Mastering - PG Music Forums

Scott.

I can assure you that here in the UK it's just a complete mess. There's no coherence at all between programs, announcements, trailers, or individual advertisements. I find the lack of professionalism, compared to say thirty years ago, to be worrying in the extreme.

It does not bode well for the future.

ROG.

Mastering, for me, is the final step to creating a pleasing, unified sound across all tracks (not just volume), sequencing the tracks into a pleasing order & pacing, and making sure there is "digital black" in between each track.

After that, making sure all the artwork is correct, the tracks are listed in order, and the final ISO is correct, with no flaws.

I consider this a separate process, and usually wait a week or so before doing it to clean my ears out.

Home single song mastering can be done using Final Mix, Ozone, and T-Racks. These are popular because everything you need is prepackaged together with presets to tweak.
If you read around Waves is also popular high end but expensive.

Home multiple song album/cd mastering can be done using Sony software - I think it's Sony Architect or something like that and probably some others.

Real mastering is done by a mastering engineer. They polish the sound. They also embed the codes and make sure all is industry standard for you. Cost varies depending on who you get.

Using any of the above - the better the mix the better the master.

Okay, many good answers later, let me weigh in with some thoughts.

When I see the reply that addresses "taking your collection of songs and getting them ready for mass production and sale", or words to that effect, here's what I have a problem with.

The songs for the country album I have written were written over 2 years, during which my gear saw MANY changes in direction and rethinking of philosophy. My skills in recording have also grown as I have learned about using effects more adeptly. Given all of that, the 11 songs are at ALL different levels of quality. (And honestly, this is a home studio project that was never intended to be printed and sold so I am not about to record any of them again.)

When a touring, active band decides to record a new album, that all gets done over a period of a couple of weeks, where you go back to the same studio conditions every day with the same producer and recording engineer doing things their way, another constant. Those 11 songs are a MUCH more level, even, "smooth" product to start the process of final mix and master. As I listen to my stuff on a thumb drive while I drive I can hear DRASTIC differences in how they are EQd, which ones were done before I discovered (after some gentle nudging) onboard effects and in the box mixing. I have never even touched a comp limiter before, and the 13 songs on the thumb drive are ALL over the place.

So this next project will see me take more accurate notes and screen shots of how the EQ, reverb, limiter, etc.... are set when I have the sound I want. For most of those original 13, it's too late as I was doing outboard verb and EQ and it was destructive editing, and I don't have clean tracks to start over with. (Yes, dummy Eddie edited originals instead of making a copy.)

It does bum me out that when that stuff plays, track 1 is okay, track 2 lacks highs, track 3 is okay, track 4 lacks low end.... and so forth. Hindsight being 20/20, I know better now. The 4-5 I am really at all proud of for content were really only be meant to send to lower level recording artists looking for material to sing for their demo. Remember, as I have said all along, I don't care about playing. I want to write.

Eddie,

In this case, I think I will highly recommend that you try a 'mastering' plugin and use the same preset across these 11 different songs. You may be highly surprised at the outcome being much more consistent than you would expect.

Did you buy Tracktion when it was $19 at Musician's Friend? If so, then you have Final Mix - which is worth much more than the $19.

Actually, even if you don't, I will do you a favor - send me a CD with .wav files on it of the songs and I will send you back a CD with 320 kbps .mp3 files, each in a folder for which I've applied a different preset or tweak of a preset for the 11 songs. It might take me awhile to get it done, but this might help out somewhat for you to appreciate the power of multi-band EQ, multi-band dynamics processing with limiting that most of these plugins combine with finesse.

To make this a shorter undertaking, perhaps limit the tracks to 3 of the 11; 3 drastically different sonic signature tracks.

-Scott

Ed, go grab the free 21 day trial of AAMS & start running the tracks thru it.

It WILL take some time for each one to be analyzed, depending on your system but after the first time thru you can start trying out all the Reference Files to see if you can find one to use.

AAMS works best on an XP x32 system, but can be made to work on Win7 x64.

What have ya got to lose? You can always take Scott up on his offer if ya don't like the results...

I am actually going to heed BOTH of those posts. Scott, if you will PM me a mailing address I will get that done, and JC I will grab that trial download.

Even when the stuff is just for me, the over achieving geek in me wants it perfect.

Eddie you have said many times and in many different threads, you're a perfectionist and you really like knowing exactly what's going on and how it all works. Very laudable. With that in mind, here's the correct, classic definition from Wikipedia. This is what you would get from graduates of those Audio Engineering schools we talked about. Big difference between working pros in the field and us hacks at home using our little setups. You want to be a pro, then here it is.

http://en.wikipedia.org/wiki/Audio_mastering

You also mentioned in the other thread in response to someone suggesting you get Ozone that you don't want simple software presets, you want to know the why's and wherefore's of all that. That's great but the presets in these top mastering programs are pretty darn good and are specifically designed for folks like us who don't have that formal education. I would be using those presets for now while you're busy signing up and working those classes. Otherwise, you're getting nothing done for the next few years.

