Getting song Radio Ready

Discussion in 'Mastering' started by IcePhixia, Jan 14, 2011.

  1. IcePhixia

    IcePhixia Active Member

    I've been wondering for the last couple of months why my last song on radio didn't so that clear and powerful as in the studio.

    The I found out that quite a few people said that when you send/email or drop your song of for a radio station, that the overall RMS must not be over the -6db range...thus, don't compress or limit your song, but keep the song as is.

    I've attached a file where the overall range is say within this -6db limit. What should be the main focus to get this sort of song radio ready, since the song contains quite a few dips below -18db...?

    How do we keep the songs overall clear and punchy without going over the -6db limit and not using any compressor or limiters to get the soft parts up?
     

    Attached Files:

  2. Big K

    Big K Well-Known Member

    Most stations use their specific "tools" to squeeze the last drop of dynamic out of the songs to obtain what they call their special station sound. The optimal levels vary quite a bit. Here, in Munich we have to deal with 3 or 4 different ones. Often it depends on if the stations use digital or "analog" systems to cue their tracks.
    I found that giving them a hi-gain copy is the worst what you can do. It sound awful and is more quiet then a more conservativ approach to levels. It gets scaled down ( or rather squashed..) by their "Optimizers" a lot.
    The best advice would be to either to ask the station Techicians or have it worked over by a pro Mastering Studio.
     
  3. RemyRAD

    RemyRAD Well-Known Member

    You are slightly confused between the mixing process & " dialnorm" like specifications. This is where audio gets a little confusing where something can sound much louder at lower levels and sound much lower at higher levels. Telling you not to compress anything before it goes on the air because the station is already going to do that anyhow. That's sort of like saying you don't need to wear any underwear since no one can see it. So when you mix, you typically take certain instruments, vocals and all sorts of things and may subjectively submit them to some kind of dynamic range modifications. Once you put your genius stereo mix together, the big timers then send their genius mixes (that generally already sound loud, louder than most) To their big-time Mastering Engineers to have it even further optimized that might include some further dynamic range limiting processes. And then it gets to the radio station that will go ahead and still add their own, competitively based, dynamic range loudness enhancers! So, keeping your levels low, not using any compression, is called " MISERY LOVES COMPANY". So unfortunately, the cruel joke is on you.

    Now, the bottom line is, to be radio ready, it simply has to sound good, not be too low in level, not be too high in level. No distortion. No clipping. No noises before the music begins no noises after the music ends. And the reason why your cut didn't sound as loud as everything else did on the air is because you didn't use enough dynamic range modifiers & highly knowledgeable mixing technique. This is what comes with the included directions to Carnegie Hall, practice, practice, practice. So you should be proud of the fact that you produced a quality product that actually made airplay. Congratulations. So obviously your product is good enough to warrant airplay. And while I'm talking about the dynamic range modifiers which we also refer to as compression & limiting, these devices provide control over what is known as " apparent loudness" levels. This means that by manipulating attack and especially release times, sources can be made to sound louder & more aggressive or slightly smoother & less aggressive. Loudness wars on the radio and other sources, are generally based upon very fast attack times, which prevents any peaks from getting through. And very fast release times. It's the release times it will make things sound most aggressive. The faster, the more aggressive and louder it will appear. Now this sounds like the magic be-all end-all but it is actually quicksand. It is already being abused by the radio station. Whatever buddy is trying to tell you is that you shouldn't abuse your material as aggressively as the radio station is going to. So use plenty of compression & limiting to try to retain as much naturalism as you want. Then if you want an increase and it's apparent progress of loudness levels, you adjust your release times to create a more upfront, faster, louder sound. This may actually require you to lower the vocal in the mix, to maintain the proper balance. Again this is a lesson on theory & practice and not necessarily a do or don't lesson.

    Of course on the other hand, if your waveform looks like a brick on your screen, it might be too fast, too dense, too aggressive which when adding with the radio station does can make your product unlistenable & unbearable. Because the faster you release the more distortion you also enhance. And while some distortion can be musical most electric & electronic distortion, not created on purpose, isn't pretty to listen to on anything.

