If all your tracks on their own remain in the green, but all combined frequently touch red, but yet do not go to he top of the bar is this a bad thing? I hear that you should have the final mix at 0, the final mix at -3, I just wanted to bluntly ask the question, what does it actually mean, and does it matter? thanks 
You should leave some headroom, -6 usually there’s some good reading here:
http://tarekith.com/tutorials-and-reviews/
If your putting it online (soundcloud or such) you can normalise it after its rendered.
I’ve read soe of that link, I will get round to the rest, it is already an interesting read so far, can I ask, what is actually meant by -6 db, how can I see that its at -6db? Honestly, talk to me like a child, because i’ve got my head around creatng complex sounds, but this ‘bit’ seems to go way over me, like many other things i might add…
Thanks
If you select the master track then at the bottom it will show you the decibel (dB) reader, then there is ‘Peak L’ and ‘Peak R’, these readings show you the maximum (peak) volume which the track has encountered so far (see attached file for details, note that the line cursor [golden rectangle] is on the master track). So play your tune through and if it’s around -6 you’re right on course. It will also tell you the peak dB when you finally export the tune to a stereo track. There is also a peak reading on each seperate track, but the master track is the one which will tell you the overall volume of what you’re making combined. If you are going into the red then the sound will clip and cause unwanted distortion which will begin to sound harsher the more you amplify it, so if it got played in a club then it would sound really bad if your tune red lines whilst you’re producing it.
It’s pretty important in terms of the overall sound of a song, and a good thing to get to grips with from an early point. You can stop tracks from clipping by mixing and the application of compression and/or limiting but that will come later.
Finally, the -6dB rule is around because mastering engineers like to have that much space to play with when they get a track sent to them and it allows them to really work on a track, but from a beginners point of view I wouldn’t get hung up on -6dB just yet, I’d just focus on making it so that your song isn’t clipping.
That was awesome. Such a great help. I’ve almost finished a few tracks, and I want them to be of a professional standard, I should get hung upon the -6db by the looks of things lmao. I thought mixing was just the final touches anyway? I use compression, equalisers, etc etc, but it looks like I was 6db off this whole time. Are there legit renoise people you can send your work to to get it mastered?
Maybe get in touch with this guy: Atlantis' New Mastering Topic
(I don’t know him - you do the talk)
Hmm, about that, why can’t the mastering engineer just lower the gain by 6 dB him/herself if s/he wants that kind of headroom? Wouldn’t that be the same end result as if you kept all the relative volume levels the same, yet rendered at -6dbFS?
Or does lowering gain decrease fidelity slightly, because extremely tiny sample values will be rounded to 0 that might not have been, if you’d rendered at -6dbFS to start with? (Experimenting on my own doesn’t seem to show this: a simple test pattern rendered at different dB levels perfectly phase cancels itself when normalized. Likewise, lowering a waveform in an external editor by a substantial number of dB (say, 40dB) and then renormalizing it phase cancels as well.)
That -6db headroom actually originates from analog era. I don’t even know if it makes any difference in digital times. I honestly don’t see any technical difference on audio file with -6db clean headroom and something that is normalized (not compressed) to 0db.
I’ll let you guys hear the piece when it’s finished. I’ve learned a lot about mastering. You guys helped a lot
Don’t get hung up on -6dB, it is a number used as thumbs up as engineers to have some headroom in case if something goes haywire in the mix the master will not overload. -6dB allows you to safely double the volume, or in other words something has to go really really wrong in your mix to start clipping on master. But in case you know your mix inside out you can reasonably safely approach the 0dB mark, although -0.3dB is recommended by some as some (not very good?) DACs might not go all the way to the 0dB, or whatever.
If you are not super sure what is going on in the mix and you have to render it just now and not a minute later, or if it is very long and boring so that you are not bothered to listen it through, give it plenty of head space, render it in 32 bits, then normalize afterwards and then do whatever you need to do with it.
Really, getting at 0dB problem is really a problem for live performance, or if for example you want your module sound as loud as possible in the listeners’ Renoise instance without clipping, you would have to set the levels manually.
I have done a few songs since I asked this question…and not one of them has been mastered. I’ve just avoided cipping :S So it seems I havent progressed. I just have more to do. Two questions: where and how are you normailising? In Renoise? and If it helps I literally have not used the send track Once, I’ve read and watched countless articles about final mixes, very lttle seems to be about on final masters, but there was quite a helpful video of a quy sending three bands of frequencies to hs send to achieve an unclipped final piece… but I barely understood how that worked, and when I donloaded his track, it was just no noisy, I honestly couldn’t figure out why he needed to avoid clipping. every one of his samples was distorted so much. I willf ind it for my next response, I’m sure the guy is a pro and everything. so the other question, it may seem redundant, but how does the ‘send’ help with the final master?
great help. did you buy the plugin he uses?
Voxengo SPAN can do K metering. Personally I can’t tell the difference between K meters and pretending -6 or -12 is peak, but I think that is in part because I don’t understand the theory as well as I understand the result.
That’s a good video clip, it affirms some good habits I discovered on my own time.
I tried a few different ones kmeter by mzuther, span and ixl but in the end bought the meterplugs one , It just seemed to detect things the others missed.
Thats what I originally thought I would set a point say -6 and and make the peaks hit around that. Using a k-meter I aim the average loudness to 0 on the meter, focusing not so much on the peaks as long as they aren’t to extreme or clipping, they are the parts the mastering person needs.
If I got it right, K-metering is about dynamics, not about absolute peaks. So you can’t simply say “let’s pretend -14 is the new 0”. There is a relatively large time frame, and the algorithm takes all the data in the given time frame into consideration, in order to ‘calculate the dynamics’.
That’s right, here’s a quote from meterplugs site:
SO, peaks established at -6 for a master? and what, averaging at something lke -8? Lets say I get all that, now we have to fight the loudness war 
I’ll ask agan, how are you guys normalising? how are you getting your tracks the ‘industry volume’.
I have an side-predicament, if you will, which feeds int this mastering talk damn well I think, talking of the indutry standard
A piece I made is too bass-heavy, its also a touch quiet, I know I’m going to have to EQ, but when I look I at the spectrum, what exactly is the ideal look? (Im talkng frequncies to decibels) What are the ‘industry’ (corporate slave maybe, but I oon’t own one apple product) preferred frequency loudnesses? should below 70hz peak or average at -12, etc, etc?
I di read something abut lower frequencies needing more presence, it has a term, read it in a renoise article, and I get that, but obviously if you give the lower end more loudness in a linear fashion cmpared to the high end, you’ll end up with a dodgy… dodgy result… as I basically di. Unintentionally.
Thanks 
I’ll purchase this K Meter when I get my next pay check
Loads of stuff on therecordingrevolution.com I really liked this one:My link
And this one as I often find myself having to do this, and it kind of answers your question:
http://simplyrecordingpodcast.com/mixing-on-headphones/
If a song is too bass heavy, it will sounds psychotically quiet, so you may want to reconsider the mix.
I avoid raising levels to proverbial industry norms, if a mix is way too quiet I’ll use Maximizer and crank… I can’t remember the last time I felt I had to do that.