Bob

Even with formal training, the presets are nice to 'start from'. If I know a certain preset has the effects I want in it (in Ozone there are about a dozen possible effects) sometimes it helps save time, as opposed to starting from scratch.
Same with any of my effects, as you use them more you'll learn certain presets. Reverbs, mastering effects, guitar amp sims .. tinkering with them over time can be a good way to learn them. The more I use them the more I know about them.
YMMV
Grab a 10 day free trial of Ozone. Try it.
Look at the presets, find a couple you 'kinda' like and see what they are doing. Look at each effect being used (and how). There is a lot to be learned right there.
Then analyze how it can be improved.
Take notes.
I like Ozone. Nice tool. Awful handy to have a collection of effects of that quality that work in one FX slot (grin).
You can have all that and three empty slots for meter, analyzer ... whatever.

Yah, thanks guys.... that's why I recommended Ozone 5 for him on the other thread.

I just didn't have the motivation to explain and try and convince, and probably couldn't have done it as well as you and Bob did.

Eddie,

You can dl the FREE guide "Mastering With Ozone" from the Izotope site.
There's a lot of helpful info in it covering the different effects and how to use them.

Man, I am LOVIN' all this awesome input!!! Once this annual show is out of the way this weekend I can get back to concentrating on the home studio and put all of this to work.

Thanks to every one who has chipped in so far!

Mastering is a BIG topic on its own, and IMHO, using presets is a wrong approach to mastering audio, for several reasons. For example:

"The difficult part of mastering (and trying to create a guide like this) is that every effect, setting, and parameter is entirely dependent on the content of the mix, the genre, the desired result, etc. With this in mind, we don’t believe in products that fool you into thinking you can just select the “Hot Pop Master!” preset and you’re done."

That quote is taken from the "Mastering with Ozone" guide, an excellent introductory text to mastering, very useful and informative even if you don't own Ozone. You can download it for free here:

http://izotope.com/products/audio/ozone/support.asp

Eddie, I'm getting confused between the different threads. Somewhere you asked about why the rendering to audio was so low on some tracks and you were asking about using Gain Change. The answer to that is tricky and totally dependent on the details of your project. Basically, individual tracks need to be kept low so when they are mixed together you don't get massive clipping. Look at that Wikipedia article again and see the picture on the right named Optimum Mix Levels For Mastering. If every track in an 8 track project was recorded at -3db and you combined them together, what would you have? A complete distorted mess. But if it were only 2 tracks maybe not so bad. Why don't you hear that when you playback your project in RB? Because of automatic limiting in the software. But, and this is a big but, you as a true Audio Engineer don't want that. You don't want the software doing that for you. If you were to disable that then you would easily hear the problem. The question then becomes what's the default setting for RB rendering or generating audio tracks? How low is it? I have no idea and really neither do the developers because they have no idea how many tracks will be in anyone's individual project. It could be 1 track up to 48. If you're doing a real, traditional mix then you don't apply destructive Gain Change to a low track, you would lower all the other tracks because without the automatic limiting in effect you're going to clip.

I have the same problem with my new Kurzweil PC3 keyboard. Some patches are much louder than others and it's a problem on a gig. According to the Kurz guru's it's the exact same principle I just described. This keyboard has such a range of available layers of different sounds (16) plus twice the DSP power of the older unit they had to keep certain programs at a very low level to allow for users fattening them up and increasing the output to the point of saturation and distortion. This means users have to figure out exactly what they need for a gig, make all the appropriate adjustments and copy their setups into the user bank. Some people on that forum are not pro's and don't understand that concept and are yelling about why doesn't my Roland, Korg whatever have that problem? The answer is those other keyboards are nowhere near as deep as my Kurz is plus it has always been marketed to true pro's and pro's need and understand that capability. Sort of like someone who has a new Porsche and about dies when he's told a standard brake job costs $2,500 or so because the rotors have to be replaced every pad change. Why? Because the car is capable of going 175mph so therefore the brakes have to handle that. Never mind the average doctor in Beverly Hills will never, ever use that capability. It's there so that's it. Can't have the brakes die coming down the pass at 150, bad publicity and all that.

Even though I have a basic understanding of mixing and mastering do I actually use it? No. The basic defaults inside these programs produce results that are "good enough" and that's true of probably 95% of users. Others really do their Audio Engineering thing properly so since you asked...

Bob

Bob, that other thread was a little different. And in fact there were 2 threads.

When I finish a recording song and mixing it to where it is good in my ears in RB, I then export all tracks to wav. Then if I import those wav files into Sonar, the levels in Sonar, withough touching the volume on my interface that is feeding my powered monitors, is so low I can barely hear it. That was where someone suggected immediately making an adjustment in Sonar before even hitting the transport.

The other issue is still in RB. If I have a song set up, let's say on 10 tracks, and I decide "I want to add a rhythm piano part", I select the 11th track, click the menu option to generate a Real Track, find my instrument, listen to the audition of it, like that particular part, and hit "ok, generate for me", when that track is done generating it is sometimes SO SOFT that I need to boost it 8-10db just so it matches up to the existing tracks. Ar ethey supposed to generate at a level where they fit the existing tracks? Do I have control over how hot a new track is generated? Someone once said it is based on where the master volume slider is. I never work with the master volume slider completely to the right. It is usually at what would be about 80%. I do that to leave some headroom. When I just start to see some red clipping, that's where I stop. Just like mixing a band. Turn the gain stage until you start to see it go into the red with the channel volume off, THEN use the volume faders to blend the band.

And be ready to pounce on anybody who dares change their stage level and goes red on me.

