Here are a couple of reasons to normalize, when, due to things beyond your control, the FOH guy gives you levels that are barely there (but in a 24 bit world are still useable):
1. DAWS are a audio-visual tool.  What's easier to work with when you have 2 hours worth of 48 tracks with weak signals, flat lines or visible waveforms?  Consider how difficult it is to replace flubbed notes by the horn section with a snip of a good performance from the same track, if you can't see them...
2.  I, like many others, have mix templates with all of my dynamics processors and other plugins for drums, guitar, bass, keys, horns, etc. already in place.  When levels are way below what you would normally record at, those devices must all be adjusted to sometimes extreme degrees to get proper amounts of compression, etc.  I am talking about situations where channel gain can not make up the difference.
3.  When you have three shows to pick and choose the best performances of in very short order, #1 and 2 become indisputable.
4. Never say never (as in "I will never apply fill-in-the-blank" to my work), (and never tell the FOH guy that he is an idiot - it does no good)

Back to the issue at hand.  I have 48 tracks, ~2 hours long X 3 shows.  Doing each track individually means I will be sitting in front of my machine for an extra day, just doing the normalization :-(

Coffee on standby?

Poco
Back to the issue at hand.  I have 48 tracks, ~2 hours long X 3 shows.  Doing each track individually means I will be sitting in front of my machine for an extra day, just doing the normalization :-(

I just tested with X3, and it behaves similarly. I appreciate your bringing this up as now I understand why normalizing multiple tracks is limited by the level of the highest-level track.
 
As you're verging on digital audio editor territory, the only alternative that comes to mind is using an editor with a batch processing function. Then you could simply drag the project folder audio files into the editor, set up batch processing, push a button, and come back later when they're all done.
 
 

Thanks for your help, Mr. Anderton.

By working blind I'm meaning the huge difference in using the Normalizing tool in Wave Lab or Sound Forge. 
First I like to see what the peak level is, because sometimes a very quiet track will report in a 0db. So now you need to look for that one spot that is peaking etc. Also Sonars Normalizing works with percentage which is totaly weirdness when your thinking db's. Mabey they fixed it in x3 as I admit I haven't bothered with it for 3 years. Tool copy works fine. 
 
Poco I now see what your doing and agree 100%. I would look into seeing what software is capable of this function. The right tool for the job. 

johnnyV
First I like to see what the peak level is, because sometimes a very quiet track will report in a 0db. So now you need to look for that one spot that is peaking etc.
 

 
If you've played the clip, you can right-click on the peak numeric value in the track view and select "go to Peak." The indicator will be a little late compared to the actual peak due to framing, but it's usually obvious where the peak is if you look about 40 ms to the left.
 

Also Sonars Normalizing works with percentage which is totaly weirdness when your thinking db's. Mabey they fixed it in x3 as I admit I haven't bothered with it for 3 years. Tool copy works fine.

 
You can enter the normalization ceiling in dB or percentage. 

According to phone tech support: WORKS AS DESIGNED. 
 
If you want details on the original issue, look for the other thread with a similar title.  I can't figure out how to change the title of an existing thread...  So, according to the helpful tech support guy (who tried it and did reproduce it), you should NEVER BATCH PROCESS NORMALIZATION.  Excuse me?  I have to process 48 2.5 hour tracks individually BY DESIGN?  He made a rather weak argument for never doing anything to two or more tracks at a time   Anyway, I am going to try putting some sort of gain boosting plugin in each track (he suggested a buss, but that won't print back to the track), and APPLY EFFECT, to see if that works. 
 
Here's an interesting question:  If batch normalizing of tracks creates one big wave file associated with all the tracks, what else does that too?  You don't even notice it unless you go to "Associate Audio Files" and check it (unless of course waveforms don't redraw, as in my case).  I asked the support guy to check with the developers to see if this really is the way it's supposed to work, and he said he would.  Let's hope he does.

I joined this post to the original thread and modified the thread title.
To modify the subject thread, edit the original post and change the subject line.

Thanks...