a little problem

  • Thread starter Thread starter Rusty K
  • Start date Start date
Rusty K said:
The only thing I'm fixated on is how the "majors" are able to squeeze a wav right up to 0db without distortions and while maintaining fidelity.
They don't. You should hear how much better those recordings sound before they smash the living daylights out of them.

I'd like you to find for me one quote from what you've read that is from an A-list, big name engineer who actually believes that the War On Volume is actually a good idea. In the meantime I can probably find you at least a dozen without breaking a sweat who think that it's a bunch of hooey. And that's not even including the several pro mixing engineers and mastering engineers who hang out here who agree completly.

But if you want to master square waves like the kind you're talking about, it's easy; boost your gain until you're clipping at least 6dB over, 20%-40% of the time (turn your monitors way down when you do this to save both them and your ears :)). Then throw in heavy limiting to pull the signal back just enough to turn the clip indicators off. Your recording will have an extremely high RMS and will be LOUD. It will also sound lousy.

G.
 
I'm pretty sure Rusty is still talking about peak and not RMS level. Rusty, read up on the difference between the two.
 
I use PT 7.0 LE. after i get it mixed i use the waves 10 band paragraphic eq followed by the waves ultramaximizer. that thing rocks, i pull down the threshhold until limiting occurs and then pull it a little more. set the ceiling at -0.1 and man is it loud and clean. mix it hot as shit.
 
mshilarious,

I'll do that cause I don't really know the diff. between RMS and peak but I think I'm just trying to flatten the peaks and boost the overall level. I don't normalize.

Most of my music (ballads, blues etc.) doesn't need to be pushed to the max but I write pop tunes as well. Now I haven't analyzed every pop CD mix out there and I'm certainly not privy to those mixes before they're mastered/squared but if you're in that market you need to be able to play that game. Your song has to go head to head with what's out the on the air waves. I'm not even saying that I could accomplish this in my little home setting, probably not, but I'd like to understand it and strive for it in certain scenarios.


Rusty K
 
Rusty,

just some thoughts i had while going through this thread. i hope it's not redundant information to you...

1. i use 7506 phones and think they can handle very high signal levels with relative ease. i wouldn't disregard the distortion you're hearing in them if it's not present on commercial cds at the same loudness. so in my opinion, your distortion may unfortunately be real.

2. unless your target audience will mostly be listening on headphones, don't use them as a total standard for placement of tracks in the mix. Bluebear has an article somewhere that addresses the shortcomings of mixing with them and could explain it much better than i can. i still work on headphones all the time and check mixes on them, but i wouldn't *rely* on them for final decisions.

3. regarding Yareek's levels, it sounds like he's referring to RMS levels at -14dB. peaks can definitely hit 0dB, and some cds shamelessly clip at 0 (where you have multiple samples in a row at 0, as opposed to a single sample hitting 0). it's fine for a mastered track to hit 0, and that takes full advantage of dynamic range. some mixing engineers will recommend not pushing levels too high though, as it will give the mastering engineer more headroom to work with when raising the RMS or average levels of your mixes, and a mastering engineer will (hopefully) have higher quality equipment to raise those levels cleanly.

4. backing down on the mixdown levels will probably not change your situation unless you're treating individual tracks differently. if you're just pulling the master down so you're peaking at -3 instead of -1, then it's not going to matter because the audio will still be pushed up in mastering. in fact, lowering the master level will technically be one more stage of processing in your DAW, which will introduce another stage of rounding errors (though whether you notice them or not is arguable). now if you're treating individual tracks differently, you may have some better results. are you by chance working with heavily compressed material, or are you using heavy compression in your mixdown? actually, what kind of music are you mixing?