I figgered it had to do with the style being used. How that part might generate to fit the mix of that style. Never thought of the master slider being part of it, but I doubt it's the final out slider. Maybe the 'All' Slider toward right of mixer.(?)

Or not.
I just gain change and go if I need to. There is a trim dial that may adjust it too.
Click the audio FX area for a track and look in the pop up window bottom right.

Like I mentioned, I can talk about how the pro's do it but I don't bother with that myself. Doing a quick gain change works for me too but, I suspect that RB is looking at your other 10 tracks, sees the levels and generates a new part low because if your mixing them the correct way you would be lowering all the other tracks not raising the new one.

As to Sonar I suspect it's the same thing. You import your 10 wav files that you created in RB. Say they're all around -3db. Sonar is probably lowering them all for the same reason I just talked about. They have to be low otherwise when you combine them for the mix the final stereo track will be way too hot and clip unless you have it set up to do automatic limiting which a good mixing engineer would not want because automatic limiting just took him out of the equation. Hitting the automatic Merge to Stereo Wav button turns a pro engineer into one of us. A home hobbyist who just uses the defaults and calls it good.

I used to do a lot of live band recording but not lately. I would take those tracks from my Akai DPS 16 HD recorder and import them into Adobe Audition. Audition does not have automatic leveling for a mixdown or if it does I never set it up for that. If I tried to mix a bunch of tracks that were too hot the result would be solid black in the waveform window and the sound would be total distortion. Why? Because Audition is emulating a real studio and if you try that there, that's the result you get. I would then have to go and lower the levels of all the tracks so I would have something useful to work with for my final mix. I might have put some effects on a few tracks but still needed to leave headroom for EQ in the final mix. By that I mean if the initial trial mix gave me a very hot mix up to -1 or 2db that's not enough headroom to apply any EQ. The slightest bit of EQ would push the whole mix over the top. This has nothing to do with me trying to get the hottest original raw recording I could using my Akai. The source recording always is as hot as I can get it without clipping to get the best S/N ratio but when it goes into Audition the levels get reduced for mixing.

Bob

We're talking realtracks volume, right? A gain change on those is usually pretty painless.
I will say I've heard some interesting things after doing so a few times.
Drummer grunting, etc.
But they are pretty clean.

Real Tracks, yes. And yes, it is no big deal to select all and do a gain change. It just seems that we should be able to set "make all my additional tracks at X level".

As I have walked down this path I am seeing that my reference monitors are WAY too bass heavy, even with the cut switch enabled. And that gets me back to the studio itself, if my room is giving my ears an inaccurate reading.

I mix something down that sounds JUST RIGHT in my room monitors. I merge it down and make it into an MP3 and test it on a computer, on the main stereo, and in my car, and the bass is always way too loud. So I guess I just need to stay mindful of the end result more than the room and mixing with very low bass levels.

Now someone mentioned Nectar. I grabbed the 15 day trial. Does it work with RB or do I need to use it as a plug in with Sonar only? I tried it with RB and it crashed me, though I may have been using it wrong. Unfamiliar territory for me.

The pro who does my final mastering, said to me a couple of years ago, ( I had asked him about 'upgradng my monitors)........"Joe, without being cheeky! Think about 'upgrading your ears!"

Which I understood to mean..... learning how to "properly use" the equipment I have, & compensate where necessary. Like Eddie........for me a steep learning curve! I was so pleased when I took my latest CD to him, he told me, "Not much to do this time Joe, your best yet!"

Joe G.

Eddie and Rharv, the reason I'm explaining this in detail is Eddie wants to know how this works in a real studio. For us just messing around at home, letting the software automatically limit your mix sounds fine in a lot of cases. Like anything else it depends on what you're doing, the type of music it is and how critical you want to be.

Example, most strong rock oriented stuff is all mixed hot anyway but a lot of songs still might have an interlude section in the middle that is much quieter than the rest of the tune. When you just listen to your song in project mode, that is in RB's main track window, each track is still individual and if each one is below the clip threshold it sounds good. But, mix that using automatic limiting and the whole thing winds up at the same level and you've just lost those dynamics that you created for the interlude section. I believe it's this that causes people to say things like why doesn't the render sound the same as the raw project?

Without the automatic limiting you're telling the software to do it the old fashioned way. Combine the tracks and after combining that audio data together the levels will be what they actually are and then it's usually way too hot so the whole mix is clipping. That's when you realize you have to go back and lower each track and remix it so you can preserve those dynamics.

Using gain change to raise the level of one track is fine as far as listening to it inside RB is concerned but it will cause you to lose dynamic range in the final mix. If it's a hot song anyway with little dynamics in it, then no problem let the software do it, it's a lot less work on your part.

Bob

How did you equate Gain Change with Limiting?

Also using Gain Change doesn't make your perceived mix in RB sound different when merged down.
A couple things you said there confuse me ..

