For your situation by keeping your volume down until the end means you don't HAVE to worry if something gets a little loud. If it sounds good you can keep it without having to constantly readjust your faders. You can just let it be loud and not worry about clipping.
 
Then when your mix is done you make up the extra volume with the limiter. It doesn't even have to actually DO anything except raise the volume. I do this all the time to check out premaster mixes. I use the limiter to turn it up as much as possible but NOT reduce any peaks at all. So the limiter doesn't CHANGE the sound. It just turns it up.
 
It's hard to explain in a simple way (or really any simpler than has been explained in this thread).
 
It will just make your life a LOT easier and likely help you have more freedom with your mixes.

Thats now how normalize works. It searches for peaks and places them at zero. The rest of the audio is brought up proportionally. If you have an audio track with peaks already near zero and much of the audio is -24 dB you will not see much change in loudness even though the audio is relatively low for the most part.

 
I bag to differ. The Normalize will increase the level of intro if it is at -7, even if middle part of the song was close to zero. I might be wrong as I learned it when I was mastering with Stainberg WaveLab. There every time you use Dinamics, it offers to normalize, so I may be thinking that it was normalizing but it was compressing and then normalizing. But Normalize does resample the track and takes everything to the Zero point. Yes, compressor suppresses the peak. That is all. It does not increase anything. In WaveLab it was 2 steps process. First you compress and suppress all peaks. Then you normalize and you can see that it changes from pine cone shape of the wave in to the sausage, and the Intro part becomes the same level as the body.  

Uh... vlad, you really should be listening to what's being said here. These guys aren't lying to you. They're trying help you.

What will make my life easy? How is limiter "just makes it louder"? I don't have to worry. My tracks are where they should be by what I hear. Some are very low, some louder. I keep the final Mix by -0.1. Then Limiter has no need to make anything sound louder- it just color it and makes it feel and sound louder.

vladasyn

Thats now how normalize works. It searches for peaks and places them at zero. The rest of the audio is brought up proportionally. If you have an audio track with peaks already near zero and much of the audio is -24 dB you will not see much change in loudness even though the audio is relatively low for the most part.

 
I bag to differ. The Normalize will increase the level of intro if it is at -7, even if middle part of the song was close to zero. I might be wrong as I learned it when I was mastering with Stainberg WaveLab. There every time you use Dinamics, it offers to normalize, so I may be thinking that it was normalizing but it was compressing and then normalizing. But Normalize does resample the track and takes everything to the Zero point. Yes, compressor suppresses the peak. That is all. It does not increase anything. In WaveLab it was 2 steps process. First you compress and suppress all peaks. Then you normalize and you can see that it changes from pine cone shape of the wave in to the sausage, and the Intro part becomes the same level as the body.  

I think you need to read more about this. I'm not sure how to address what you wrote. You are in effect telling me that normalize while using a compressor will compress the dynamic range. Yes that is true. Normalize by itself will not alter the dynamic range. That is also true and you need to understand that. 
 
This thread was about clipping. The only way to not clip is to lower the volume. How you do this is up to you but you can't stop clipping without an adjustment to the volume. I hope you understand this. 
 
 

vladasyn
What will make my life easy? How is limiter "just makes it louder"? I don't have to worry. My tracks are where they should be by what I hear. Some are very low, some louder. I keep the final Mix by -0.1. Then Limiter has no need to make anything sound louder- it just color it and makes it feel and sound louder.

*sigh*
 
You started this thread by saying things were clipping. What is being said is that if you mix at lower OVERALL volumes from the start you NEVER have to worry about clipping EVER. Then you make up the difference at the very END of the signal chain by turning up the master bus. Nothing needs to get louder or quieter in relation to each other. Nothing needs to change for your mix.
 
By "making it louder" I am saying it can be just like turning your speakers up (your mix doesn't change when you do that... right?). The limiter is ONLY there just in case something DOES clip. If your mix is completely perfect already then you can just turn up the gain knob or the volume fader on the Master bus as loud as it goes before clipping. That is the EXACT SAME THING I am saying you can use the Limiter for on the master bus.
 