5. i'm so glad to hear glen's comment on the sound of masters. it wasn't until i started seriously attempting to master my own material (and we can save that debate for another time since it's heated :)) that i started hearing how crappy commercial cds could sound. i'd love to compare mixdowns and masters of the same tracks at equal loudness. just hearing clarity and openness would be pretty amazing. i love compression, but i'm not the biggest fan of "loud". my "finalized" mixes had RMS levels around -15 give or take a dB or two, and i was content with their sound. they didn't light up my stereo as nicely as i'd have liked, but they had a cleaner, more focused sound which was more important than volume for me.

this process really helped me: if you've got time to experiment, try remixing a problematic track with less or even without any compression (especially on the master bus). just save it as another project, bypass every compressor, and just remix it with the track faders. then put it through your mastering chain and see if the sound is improved relative to the levels (both peak and RMS). limiting is a compromise. you gain loudness at the expense of fidelity, and you'll have to draw your own lines on what your goal is. MassiveMaster has a quote in his signature stating that the process of getting a good mix starts before you hit the record button, and it's very true, especially for a loud mix. of course, if you intend to have these tracks mastered, then screw it all and just go for the best sounding mixdown (without any master bus compression!), regardless of where your levels fall**.

good luck,
marcus

** as long as they peak below 0dB!
 
Last edited:
Marcus,

Thanks for your post.

1. 2. Cool, I AB between the phones and monitors and come to a decision after both. I'm sure now the distortion is real.

3. Yes I was just monitoring a friends CD done here in a pro studio and mastered. Sounds great but it's clipped. Can't hear it but it's there when analyzed. Yes I don't really understand RMS vs. peak. I need to bone up on that.

4. Yes I've also come to the conclusion over the course of this thread and my experimentation that the level of my mix doesn't seem to make much difference in my result....except for the leaving of headroom for another to master.

5. I say to all my friends when we're talking about home recording that I really only consider my product as demo. That doesn't mean I'm not going to try to do the best I can and learn the process so that I'm more knowledgeable when I hand my work over to a professional. On the other hand about "lighting up your stereo". I don't know what your purposes are but here in A it's amazing what gets played on the radio. It's obvious that a lot I hear on the ecclectic stations could have easily come out of a home setting. For that reason it should measure up well just in the offhand someone should deem it worthy of airtime.

I'm only compressing as I record these days. It could be that I compressed to much but it does level out those peaks that become problematic later....a trade off. When I get to the mastering phase, my master suite has many presets and I start from there, trying many different settings and listening to the differences. Sometime I squeeze the hell out of it but with respect to the subject of this thread I seem to be getting these distortions no matter what combination of compression or limiting I use when ever I start trying to push the level up.

I'm beginning to wonder if I have maybe clipped my preamp during the recording process. I have a Grace 101 which I love but the red input light starts coming on at -10db and just gets a little brighter as the level increases. It's hard to tell whether or not I've clipped it. I'm not hearing anything. In the manual the red light is ok up to a point.

Thanks again for your helpful post.

Rusty K
 
there's not a whole lot to bone up on between RMS and peak levels. just think of peak levels as the highest peak in a waveform. RMS levels can be thought of as an "average" of the power in a waveform. So while your friend's mastered demo CD may have peaks that go all the way up to 0dB (and that's fine), the average loudness is not 0dB. The RMS level may be somewhere around -20dB (relatively soft) to -10dB (relatively loud). RMS doesn't represent an individual point. It's a reflection of the changing levels of the music over time.

Peak and RMS are somewhat related, but certainly not closely related in all cases. A 5 second sample of a bass guitar with a big sustain can have a low peak value when compared to a 5 second sample of a single snare hit, but it may have a higher RMS level because over that 5 second sample, it's still at a high level, whereas the snare drops off quickly into maybe 4 seconds of silence after its hit.

Hopefully I explained that clearly enough. If you just keep in mind that RMS is close enough to "average" level over time, and peak level is technically a single sample, that's probably all you'll need.

If you're a glutton for punishment, here's some math stuff:
http://en.wikipedia.org/wiki/Root_mean_square

When i mentioned lighting up my stereo, i was referring to the spectrum analyzer on it. A lot of commercial CDs are very loud and heavily compressed/limited. Their peak levels are lowered, which means that the master level of the entire song can be raised without clipping (or in some cases, without absurd clipping). When the master level is raised, all frequencies are boosted, which means that there's more energy present in each frequency band on my spectrum analyzer, which lights it up a lot across the board. My music isn't as compressed or limited, so the levels in each band aren't as high, which makes for less exciting viewing of the stereo, but a more satisfying sound in my opinion.