Ok, I'm not an educated, experienced Audio Engineer and I'm probably not explaining this right. When I said Limiting earlier, I'm referring to automatic limiting the software is doing. Think about it, you recorded 4, 6 or whatever tracks at the absolute max level you could get away with without clipping any individual track. How would you mix those tracks down and not wind up with a distorted clipped mess? Just lowering the sliders in RB's mixer isn't good enough. I can't explain it exactly but just that alone won't do it. When you select "Merge to Stereo Wave" RB is doing some automatic limiting to avoid clipping. The best way and the pro way is to not record your individual tracks so hot. They have to be recorded down at -12 or so. It's all about leaving some headroom throughout the whole recording process from tracking individual parts, to the mixdown to the final master. Here's a pretty good article about it:

http://karma-lab.wikidot.com/misc:setting-gain-levels-for-recording-and-mixing

Note what he says here

The executive summary

If you want to avoid reading a lot of details, the short story is this:
•Record at 24-bits. 16-bits is antiquated now and doesn't give you enough dynamic range and headroom to work with.
•The analog input signal from the M3 into your audio interface should be loud enough to start causing red flashes on the peaks (in your audio interface's input meters), and then you should drop the volume just enough to make the red flashes disappear.
•The digital signal from your audio interface into your sequencer's individual recording tracks should be set at -12 to -18dB on the input meters for each track being recorded. Don't go near -6dB and certainly don't get anywhere near 0dB when tracking.
•The digital signal in your sequencer's Master meter should peak at -6dB when creating your mixdown. If you go over -6dB peak on the Master, lower the volume of all your tracks slightly, and equally, to reduce the summed volume of all tracks as needed to keep the Master peaking at -6dB
•Render your mixdown to a WAV or AIFF file still at -6dB because normalization and limiting from within all major sequencers is pretty weak.

I take this to mean if we're trying to do this the correct "pro" way we don't want RB or any other sequencing software to handle any of this automatically for us. This is why I use Audition, I have complete control over the mix down. RB may allow for that too, I just never use it for that.

Again, this is the proper educated way to do this. I suspect most people on this forum including me think RB's automating rendering down to wav sounds good enough and that's the end of it. Eddie has talked about wanting to know how this is all really supposed to work and is talking about going to school about it.

From the quote above:

The digital signal from your audio interface into your sequencer's individual recording tracks should be set at -12 to -18dB on the input meters for each track being recorded. Don't go near -6dB and certainly don't get anywhere near 0dB when tracking.

Here's where my comparison to Gain Change comes in. I've seen the result of generating a RD track in RB. The wave form is so small you can hardly see it until you blow up the track. I'll bet it's down around -15db or so and that's exactly what this guy is talking about. To put it back to Eddie's question, if the RD track is that low, it's because RB did it that way on purpose so no, you shouldn't apply Gain Change to raise the RD track up, you should be lowering all the rest of the tracks down. Iow, those other tracks were recorded/rendered, whatever too hot, not that the newly generated RD track is too low. RB is doing it right.

Hope this makes some sense. If it doesn't then just hit the render button and be done with it.

Bob

I typically gain change the track to just before it would clip. I can always mix DOWN in the final mix but I can't go past as loud as it will ever get. That's like people who prefer hot to cold. I prefer cold. I can always wear another sweatshirt. When it's hot, there's only so much you can take off and not get arrested.

If my tracks are on the hot side, I can make them softer, but if they are on the cold side, I can only boost so much until I hit 100% and then there is no more boost to boost.

This is where you get into amplitude vs volume. The gain stage (amplitude) has to be strong so you DON'T need to push the track during mix (volume).

That being said, I gain change 2db at a time, and never to where it flatlines against the top and bottom of the track.

Many years ago I posted something similar on the old Syntrillium Forums. Here it is.
Something you are missing, except Bob at this point, is that "0" on a digital meter is IT."0" on a pro console still has 26dB of headroom left before clipping.Thus the theory of mixing at around -12bB or less.

A quick add to this post. (busy day ahead)

I was confused about levels as well when I started learning about this and I'm am certainly not an expert but Bob and John are right, you have to mix lower than the old analog console way.

Eddie, go to the website I put up in the other thread and have a read on the "Mastering & Gain Staging" section. 0db on the analog scale is equivalent to -18db digital. If you are working with the analog levels on a digital system, then all you levels will be too hot.

On the videos I bought,the mixing engineer,(a pro with 30 years experience,the audio engineer at Columbia Records, etc, etc) had all his levels as he was mixing low. Around -18db, -12db even lower in some cases to allow for further processing. It was an eye-opener for me.

This is the link for the article on mastering & gain staging (not to the recording school where I get my videos from)

http://www.independentrecording.net/irn/resources/index.php

If you are interested in the videos I use, here is the link (I'm not trying to push this on you, it's just there if you want to have a look)

http://recordingschool.biz/homerecording/index.php

Bowing out, too much 'splaining required to even address.
Analog and digital are different animals.

You're right Rharv, they are completely different. It took me much reading,study and web articles/info including the threads on this great forum to final realize it.

Still along way to go.

"Cleanup in aisle 9, stat!"

Lots of terminology is starting to be thrown around here. One of the items that is likely being misinterpreted is the term introduced by jazzmammal as 'automatic limiting' when saving the 2 track mix to a stereo .wav.

In most DAW software, when one does a saving of the final output bux to a 2 track file, there is an option for 'normalizing' or 'normalize' or 'prevent clipping' or something along those lines.

What jazzmammal is describing is a normalizing process. What normalizing will do is look to see where the absolute maximum level would be in the samples of the mixdown, and set that so that particular sample's digital output value is 100% of the max possible digital output - coded to .wav - and then scale every other sample in the mixdown by the same scale factor.

It can be a greater than 1 multiplication, or less than one (as is being discussed in this thread) multiplication.