It's just another volume control with "clip protection" at that point. You can however use it to tame peaks as well or completely squish the song. It depends on how you set the limiter.
 
Anyway... if you are happy with your mixes then by all means keep working as you have been. I guess that;s all that matters.
 
Cheers.

vladasyn
But this blog sounds immature. Who in the world care about where the Faders are. There is no need to pay attention where the faders are. That number has no value...[snip] he says- the effects may be clipping? Come on now! No visual working? My effects do not randomly clip. There usually is a meter and you can see or hear if it clipping.

 
A few things to remember. First of all, if you never use any plug-ins, you can stop reading now. Keep the channel faders as high as you want, the master fader wherever you want, and godspeed.
 
The world of digital audio is changing constantly. I see people on the web offering advice that is very relevant if you're living in 1991 and using Pro Tools, but has no relevance to 2016. So, beware of misinformation. I have a feeling you are thinking 1999.
 
Next, there is a difference between what makes sense theoretically and what makes sense practically. In theory, you don't need to keep the master fader at zero; in practice, if you're putting any plug-ins in the master bus, it does.
 
Okay.
 
In theory, with SONAR's 64-bit engine and summing bus which have virtually unlimited dynamic range, you can run the channels totally into the red and not get distortion. In practice, you probably use plug-ins and they are not part of the summing bus.
 
People in these forums get all freaked out because some ProChannel module clipping LEDs may go on when the channel fader isn't in the red.
 
The ProChannel clip LEDs relate solely to the ProChannel plug-ins. They have nothing to do with what's happening in the channel hosting them. This is true of plug-ins in general.
 
Think about it. You have a module that emulates analog gear. What happens when you push way more signal into analog gear than it can handle? Do you want to emulate that too, and use up a gazillion CPU cycles to emulate the sound of analog audio crapping out? I would hope not. So, those clipping indicators show when you're about to enter the range of irrelevance. And many plug-ins don't indicate clipping, or have a limited dynamic range just because the designer saw no need for it.
 
Conclusion #1: You want to keep channel levels within rational boundaries if you use plug-ins. Peaking at -6 to -12 is fine.
 
Next, stuff happens. A peak hits on the boundary of the 40 ms meter framing window. A random modulation effect causes a boost. Acidized loops can produce different output levels at different tempos, so the loop that sounded great at 120 BPM when you loaded it may be going "into the red" at 127 BPM. Which you think is fine because SONAR has an ultra-cool audio engine that can take anything you throw at it, but bearing Conclusion #1 in mind, the plug-ins it hosts may NOT be able to take anything you can throw at it.
 
Conclusion #2: Stuff happens.
 
Mastering and mixing are two different things. Mixing is all about getting a balance among various signals. If two signals sound right when one is 6 dB below the other, then it doesn't matter if the signals are at -12 and -6, or -15 and -9. Mixing is about balance, not level.
 
Mastering these days is about level. So you take your properly-balanced mix and beat the living daylights out of it with a multiband limiter so it's nice and loud, causes ear fatigue, and turns people off to music. Or not...it is possible to get a loud master without having it sound horrible, but people pay me a lot of money to do that and it would take a lot of pages to describe the processes, so we'll leave that for another day.
 
Conclusion #3: Use mixing to get the balance right. Use mastering to get the final level of the stereo mix right.
 
Finally, normalization is not the same as limiting. Normalization analyzes the difference between the highest peak and 0. For example if the highest peak goes to -4.3 and you normalize to 0, everything will be amplified by 4.3 db - loud parts, soft parts, whatever. Your dynamic range stays the same. It's like turning up a volume control.
 
Limiting restricts dynamic range by reducing peaks. After reducing the peaks, your maximum peak is now below 0, which opens up headroom, so you can apply makeup gain to make the overall signal louder.
 