As far as radio goes, I'm definitely more interested in the college and public radio stations. I totally agree with your fidelity assessment of some stuff. Lots of tracks on a local college station sound like demos in comparison to mainstream stations, even though they're not demos. Personally, sometimes I find myself analyzing their sound and wishing they would have miked the drums different or lowered the vocal levels, but that's the tradeoff you get for hearing the variety of music you don't get on other stations. I'm kind of bitter about mainstream radio :)

Regarding compression on recording, I'm sure you've read on this board all the do's and don'ts, so i won't go down that road. I'll just underscore that it can't be undone, so tread lightly. Regarding mastering presets... uh... I'd avoid those. Not that they aren't useful, but I think it's more advantageous to think about what your music needs before you start affecting it. Ears can be tricked pretty easily and they also adapt quickly. I would suggest really listening closely to your mixdowns to find out where their shortcomings are. Reference your mixes against other CDs through the same signal path if you can. From there, try to fix the mixdown first! You have so much more control of the multitrack project, you may as well take advantage of that. Get the mix as perfect as you can before even thinking about what you'll do in mastering. Then, when your mixdown is just right, give your ears a rest (i'm talking days not hours) and come back to it. Reference again with CDs, and know what you need to do before putting on effects or enhancements. I'm not a supporter of slapping things on just because you can. Everything in the mastering chain should serve to treat a *problem* in the music. If you're not hearing a problem, i feel it's best to leave it alone, even if your plug ins are world class. Oh, and bypass frequently to make sure you're staying on the right track. oops. I didn't mean to preach there. i'll shut up now.

good luck,
marcus
 
You might want to make some test tracks with your preamp to see if you are indeed distorting at that stage. The results of that would tell you a lot.

Also, recording at sustained levels of -10 and above are much louder than they should be. When recording, you might want to light up that red light for an instant on a peak, but not stay in the red for any amount of time. If the red light comes on and stays on, any resulting distortion is not the preamps fault, it's yours for putting too much level through it. The red light is there to say "you're too loud!", that's it's whole purpose.

Keep you recording levels between -20 and -10 and you'll have a good clean recording and plenty of level to deal with when mixing.
 
What limiter are you using, anyway? Not all limiters are made equal IMO.
And don't rely too heavily on the limiter. Most are not made to take much abuse; merely tame the stray peaks. If your limiter is constantly reducing gain, it will sound like crap. If the release is too short, it will sound like crap.

BTW, before I go any farther, your mix doesn't have to be smashed to heck to sound good on the radio. The FM stations will do plenty of smashing for you. In fact, some of my old lightly-mastered songs sounded awesome when they were on the radio, even compared to the mainstream songs that came before and after. A lot of times, overcompressed stuff sounds a little poopier when played over the radio.

OK, back to self-mastering. I don't know what you are doing to your mixes, but don't count on finding a preset that works for you. I wouldn't even bother with all-in-one mastering presets at all. The program has no idea what your mix needs. You need to be able to hear what needs to be changed (this is the hard part of self-mastering), and then you need to know exactly how to make the changes for yourself. Here is a simple method I might suggest. First of all, have a few reference CDs handy; check back to them throughout the process. Take your cleanest, least abusive EQ and start a high pass filter at 20Hz, and see if you can move it up any higher without messing up the bass. Play with different slopes if you have the option. Or you could see if a more gentle low freq shelf starting around 40-50Hz sounds better. Next, dig around in the low mids (100-450 area) and see if there is any mud you can dip out with a db or 2 of EQ. Play with the width (Q) of the cut and see how wide or narrow the cut can be. Next, move to the mids (750-2kHz area) and see if your mix needs a narrow cut somewhere or a little bit of a broad Q boost. Probably going to be one or the other. Next, you can try messing around in the 3kHz-12kHz range, but this area can be dangerous if abused. I'll just leave that one alone.