This is entirely different than automatic 'limiting', which looks to see if any part of the mix-down file would clip and reduces only those parts which would clip, leaving the others unchanged.

Does everyone understand the difference? Normalizing scales every single sample in the .wav file by the same amount, in order to achieve the end result that the absolute loudest sample in the mixdown, gets the max that the .wav file allows, where limiting only changes samples which exceed some threshold value, leaving other samples unchanged.

Are we clear? If not, please post as such and I'll try to explain in a different way.

Scott just answered that. As I was writing these posts something didn't sound right and this is it, it's not limiting, it's normalizing. Thanks Scott. Just switch those two words. My main point hasn't changed and now when I search mixing articles using the keywoard "normalizing" I'm getting lots of hits all saying versions of this:

Normalizing
Do not use any kind of normalizing on your mixdown.
 Normalizing raises the signal level in an unnecessary fashion, and it will change the
amount of headroom left in the mix. The final volume level of the mix will be optimized by
me.

That's what I was missing. I'm not going to bother posting the full link, they all say the same thing. Pro's don't normalize in most cases so my overall point is correct.

Bob

Pros are often feeding signal to other hardware equipment. In this scenario headroom is important.
If you are staying in the digital realm, you are a little safer going up a bit on the levels.
Besides headroom there is signal-to-noise. Many hobbyists focus on this aspect. Numerous tracks add noise and if you record a strong signal you can lower signal-to-noise, but at the cost of headroom.
I'd worry more about whether I was using line levels that are consumer (-10) or pro (+4). That may also come into play for the reason 'pros' say -12; they are likely using signals with more energy than hobbyists are to begin with.
Here, maybe this will help-

http://www.independentrecording.net/irn/columns/jwal/index.php?id=79

As analogue levels and digital levels are not the same, I thought this article on recording levels might be of interest.

http://www.massivemastering.com/blog/index_files/Proper_Audio_Recording_Levels.php

There seems (at least to me), to be two different things being discussed in this thread:the recording levels (how hot to record the tracks) and the slider levels on the mixer.

I'm trying to learn how to do this stuff properly and was getting confused so I thought I'd post the link and see what everyone else thought.

Pretty good article, but of course he doesn't say why it is harder to make hot tracks loud(er) while it is easier with track with lots of headroom.

My take: You use some kind of compression/limiter to make things loud at the end. If everything is too hot, then the "brick wall" limiter does all the work and everything gets squashed and you lose dynamics and clarity. If you have headroom, then the compressor does most of the work and the limiter only kicks in at a few transient peaks.

Well, I think that is what happens. But I don't really know for sure. I mean you could just slide the faders down and "create headroom" -- but I don't think that works as well. Ya know, I think I have hit the 0dB of my knowledge on this subject.

Quote:

Pretty good article, but of course he doesn't say why it is harder to make hot tracks loud(er) while it is easier with track with lots of headroom.

My take: You use some kind of compression/limiter to make things loud at the end. If everything is too hot, then the "brick wall" limiter does all the work and everything gets squashed and you lose dynamics and clarity. If you have headroom, then the compressor does most of the work and the limiter only kicks in at a few transient peaks.

Well, I think that is what happens. But I don't really know for sure. I mean you could just slide the faders down and "create headroom" -- but I don't think that works as well. Ya know, I think I have hit the 0dB of my knowledge on this subject.

I DON'T user compression/limiter until the very last step if at all.

Think about this. If you have a quiet piano track, once you pull that slider all the way up or right (depending on your screen view) you are done. That's as loud as you can go. That's where I would start boosting 2db at a time to see if that created enough headroom.

And that is why I prefer to record hot. I can slide that puppy down to nothing if I want to, but I can only go so high before there is no more "high" to go to.

Quote:

... Think about this. If you have a quiet piano track, once you pull that slider all the way up or right (depending on your screen view) you are done. That's as loud as you can go. ...

In sonar 8.x you have four ways to crank the gain on a single track (and there might be a couple others!). Of course clipping is your enemy with these steps.
1.) push the fader all the way (maximum +6db, it looks like)
2.) Crank the trim control (max +18db)
3.) Just add gain to the track. In Sonar 8x, you highlight that track, go to process --> audio and add max of +18db as many times as you want.
4.) You can use compression with make-up gain to make it louder.

It looks like you are just using the fader.

I am also not in Sonar 8.x. I work in Real Band 95% of the time. I go to Sonar when I want to play with an effect that RB doesn't like (Izotope, for example, crashes my RB but works fine with Sonar).

The original question, 3 pages ago, was asking what people call mastering, and why some of my songs sound like the top half of the spectrum is washed out. Early on I got like 4 different thoughts as to what the definition os mastering is. The resulting conversation has been awesome on many levels, too. Links and links from those links and all good reading that I copied and printed for highlighting.

This is why I love this place!!!

Just checked -- In Realband you seem to have everything that Sonar does.

1.) The fader is the same.
2.) The trim is found by clicking the fx icon on each track (next to the Solo button). The trim control is in the lower right of the fx window.
3.) To add gain to a track: Right-click on a track and "Select Whole Track". Then Edit menu --> Audio Effects --> Gain Change
4.) And I think you know how to find the compressor.

Eddie, try using DXi version of Ozone. Works here. If that's what you have running that crashed RB then try the VST version. I've heard of both ways having success, so worth trying 'the other one'.