Conclusion #4: To get the maximum level so that your music can be heard above the sound of vacuum cleaning, buses going by, living in a flight path, or road noise, limiting does the job, not normalization.
 
 
 
 
 
 

One more thing: You may not perceive clipping as such, but it's cumulative. If there's a lot of subtle clipping going on, it will destroy the clarity of your mix and you won't know why, because it's not obvious.
 
Often pulling back on individual channel faders and increasing the master fader will all of a sudden make the mix sound "clearer."
 
Also, remember that how effects react usually has more to do with what's going into them than what's coming out of them. Slam the input of a compressor, you'll get more compression. Slam the input of an amp sim, you'll get more distortion. The fader is determining how much they contribute to the mix. What contributes to the effect is the signal level that precedes them. Remember...balance.
 
And ultimately, in response to your thread title, the meters are accurate at the point where they are taking their reading, withinthe window during which they are taking that reading. 

The reason we got to talk about how limiting works is because somebody said- they would mix to -3 and then use limiter to bring it to -0.1, which to me is not a good idea. I mix to -0.1, then limit peaks by master limiter and bring it slightly higher. I still do not see a reason to mix under -3.
 
I do use plugins almost on every track. I like what you wrote: Conclusion #3: Use mixing to get the balance right. Use mastering to get the levels right. Still I think when you mix, you may as well have it as close to where you like it to be as possible.
 
What happens with my mixes is this: I have 7 minutes song with 70 tracks. About 250 measures. I play song over and over and over and over for years (I still mix song I made in 2008 today) and make adjustments to individual instruments to make right balance between the instruments. The problem is that when all tracks are at the right balance, some tracks sound low but clipping. Let's say Vocal. I want it at this level. 90% of the track is under the Zero, but it has let's say 10 peaks above zero. I can not lower the entire vocal track to get rid of it because softer parts will be too low. I can not lower all 70 tracks down without risking of affecting overall balance. So I have to use automation. I enter node and lower that individual part which was clipping. (Now word clipping is wrong. My tracks do not clip as being recorded above zero- that would cause ugly vertical line and distortion. I am talking about normal recording level and peaking on playback).So when I use automation and I set this part to -7, the clipping disappears. Then I play from beginning of the song and that same part clipping again. May be I should put the limiter on that vocal track, instead of entering nodes on automation. I started another thread about clitch on entering nodes. It is very painful trying to lower one second long peak by entering 4 nodes. It stops accepting nodes and I have to close and reopen Sonar to get it back to work. So that was the cause of frustration when I started thos thread. After struggling for 3 hours fighting peaks, they were partially back when I played from beginning. This musty be the effects on the vocal track.

This is a very interesting and informative article on levels and gain staging, and I would highly recommend that you read it. 
 
https://www.soundonsound.com/sos/sep13/articles/level-headed.htm
 
Of interest:
 

Bad advice aside, I believe there are also several reasons why the DAW software itself leads people to make mistakes. Partly, I think it's due to the huge amounts of 'internal headroom' offered by modern 32- and 64-bit floating-point software — you could theoretically describe a dynamic range of a whopping 1500dB in a 32-bit floating-point system without causing problems. That enables you to apply serious amounts of gain if you wish — in theory! In practice, many people, including me, believe that the summing engines in different DAWs don't always produce the same results when summing lots of very high-level signals, and that these differences can be audible. The probable reason for this, to put it simply, is down to differences in how the 32- or 64-bit floating-point calculations are rounded to create the 24-bit audio part of the floating-point data.

 
 And also this: 
 

The second way in which DAW software can mislead is in its digital sample-peak metering. That's used by default in all DAW mixers (although, thankfully, progress is at last being made in meaningful loudness metering on the stereo mix bus). The sample-peak meter indicates the amplitude of the highest audio sample at any moment in time, and provides an approximation of the actual peak level of the reconstructed audio waveform. The approximation is perfectly adequate for use on the stereo bus, or any channel where you plan to send the signal out into the analogue domain — if you're working with a sensible headroom margin — since any amplitude errors are non-critical.