    Now you know the ups and downs
    Mx. Remy Ann David
     
  4. IcePhixia

    IcePhixia Active Member

    Hi RemyRAD. Thanks for the major detail post regarding this "radio ready story". This is probably the best advice regarding compressors and limiters I've seen in years and you've helped me with this radio ready question.

    Just to clarify the radio song story - the above wave file picture was before we really compress the living crap out of it to end up to look like a brick wall sound file, before we dropped it of at the radio station. So I thought it would be as a staring point for a question like this...using our big dynamic range song

    After reading your post, it makes a hell of a lot of sense that we should really focus our attention on the mixing side of things, -and apply proper compression and limiting on each instrument to get them to "sit in the mix". It's up to the mixing engineer to really get the song to a good starting point

    I've downloaded Chris Lord Alge's CLA compressor (demo) and also been reading up quite a bit on this mixing and compress process. Also seen CLA in his LA studio mixing a song using the wave plugins and the I realized another flow we've been doing as mixing engineers. To pro people really do what he calls "riding the faders", which is like painting almost. You can have a big red sun with shocking red colors and thin blue sky and dull green grass with a small home in the background, but you will find a few people buying a painting like that or even look at it. You need to first sit down and work out what this painting is all about, work with the colors and depth to bring out a master peace. Same for music and those big SSL mixing console. Find the thing that make you listen to the song...is it the beat, the vocals...and work with that while changing the overall volume and dynamics of each instrument. Thus, your dynamics in this case is how strong or weak your color is in the painting :), while the volume is the size of the overall image. How much space is the sun taking up on your paper versus the house. And is the sun the main focus point of your art peace.

    SO, getting back to your good post. Without compressing your instruments as per track (vocals, drums, bass...) you will end up with a dynamic range song that is huge. The gaps between the hi(hard) and low(soft) will be to wide and even if you apply compression on the final mix-down, the song will sound distorted and bad.

    Using the attack times of something like a 1176black Compressor on bass-guitar to keep the bass guitar nice and tight (not to much dynamic diff) you can help with your dynamic range difference and if you want it to sound a bit more aggressively you get ride in on the release time and bring that down. BUT never do that on it's own. Keep the whole song playing while you apply this per track.

    So, If I understand your post:
    1) Your song must be loud and compressed already before sending it to radio
    2) Mixing is the key to good compressed overall sound
    3) Apply compression/limiting per track - using attack to bring the peaks down, release to get more power/aggressive sound
    4) Radio ready wars on loud sounds = use faster attack times
    5) Cheap compressors will distort if you use fast attack times - so get a GOOOD compressor/limiter!
    6) The wavefile as seen in post above is going to sound bad on radio - due to huge dynamic range...need to go back to the mix and start compression

    Now my final question regarding this radio ready song is:
    1) Lets assume the mixing engineer used proper compression
    2) Used proper fader skills to keep the song in motion
    3) Applied proper EQ on various tracks to keep the muddiness out
    4) He renders his final track to a stereo 48K, 24bit track
    5) The overall dynamics of the stereo track looks good

    Question 1:
    Should we then take this track and apply a final limiter on it to get it really loud and proud ...say (ceiling of -0.3dBFS) or is this where we leave the limiter part out.

    Question 2:
    If we leave the limiter off, at what level should we be at for this track ... -6 dBFS for a radio track? or should we look at the overall avg level?
     
  5. RemyRAD

    RemyRAD Well-Known Member

    You're getting pretty close that you're understanding better. Unfortunately, I've made some generalizations that you have misinterpreted. Not every instrument always has to be dynamically modified. Cheap compressors such as the Alesis 3630 or whatever the heck the model number is, can be just as effective as a Universal Audio 1176 if you understand its limitations. Fast attack times isn't one of it's limitations but it's operational limitations are more severe than a 1176, for instance.