Oh, and if you find yourself boosting lots of tracks, consider pulling the other tracks down instead.
Then worry about final levels (mastering) later. After the mix is done.

Hi Kevin,

Referring to your quote:

"My take: You use some kind of compression/limiter to make things loud at the end. If everything is too hot, then the "brick wall" limiter does all the work and everything gets squashed and you lose dynamics and clarity."

Your knowledge isn't 0db at all, from eveything I have read and studied, you have it right.

If the wav is already filling the entire window, you have no more room to add effects and the mastering engineers are not going to like it. They have nothing to work they magic on. Pulling faders up or down doesn't alter the wav itself, it doesn't give you more headroom (as referred to in the article),it only the changes the volume.

At least that's my understanding of it all.

Quote:

Ya know, I think I have hit the 0dB of my knowledge on this subject.

Now THIS made my day!!!!

Quote:

If the wav is already filling the entire window, you have no more room to add effects and the mastering engineers are not going to like it.

I hate to keep disagreeing with people, but this isn't strictly true. (Or even at all.)

Suppose the mastering engineer is sent a wav file with very little dynamic range and the recording normalised (maxed out). Because this is digital, the first thing he does is make a safety copy to work on. He then performs a gain change down a few db to create headroom without effecting the original mix in any other way. He is then free to work on the track in whatever way he sees fit, including increasing the dynamic range by use of an expander. When he's finished, he can bring the whole mix back up to max. Job done.

ROG.

When I listen to your songs Eddie it sounds like they have a poor job done on the EQ. That may be the problem with sounding washed out on the high end.

Yep. Exactly what I am trying to learn. Plus some of the earlier songs were recorded through an outboard mixer with outboard reverb and EQ and I need to do them all again without all that stuff. Once those destructive effects were recorded, there is no going back, so I need to sing some of them again. First I have to hope I still have the original SEQ files.... it'll come. With all the great input from the userbase here, it'll come. The later stuff is MUCH better as I have learned more.

Quote:

Fixing a bad mix, while in the realm of a mastering engineer, is probably the last thing he would want to do to try and make it sound as good as possible.

Kevin, I really hate doing this, but not in my experience.

Sometimes you can be given an excellent product to work on - something which is well recorded and mixed with a nice leveling amp applied at the end. In a case like this, there's not a lot of work, if anything, to do, yet people still expect you to make it sound better.

On the other hand, you can take someone's huge disaster and very quickly make a noticeable difference. Easy money and less stress.

ROG.

I'm not trying to drag this thread out but I've been ill for the last few days and wasn't able to check back to see how it was going. I'm finely able to look today.

Thanks ROG for adding the gain reduction info in your reply to my last post, I completely overlooked that.

I know in the analogue realm it is good to record tracks as hot as possible and that a bit of distortion or tape saturation is also a good thing.

Everything I read: articles on various websites, videos and even in the BIAB manual all recommend recording at lower levels (around -12) when working in the digital realm.

I know there are quite a few experienced mixing/mastering engineers here on the forum, so my question in the pursuit of better knowledge is this,

Instead of trying to record tracks as hot as possible in the digital realm and then having to do a gain reduction so compression and other processing can be done...wouldn't it save time and an unnecessary step to just record at the recommended levels? You could do your processing, get your mix as you want it and then bring up the gain in the mastering stage.

Everyone has their own ideas and I guess in the end how it sounds to your ears rather than sticking to levels is the deciding factor but some thoughts from those experienced working digitally would be great.

To help understand this -12 thing I need you to do this. If you have a pro level mixer look at the leds. You'll see that the "clip" led is at +12.You say "what difference does that make?".The reason is we didn't go from analog tape to digital computers and CDs. There was another step. We went from analog tape to digital tape machines BUT & this is a BIG but. All of the rest of the gear was still analog.
Compression is a good tool but often misused.It's real job was to take recored material with a large dynamic range and "squeeze"it so that it could be pressed onto a record that only has a dynamic range of around 54dB.

Quote:

To help understand this -12 thing I need you to do this. If you have a pro level mixer look at the leds. You'll see that the "clip" led is at +12.You say "what difference does that make?".The reason is we didn't go from analog tape to digital computers and CDs. There was another step. We went from analog tape to digital tape machines BUT & this is a BIG but. All of the rest of the gear was still analog.
Compression is a good tool but often misused.It's real job was to take recored material with a large dynamic range and "squeeze"it so that it could be pressed onto a record that only has a dynamic range of around 54dB.

Agree and disagree here...

Some of us use compression because we stink at playing our instruments. Like when I play bass for church, I compress my incoming signal into the amp like the dickens because I'm still new at it and it makes the sound engineer's job easier not having to deal with inconsistent levels coming from my amp simulator. He can control the amount of content I contribute to the overall mix with much more ease than if I fed him a tentative player's output.

Compression has all kinds of uses both in helping to record, taming existing individual tracks for anomalies, and on final mix-down signals. Used in many different ways - one of them which was to prepare tracks for recording to tape and eventually for radio airplay and it's own signal to noise ratio challenges.

Some even use it as an effect intentionally - not to transparently hide overzealous transients, or to pull up too quiet sections, but to make use of the distortion that results intentionally. This has been done for years on drum signals in rock and other type of recordings and even live use. I let the kick and snare and overheads saturate a bit through our PreSonus ACP88 at church.

