I can only speak for myself, but I always use -3.0 as my peak. The reasons I don't use 0.0, besides a few distortion/clipping problems, is that somehow, when I render AC3s, sometimes I get nothing or what sounds like a backwards recording. I never noticed a difference in rendering speed though. (never checked)
-3.0, ...... I almost never have a problem with AC3s.

Another reason to never go higher than -2dB is that some oversampling systems will actually generate distortion. but -6 as a peak...that's a bit low for most applications.

The thing is unless Grazie's audio was already hitting 0 is it really shouldn't make a shred of difference, well actually running another 6dB down is making things worse (wasting headroom)
But the big question is how are you evaluating this, what are you encoding to etc, etc. Way too many variables I think to offer any concrete opinion.

The other thing we need to know for sure is....the Title says -6dB and his 1st mention says -6dB, but his 2nd mention says -0.6dB. Which is barely lower than 0.0.

Thank you. Just waking up! OK .. :-

1/- Typo: 0.6 dB was a typo. It should be 6.0dB.

2/- Render: Rendering to AVI using the default for Audio: PCM Uncompressed; 48kHz; 16 bit depth; Stereo.

3/- Evaluating: Over Speakers and looking at Wave forms.

4/- Trialing: I've always wanted to try this out. I tried -3dB and thought, now THAT sounds good! What happens if I go another -3dB?

5/- Bob? Please explain your, "However be aware, the level you set on the header is post track FXs, this can be very confusing when dealing with compression."

6/- I'm going to experiment and then use a combination of knowledge from Steve and Douglas = -2dB and -3db and keeping a bit more headroom. Bob's first comment about getting "internal" clipping is another pointer 'cos as I was watching the meters push/float into the higher yellows, I thought 'ello?!? Going to keep an eye on that!

So, I'll NOT go as far as -6dB, but rather keep within the -3 to -2 range.

. .and the answer generally is that it is "recommended" to drop a tad under 0.0 dB - but certainly not -6dB. Yeah?

Thanks!

Grazie

re 5)
'Signal' goes through the FX chain before the volume envelope and the track level control. On many things this doesn't make any difference but for FXs that affect dynamics it makes all the difference in the world.
If you've got a track that's peaking to -10dB and you apply compression with a knee at -6dB it'll not be doing much and using a volume envelope / header gain to boost the signal is lifting the level of what's coming OUT of the compressor, not what's going into the compressor. In other words the FX chain is in the same place as the FX Insert jacks on a real mixer.

Now if you want to do it the other way around you add the compressor to the Master Bus say. The master bus is post the envelopes etc so now you're controlling the level going into the compressor, the compressor will work properly.
All this is best explained with a diagram which I cannot post here but there's a good one in the manual.

Typically I set up several busses with different FX chains and route tracks to them. Music gets left alone, speech might go into Graphic Dynamics (used as a smooth noise gate) and then medium compression. To control levels in the final mix I use the Buss Envelopes.

Bob.

Bob, thank you for taking the time out to explain "5" - very much appreciated - AND what you say makes admirable sense.

I can see I gonna need to set up a CTRL+ALT+D for this type of audio requirement.

Cheers,

Grazie