love someone to point me to a -6 RMS song.
Click to expand...

Iggy's late-90s remix of Raw Power is deliberately at -6-ish and it's generally regarded as the "loudest" album ever. There isn't much else out there that really even comes close to that. Modern shit is loud, but it's not -6 loud. A professional -10 RMS will seem way louder than a home rec -10 because pro mastering engineers know what the fuck they're doing and have thousands and thousands of dollars of equipment to do it with.
 
Greg_L said:
Iggy's late-90s remix of Raw Power is deliberately at -6-ish and it's generally regarded as the "loudest" album ever. There isn't much else out there that really even comes close to that. Modern shit is loud, but it's not -6 loud. A professional -10 RMS will seem way louder than a home rec -10 because pro mastering engineers know what the fuck they're doing and have thousands and thousands of dollars of equipment to do it with.
Click to expand...
Exactly. "Most modern mixes are -6 RMS" or whatever his exact words were, is false. In fact, if there are degrees of falseness, it's VERY false.
 
When I was using soundforge for mastering, I would just edit those 3-4 peaks before I normalized. It's even more transparent than a compressor.

If it became necessary to get the mix louder, then I would go with the compressors and limiters. Shooting out the huge peaks helps keep the compressors smooth.
 
Farview said:
When I was using soundforge for mastering, I would just edit those 3-4 peaks before I normalized. It's even more transparent than a compressor.

If it became necessary to get the mix louder, then I would go with the compressors and limiters. Shooting out the huge peaks helps keep the compressors smooth.
Click to expand...

How did you find the peaks and how did you edit them, manually find them and pull them down? You scan the wave form by eye? Is there a way to automate this?
I'd rather not use compression most I can avoid it. I have a little on bass and vocal, but I don't want to compress the entire mix.

I keep having a slight mismatch between left and right peaks and rms. Would this require a limiter to fix or is there a way to manually figure out why?

When I loaded some of my favorite records and analyzed the RMS, it was around -19db. I believe I read some of this had to do with limitations of vinyl? If so, my ear is accustomed to that level and finds its pleasing. I don't find loud pleasing. In fact, I find weird distortions (at least in home recording mixes that are at high volume) that hurt my ear. Maybe the pros do it great as mentioned, but I'm a home recorder and don't have that ability.
 
RAMI and Greg, just listen to a Foo Fighters record. It is compressed and loud to the point it reaches -6db of RMS level.

I think you two are so busy trying to think that you guys are good and know everything better than everyone, that you have to try and be rude everytime with everyone.

Check out this link: Album details - Dynamic Range Database
 
4tracker said:
How did you find the peaks and how did you edit them, manually find them and pull them down? You scan the wave form by eye? Is there a way to automate this?
I'd rather not use compression most I can avoid it. I have a little on bass and vocal, but I don't want to compress the entire mix.

I keep having a slight mismatch between left and right peaks and rms. Would this require a limiter to fix or is there a way to manually figure out why?

When I loaded some of my favorite records and analyzed the RMS, it was around -19db. I believe I read some of this had to do with limitations of vinyl? If so, my ear is accustomed to that level and finds its pleasing. I don't find loud pleasing. In fact, I find weird distortions (at least in home recording mixes that are at high volume) that hurt my ear. Maybe the pros do it great as mentioned, but I'm a home recorder and don't have that ability.
Click to expand...

It has definately because you have to make sure that when mastering with vinyl, the peaks dont even come close 0db, because the chances of clipping and ruining the vinyl are higher...you know that.

The best mastered album I've ever heard is Abbey Road, from The Beatles...the vinyl version, of course. I think that the RMS is about -19db, around that.

How do you master your songs, man?
 
CMolena said:
RAMI and Greg, just listen to a Foo Fighters record. It is compressed and loud to the point it reaches -6db of RMS level.

I think you two are so busy trying to think that you guys are good and know everything better than everyone, that you have to try and be rude everytime with everyone.

Check out this link: Album details - Dynamic Range Database
Click to expand...

:facepalm:

You need to stop posting so much and do more reading, learning, studying, understanding, then practice. A bunch.
 
Farview said:
Unfortunately, different programs have.different calibrations for their respective numbers. I've got 4 sets of meters that don't agree with each other. One of them reads ridiculously high. It will be at -9rms when everything else is closers to -12rms.
Click to expand...

This is important to account for when discussing RMS values. There are two ways of determining the 0dB against which the RMS is measured. One uses a 0dBFS peak sine wave and one uses a 0dBFS peak square wave. Measurements using the sine wave reference come out something like 3dB higher than those using the square wave reference. If you don't know what the reference method of the numbers given you don't have an accurate idea of the loudness.

For example, in Sound Forge I have a song that measures -17.5dB/-17.2dB (left/right) but in the TT DR Meter it reads -14.6dB/-14.2dB RMS and 13dB/12.5dB dynamic range. That's not a very loud file. A moderately loud song would come in around -11dB, -8dB and -6.5dB by those three methods.
 
Status
Not open for further replies.
Back
Top