-Scott

yjoh,
You and Jazzmammal asked "why" pros recommend recording at -12. Well here you go. Ethan is a well known pro -
"Folks who have used analog tape recorders but are new to digital recording tend to set the record levels too high. With open reel tape and cassettes, it's important to record as hot as possible to overcome tape hiss. But analog tape is more forgiving of high levels than digital systems. Analog tape distortion rises gradually as the signal level increases and becomes objectionable only when the level gets very high. Digital recorders, on the other hand, are extremely clean right up to the point of gross distortion. Therefore, I recommend aiming for an average record level around -12 dB. or even lower to reduce the chance of distortion ruining a good performance."

Source -
http://www.ethanwiner.com/mixer2daw.html

Quote:

this is a BIG but.

As the one who started the thread, I respectfully request that you leave my big butt out of this...

Thank you everyone for answering, it is much appreciated.

John,

I only have an old Yamaha 4 track but I understand the +12 levels you are talking about and any recording we did was done hot as possible. Even though it is still analogue gear, my understanding is, (and I'm willing to be wrong) once it is recorded into the computer, it is converted to the digital realm and the lower levels on the meters in the computer software program then apply. This is why I felt there was 2 different discussions taking place and it was getting all muddled up.

Scott,

I agree, from watching my compression video I have been quite amazed at how much more it can enliven a track if the user knows what they're doing (which I'm definitely don't as yet, but getting better)

Rharv,

Thanks for answering, that was my understanding exactly. I don't feel as though I'm going mad so much now. Thank you!
I've read Ethan Winter's articles before and have his site bookmarked. He is very good and makes a lot of sense.

Eddie,

Don't worry I'll leave your but out of it, but (no pun intended) thanks for starting this thread, it has been a learning experience.

Thanks again everyone, I feel I can move ahead now and don't feel like such a beginner. Whew!!!

I am reading the manual from Izotope's Ozone, which I really like, by the way, and on page 12, it states

"The bass will typically be under-represented on small studio monitors".

This reinforces one point made here as to why my bass is always too loud. I make the bass sound good on those monitors and it is too hot for anywhere BUT those monitors. Point noted and filed away in my aging memory bank.

Now to practice with their compression tools.

It's funny how we get locked into a routine and that routine becomes akin to a horse wearing blinders. There are things in RB and Sonar that I never even noticed because I never had a reason to look for them. Someone mentioned grouping tracks so they can be adjusted at the same rate. I don't know how much I'd do that, but the point is I didn't even know I COULD create subgroups like that. So many threads have gone past that refer to some specific part of RB and I said "Real Band can DO that?" Need to experiment more and not get so focused on finishing a song in exactly 57 minutes (if you get what I mean). I call it "done" too soon.

Quote:

Therefore, I recommend aiming for an average record level around -12 dB. or even lower to reduce the chance of distortion ruining a good performance."

Here's the point that is getting missed and is also where I was wrong in my comments about riding the faders.

Since I have been corrected about my misstatement concerning you can't get a good mix simply by riding the faders, that still begs the question: If there's no problem with riding the faders then what difference does it make if your tracks are all at -2db or -12db as long as there's no distortion? There is no automatic normalizing, limiting whatever going on. The faders themselves simply act as a gain changer so just ride the faders down until you're not distorting the output bus and you're good to go. Eddie summed it up himself when he said so I can record as hot as I want as long as nothing is clipping and just mix using the faders. Right? Not exactly.

What about dynamic range as Silvertones mentioned? You can create some dynamics between tracks just using the faders but what about dynamics within any given track by itself? If the whole guitar track for example is up at -2 and that's where it is for the whole song maybe that's not what you want. A bridge, chorus, solo sections can all be at different levels during the song and most times they should be. Very few musical tracks of any song are going to be recorded at the same level throughout. If a quiet section is at -10 and a screaming solo is at -1 then how can the overall track be at -2 unless it's been compressed or limited already? When you apply Gain Change in RB almost always automatic limiting is checked so it looks like simply applying Gain Change is fine. Uncheck that Limiting box though and then apply some Gain and see what happens. Most raw recorded audio has all kinds of peaks in it whether it's you strumming a guitar or singing. You soon realize you can't apply much Gain Change at all without checking that Limit box without clipping.

In the other thread I posted my discovery about how RB generates a Real Drum track. It came in at -24 on the meter but blow it up inside the audio edit window and peaks were at 0db and RB is not going to generate (record) an audio drum track using automatic limiting or compression so it has to keep the peaks under 0db. We all know virtually every audio engineer compresses the drums but the question is when will they do that? There's many different answers to that I'm not going to go into now but my point is if you want to hear what a virgin drum track sounds like before any manipulation then it goes back to my original point, you don't do a gain change or anything else to that track, you have to instead turn up your studio monitoring system and using the faders, bring down all the other tracks so you can effectively mix your new drum track. Then you dscide what you want to do with it. This is the same principle for all your other instruments and vocals. Unless you're doing some tune that is going to be the same exact boring level all the way through, then that is the answer as to why you should record and keep your individual tracks down around -12 or less. To maintain dynamic range at least initially. You may later decide to reduce that range by using compression and limiting but you don't want to paint youself into a corner by having your initial tracks too hot. Then you're doing what ROG referred to, applying an expander after the fact to try to recreate some dynamics and then mix it in order to look like a hero to a new client who did it wrong in the first place.