    In answer to your question 1: For instance, applying a final stereo bus limiter is a matter of personal taste & preference. If you don't do it? Your mastering engineer may? And generally they do. If you don't have a mastering engineer, which is a luxury, you may establish that your mix is good but might feel that some final stereo bus compression is the icing on the cake. If it is not too thick. Especially if you know it's going to be done again somewhere down the line like a radio station. So that brick like waveform is already overblown and when it gets to the radio station, even though their equipment is geared towards making things sound competitively louder, you're not giving those little automatic devices anything to work with. So then they do the wrong things like not making your really loud mix sound really loud.

    Now the careful considerations to take is that fast attack times while they are good for radio stations to keep them within proper operational standards is something akin to having your dog or cat fixed. They weren't broken but their lack of optimization makes them do nasty things. So in that instance while they're not broke we do fix them. Conversely, speeding up release times makes things sound louder and more aggressive. Keeping your release times down, slower will retain more of the natural sound by not responding so quickly. The trade-off there is if a particular vocal or instrument needs that greater sense of apparent loudness and/or placement in the mix, you might have to make your release times on that particular track just a little faster. Frequently selecting the fastest or the slowest anything on an adjustable dynamic range modifier, whether hardware or in software rarely sounds acceptable. The analogy here is religious or political extremism. Anything taken to an extreme on any kind of audio mix will generally result in disappointment. Good in theory, bad in practice. This is the biggest problem with learning how to make good recordings.

    Should anything be turned down to 0 (zero) or, turned all the way up to 10 (balls to the wall)? The answer is very few things. Since everybody has different techniques you'll find just as many different suggestions. This may also be the reason why more folks today, with all of our fabulous capable computers & software really have more problems making recordings than we did when we didn't have all of those important selective decisions to make. Most studios back then had a couple of simple equalizers and maybe just a couple of compressor/limiters. So you couldn't put all of this gobbledygook that we have available today in our computers in and overused fashion. Which I guess has become fashionable. And when you listen to those favorite old hit recordings from the 60s, 70s, 80s and wonder how they got such a great sound, it's probably the fact that they didn't have a whole lot of junk to work with. So everybody had to grab a fader to accomplish a "dynamic mix" without automation or a compressor limiter on every channel. In a related situation, I had to go to another studio to mix some tracks for a friend. She had recorded her latest CD at a US Maryland state University studio. The studio was essentially built, run & taught by a fellow with a PhD in the recording arts and sciences. ???? At first, I thought this guy was pulling my leg? You can't get a PhD in that. Ahh but at the University of Maryland, if they don't have a PhD program for which you want a PhD in, you can create your own. ??? What kind of fraudulent degree is that? Who are you going to give your thesis to? A physics PhD that knows nothing about audio?? And not only was this control room totally unprofessionally designed & built, thought out, almost shockingly so. Yes shockingly so. While this fool had a " Brand-new Digital Limiter", it had been normaled into the stereo mix bus of the console. I told him to patch it out. He told me you couldn't get a good mix without it. I told him I didn't care, I don't use bus compression when I mix except when the producer or whoever hired me requests it. He assured me that my mix would not come out acceptable because he knew. Obviously he's a knowledgeable PhD which I think stands for " Pretty horrific Dysfunction"? He was absolutely amazed at how fast, fat and dense my mix was without the be-all end-all digital bus compressor. That's because, I don't have a degree, I have knowledge & technique along with a great ear.

    If I want to play mastering engineer, sometimes after I complete a mix and like it, I might go ahead and see what a little overall stereo compression might provide me with? And then I'll do a goofy thing. The goofy thing will be taking my uncompressed mix and combining it with the newly slightly compressed mix. Then I get the best of both if they are carefully balanced together. Everybody has their own way of doing the same things sometimes. Some hardware & software compressors already have the ability to fade between the unaffected original source with the newly affected dynamic range enhanced portion. So moderation is the key with all of this stuff even when you want loud and aggressive. I mean when we want to have a drink, an alcoholic drink, wouldn't it just be easier to have a single shot of 190 proof grain alcohol? Maybe a couple of shots of scotch instead? Perhaps a case of beer for a single person? (A bottle of Red? A bottle of White? Or, perhaps a bottle of Rosé instead?) It's all the same thing but is it? No, it's not. One is a flash in the pan. One is to be savored. And one can be enjoyed throughout an entire day's worth of football. So the 190 proof is fast and aggressive quick and dirty. Then there is slightly more intellectual. And the other is a working man's good time. Which one is right?