OK, now put that through your nicest compressor; start with a ratio of 2:1, attack of perhaps 10ms or so, and Auto release if you have it. Bring the threshold down to where you can see gain reduction of maybe 3-6 db, but it isn't making the music pump too stupidly. Now play with your ratio and see what happens with less and with more. Threshold may need to be adjusted as well. Also, make sure you bring up the makeup gain so that the compressed signal has the same apparent volume as the bypassed signal. Makes comparing easier.

After this, put on your coolest sounding shelfing EQ (maybe Pultec or linear phase) and give it that shiny sound with a fairly steep high shelf boost at 16-22kHz. You can also try making this shelf with a normal bell-curve boost that peaks at like 22-24kHz. You want to make sure that the boost is a bit steep so that the harsher 3-12kHz range isn't getting boosted too painfully. Now do some careful listening to the bass range and see if you need more or less bass relative to your ref CDs. May need to take it to the car to check too. If adjustment is needed, use a gentle low shelf starting at wherever (80-250Hz?) to raise or lower the overall bass content if needed.

NOW, throw your fanciest limiter on the master buss (UA Precision Lim, Waves L2), preferably not multiband, and raise the gain until you get lights bouncing on and off on the gain reduction. THE GAIN REDUCTION LIGHTS ON THE LIMITER SHOULD NOT BE CONSTANTLY ON THROUGHOUT THE MIX! NOT EVEN THE FIRST LITTLE LIGHT! OK, nuff said on that. If you have control over attack and release, you will prolly want the fastest attack available, and Auto release. If you have to manually adjust release, beware of fast settings as this will cause distortion in your mix. Peak level should hit -.5 to 0.0 db depending on your preference. I do -0.1

Who the heck am I? just another schmuck who has learned a few things and done a few things. Nothing major-label, but some of my work for people has been mass replicated. Hope this rediculous-long post at least gives you some avenues to look into. These techniques are guaranteed to not work all of the time.

Gawd I must be bored. :rolleyes:
 
This has been a very rewarding thread for me. I hope is has been for others as well.

I just tried running my monitor amp very low and maxing my soundcard volume fader....couldn't hear any distrotions. I'd definitely recommend using headphones during mastering for at least one playthrough to catch these problems.


As far as the RMS vs. Peak, I pretty much understand average vs. peak and right now I don't have an RMS settings on any of my compressors. My software does analyze RMS data. If I understand then if I depressed an RMS button on a compressor I would be getting compression over an average range rather than just peaks whereas the peak button would act more like a limiter.

SonicAlbert....thanks I will follow that rule from now on although I don't think I've clipped my pre amp. I usually check the individual track by ear for suspected clips. Just one question...many have said that they record to -3db. Are you saying ignore this rule of thumb or were you just refering to the pre amp level?

I try to follow most of the suggestions I've read posted. I trust my ears for the most part. I'm a musician first and engineer second but I do know there have been many a nights when I've gone to bed thinking that I had a mix only to come back later (ears rested) to find a mix that sucks. I also know that I have to be very careful with eqing at the master stage. Like my professional engineer friend as well as others here have told me.... If you have to eq a lot at the master point it just means your mix is not that good.

I've noticed that cutting at the low end cause the hign end to jump, sometimes too much if I'm not careful. I do have a bit of trouble with 200hz and often cut there but just a little (trying to learn how to get rid of all this at the mix stage). Also harshness...I try to smooth that out a bit sometimes at 2k or so. And I usually cut at 20hz. I have the most trouble with bass (ironic since I'm a bass player) but I'm getting it under control better now at the mix stage.

Thanks,

Rusty K
 
Me again....

This thread is getting very long so if someone thinks I should start a new one I will, however the original question is still unanswered and ongoing and I'm learning as posts come along and I hope others are as well.