I think this makes sense now or did I screw this up again?

Bob

And then on the 8th day He created MultiBand Compression.If compressors have got you wait until you try these. They really are a MUST for Mastering.

Don't remember if this has been mentioned, but besides the faders there is also a trim adjustment in RB for audio tracks, it's in the RealtimeFX window and many don't even notice it. Not sure if that has any limiting on it or not, but it should be easy to tell. I suspect it doesn't.
The more you know!

Okay, so as the discussion continues, let me present this.

4 rhythm tracks

drums
bass
guitar
piano

3 vocal tracks

lead
bgv1
bgv2

2 solo tracks

guitar
sax

All of the above are recorded at this mystical magic number of -6db.

I start playing back. (I am not even looking at the index marks on the channel faders. Just doing things by ear.) Bass is not loud enough. I push up the fader a couple of lines. (Here I did not raise the recorded db level of the bass track, it is still -6db, but I DID raise the db level as far as the VU meter sees it.) So now I have the bass sounding nice and full, but man, now that scratching guitar track seems kind of week, so I slide that fader up a line. Nice. Bass thumps in harmony with the kick, the scratch guitar is playing off the hat real nice, but man, this piano is really lost now. Let me move that up a notch.

So here I have had to slide 3 faders up to give the thing some balls. Why did I not just record those three tracks hotter, at maybe -3db? In the logic that has been prevalent in this thread, recording low and boosting is better than recording hot and cutting. However, let's use that quick example above to say "Maybe that piano is up as high as it can go and there is no more fader left unless I take a saw and cut a groove in the mixer." How did recording soft satisfy me there? And if the answer is "do a gain change", how is recording at -6db and then doing a gain change of +3db not the same as recording at -3?

Why don't I want the wave form to fill as much as the track limits as possible as long as I don't hit the top or bottom barriers?

A couple reasons to record a little lower:
"The performer may play/sing louder than normal and actually clip the track. If you've ever heard digital clipping, you know this is a bad thing.
You want to leave room (called headroom) to add effects..."

source: http://audiominds.com/recording.html

Quote:

You want to leave room to add effects...

I don't grasp why adding effects necessarily increases the decibel level. You are augmenting what is there, not adding anything more.

I also don't get why taking a wave form that is recorded at -6db and doing a gain change of +3 db does or does not equal a wave form recorded at -3db. If I am standing on a scale, and you hand me a 5 pound weight, the scale will show to be 5 pounds heavier. Or if I stand on the scale and eat hot dogs non stop until I consume enough that I weigh 5 pounds more, either way the result is that the scale shows 5 more pounds than when I started.

I mention that scenario because many times while playing with mixdown I have watched the VU meters register "X", and then I go to a track here or there and do a 2db boost, and the VU meters never change, so I think I need to get clear in my mind the db level or each track vs how the summed total of those tracks affect the overall db level of the output.

I am really SO new to this side of the mixer....

Hi Eddie,

I haven't time to do a big post, my long working day starts very soon, but read what Rharv said. You have to leave room to add effects and processing.

From my learning as it progresses, you get your gain from using compression etc. It all adds up as gain.

Your thinking is ok I guess if you just want to balance (I'm using this instead of "mix") the raw recorded tracks, but if you want to "mix" which entails so much more than just getting the balance between the tracks nice, you have to leave room for your processing.

Remember, you are where I was a couple of years ago. My humble suggestion is to start learning about compression and all the other processing that can be done in the mixing stages.

All the best, go to go now!

Compression ADDS gain? By definition, doesn't "compression" mean "to compact 'more' into 'less' and save space"?

Compression can add gain when there is a 'make up gain' control involved.

Eddie, adding effects can increase levels if there is an output gain control on the effect like all of the good old Kjærus classic series or at least a good number of them.

-Scott

Simple test Eddie. Take any audio track, highlight say 8 bars and open the audio edit window and look at the waveform. Don't just look at the VU meters because I already talked about how those meters are not as accurate as I thought they were and ROG explained that none of them are. They are designed to ignore transient peaks. That's why the RB generated RD part showed -24 on the meter but has -0 peaks in the waveform.

Inside the audio edit window with your 8 bars, apply some EQ by using the PG 10 band. Make a smiley face and add +3 to the highs and lows and look at the waveform again. You'll see that the overall gain of those 8 bars have gone up.

As to your overall question, I just answered it in my last post above. Apparantly correctly this time. If those tracks you recorded were all at the same level with no or very little dynamic range then yes, you're right just record them hotter. However, if they're all that level in the first place then you recorded them wrong as a player. You shouldn't be playing your parts like that with no dynamics. In the old show group the leader would have been all over your case because we were big on dynamics during the show. A song was not performed at the same level throughout. "Turn down that flippin rhythm guitar part during the verse, you're overpowering the vocals!" "And simplify that bass part too!" "You guys are playing like a bunch of school kids, just blasting away for the whole song!"

Iow Eddie, if you've recorded those parts with good dynamics then those tracks MUST be overall recorded lower or you lose that. If the quieter parts are recorded at -3 then where do you go for the loud solo's?

And turn up your monitors if you've run out of fader. Just crank em up. It's after you've achieved a good mix and bounced it down to a stereo two track that you then master it and increase the db up to where you want it.

Bob