    Now on your question 2:
    Your levels should be above " too low" & lower than "too high". Or the easiest thing to do is to simply Normalize your final mix, that is if you consider yourself to be finished and if you already like the way it sounds. Most of the time, depending upon your software, Normalization may have numerous options or none. Some normalizers have limiting functions also built into them. So you have to be careful there. Generally, you only want a normalizer that have no compression or limiting. You want it to be a Peak Normalizer not an RMS Normalizer. And you don't necessarily want it to default to 100% but instead, 99.6% or, -.6 DB instead of 0. This just gives you another margin of safety because anything that is peaking at 100% can actually cause some audible click like clipping. It's like not over topping off your gas tank when you fill your car up. And if you do that, how are you going to pour that STP gas treatment in if it's already overflowing?

    There really is no one-stop answer to your question. For that, I'm sorry. There are really so many variables to consider. Excuse the pun but you really are already on the right track. As you surmised in your observations, use of no dynamic range enhancement on anything will give you a huge dynamic range. But for certain kinds of genres that is what might be appropriate such as orchestral & operatic. It might also be applicable to straightahead jazz as opposed to avant-garde jazz? And it's generally an integral part of rock 'n roll of almost any variety. I know some wonderful old school engineers that think so much of today's stuff is over ProTooled. Then there is some pretty fabulous young engineers who are extreme ProTools fanatics think we're a bunch of fuddy-duddies. All would be correct. So what is the correct criteria to follow? There isn't any, Criteria is now called the Hit Factory in Miami Florida.

    As a side note, I walked into Criteria in 1979, with a resume & tape in hand. I was working right up the street in Fort Lauderdale for a multimillion dollar international advertising agency and cutting some pretty fabulous jingles with many of the musicians that regularly played at Criteria. The hip, cool, groovy dude sitting at the receptionist desk told me they "didn't need my resume and you can leave". Wasn't that nicely mannered of him. He must've had a lot of hits under his belt to be so self-important? Especially since after saying that to me. What I really wanted to do, I wanted to give him another big hit to remember. Right in middle of his nose! Rude SOB.

    Shouldn't Gold records be like the Golden rule?
    Mx. Remy Ann David
     
  6. IcePhixia

    IcePhixia Active Member

    Hi RemyRAD

    Thanks for this post. Once again, I learn new things :) No wonder so many good engineers still use vintage tools. They know how they sound and have been around those powerful tools for many years. Looking at todays wonderful world of powerful computer software, we have TO many tools, and have no phreaken clue how to really make the best of this limiters, compressors, reverb ext and have no frame of reference how they really work. Don't get me wrong, the Protools world is not a bad thing, but I bet you if you give a entry level student a protools rig with all the best plugins, and you take a veterane engineer with his vintage gear, the protools engineer will not come close to any good mix at all, even with the best tools :) It boils down to KNOW YOUR EQUIPMENT

    Now looking at my self as still in the learning stages of mixing/mastering. I somtimes feel lost looking at my variaty of options on a computer. Not to add that I have no frame of refence really what this plugin is doing. I've looked at various DAW systems like cakewalk, Cubase, Repear, Ableton, and the all have compressors, limiters, reverb, EQ and other tools to create the perfect song. Each DAW have there own flavour of those tools, and then you can buy or download free plugins and tools to add to your arsenal of things. What we should realize is that DAW is just a recording platform, to capture the sound in a digital format. The mixing part is a art in it's own right. Using the correct tools is what makes this recorded song go from zero to hero. But you must know you equipment, just like any good guitarist will tell you the difference between his Gibson LP and Fender is the tone he gets from his wood, the body of the guitars, the pickups and the amp he is using and that different styles requires when you grab the Gibson or Fender. Same applies for mixing using the Compressors and Limiters. Not knowing the "tone/sound" your getting and after, you will distory the quality of your mix.