I've had a breakthrough because of some of the replies to my thread. The mastering suite that I've been using is a stand alone program. Because of a response, I took a mastering eq plug and a mastering compressor out of "moth balls" and tried using it with a master buss....what a dream. No more mixing down and deleting over and over, sometimes envolving resampling and bouncing between programs. With the plugs I can master on "the fly" saving tons of time while getting a more concise mental picture of what is really going on with my mix.

My master eq plug is fantastic (Timeworks Millinium Master Eq) Their master compressor is also very easy to use but it doesn't have a way to boost input...I'm still trying adapt to that. I'd appreciate any suggestions.

The distortion problem is still there on the song I've been using to experiment with. I'm getting distortion (phones only) in places where there is no indication of clipping on any of the plugs or in my multitrack software. I tried working with another song with greater success ( the distortion was there when I pushed too hard but that was explainable). I'm leaning more toward the possibility that I clipped my preamp but since i can't tell by listening to the individual tracks I'm at a loss as to how to fix the problem. I'm even hearing the distortion pre mastering if I mix too hot though again there is no indication of clipping that I can see. I'd recut if I knew which track...even if it was multiple tracks.

Any solutions would be greatly appreciated.

Thanks again,

Rusty K
 
Rusty K said:
SonicAlbert....thanks I will follow that rule from now on although I don't think I've clipped my pre amp. I usually check the individual track by ear for suspected clips. Just one question...many have said that they record to -3db. Are you saying ignore this rule of thumb or were you just refering to the pre amp level?

This is where it gets dangerous, things like this "rule of thumb" that someone tossed at you. We are all susceptable to things like this. Someone says something and it just kind of sticks in the brain and becomes a matter of course, without us giving it the full evaluation it should deserve.

You need to understand how loud -3db is, and only then will it really hit you that recording every track that hot pushes your gear to the limit in ways that are not necessarily good.

When you are listening to your tracks at -3 or mastered to just under 0db, it's important to remember that your master volume control is functioning as an attenuator. In other words, you are turning down the volume in order to bring it to a level that you can actually listen to without blowing your ears or speakers out. If you calibrate your playback system to actually deliver the true volume of the tracks, you'll seriously risk your ears and monitors.

So recording every track at -3 is really loud, you've got nowhere to go. On your next song try keeping levels peaking at -6 and generally between -20 and -6 overall, when tracking. Don't be concerned about smashing everything up as close to 0 as possible. Then when you are mixing see if it overall sounds better and is easier to mix without having the kinds of problems you are having now.

The important thing to remember is that you can always turn the overall volume of the track up at the final mastering stage. I'm not saying record soft, just don't slam everything so high when tracking.
 
Rusty K said:
The distortion problem is still there on the song I've been using to experiment with. I'm getting distortion (phones only) in places where there is no indication of clipping on any of the plugs or in my multitrack software.

Just to add, have you tried listening to the tracks on different phones?
 
SonicAlbert,

Once again thank you for your helpful responses.

When one is a newbie and looking for starting places you tend to reach for those "rule of thumb" life preservers. I'm not exactly a newbie because I have found many of these "life preservers" to be more like heavy weights. The -3db rule is a barnacle I will now let fall to the bottom of the tank.

I only have one set of headphones. Maybe I can borrow another but an earlier post from someone using the same model I use implied that I should trust them. Of course I've had them awhile and they could be breaking up but I can't ever remember pushing them over 50 or 60%. These 58 yo rock n roll musicians ears can't take too much.

Thanks again,
Rusty K
 
The headphones could be breaking up, so it's just a matter of eliminating that possibility. Even good headphones can go bad, so you just want to make sure that's not the problem. It probably isn't, but it's worth looking into for peace of mind.

Your final mix can certainly have peaks around -6 to -3, and after mastering you can have peaks up to -.1 (I prefer not to go higher than -.3). You just need to leave room for those later stages of the process.

I'm somewhat aware of these kinds of issues because my brother is a music hobbiest and spends a lot of time in his home studio. I set him up on my old Mac with ProTools. He sends me sessions sometimes that just blow my mind, everything is cranked up way too far for its own good.
 
Back
Top