    Very valid point! And I have to say I agree with you that no PHD will help you in this field :), but know your equipment and trust your ears. So, I've decided to really keep things SIMPLE and find myself a good quality compressor that stood the tests of time and learn how to use them in my mixes. That is why the CLA-bundle from waves looks like a good starting point. It's the true compressors of all times that you will find in any studio today. 1176, LA-2A and LA-3A and learn how those work on and in a mix, and understand the tone/sound that I am getting out of those. In time my ears will know what to use when and where, and my mixes will start to get the same qaulity. Plus, using only a few compressors keeps things really sime and you get the same "feel" for those dynamics. If you use somthing that you can tweak and change, you always get a different sound. One day you like this setting with all this buttons on. Tomorrow, you like another setting. With those 1176, 2A and 3A compressors you have only a few things to dail in on. So, it kind of telling you - LISTEN to your track and work those dynamics, rather than impressing your friends with the cool buttons and your SO CALL knowledge of things :)

    On your point about the Normalize with a setting round about the -.6db is what I normally do with my final tracks, keeping it away from the 100% level that will introduce clipping. I normally test the final mix on my weak small speakers, and it really clips and break on those small speakers when you hit the track with a 100% levels, so stayin round about the -.6db range keeps it in the realm of small speakers, which is what most people listens on today. Mastering for those big speakers system is good, but if your mix soundds good on small speakers, it 9 out of 10 times sounds big on big speakers. The radio air play story problem we've been having is because we didn't control our mixing and then we wanted to fix the stereo mix-down with a compressor/limiter to get it down to a -.3db level. Sending that track in will sound weak and sometimes the louder parts gets ducked in because the radio compressor is controlling it down.

    So to recap what I've been learning so far :) (and your post is really helping me realize all my mistakes!!!)

    (1) Keep it simple
    (2) Know your equipment
    (3) Get only a few good compressors/limiters and learn them
    (4) Trust your ears
    (5) Keep the final mix under the 100% rader, close to -.6db
    (6) Rather use a Normalize for the final mix than a limiter/compressor
    (6) Know the style of music your working with when applying compression/limiting
    (7) No Phd will help you but all fasion experience will
    (8) Gold records is the Golden rule, so learn from those
    (9) HAVE FUN while doing this! ;)

    Once again thanks for the infomation and years of experience you share - it helps our younger generation to learn from the people who still have a soul for music and love what they do!

    That sounds like something out of a movie :), and that happens so often - but I found that those clever people is all around us. I've been to a studio once and the guy was the so call "best of best". He took our mix and did something on our guitars that ended up clipping in the final mix, but he said that is how it's been done and he knows the best. If you don't listen to your clients you will end up mixing the song wrong and lose clients, because you know more based on what? hehe, so the rude guy - I hope he realized his mistake ;)
     
  7. audiokid

    audiokid Staff

    Good thread.

    Some of the local stations here are using such poor equipment even the best produced music is pumping. If you are 3 db hotter in peaks than the norm, your track is going to be pumping more and it will be very noticeable.
    I'm somewhat confused over the general answer to this thread though, but I like where it is going! No matter what song you have finished, it shouldn't need to be optimized for that station. It should be radio ready and competitive and comparable with the norm so you at least sound like the rest of the music.

    To the OP, you have room to move on that example for sure. I'd love to hear some juice talk from the ME here.

    Back to what Big K and Remy said!
     
  8. audiokid

    audiokid Staff

    Would it ever be cool to get a bunch of us to do a 30 sec clip of the same track and compare notes.
     
  9. Herbeck

    Herbeck Active Member

    Yes it would, a 30 sec clip can be worth a thousand words.
    Just the other day I learned a lot from a 30 sec clip about analog/digital limiting.
    The hard thing is to avoid the aggressive ego debates that often follows.
    I guess that's one of the reasons why there is so few audio clips on ME forums, and that's sad.


    Cheers,

    Herbeck
     
  10. audiokid

    audiokid Staff

    Lets change that.
     
  11. Herbeck

    Herbeck Active Member

    Sounds good.=)
     

Share This Page