so are you saying that renders are 64 bit?

It renders at whatever your render settings are. Your final master audio buss meters should still not exceed 0dBFS no matter what your bit depth settings are or audio file format.The 32bit floating point is the internal structure, nothing to do with the rendered output.
FWIW, SloTools just switched to the 32bit floating point architecture.

Also, regardless of the bit depth, all the extra bits are at the quiet end of the scale. 0dBFS is still 0 no matter how many bits, and any amount over it exceeds the maximum value that can be stored and ends up as distortion.

While processing within the software, Vegas/Forge/Acid do allow substantial headroom while calculating mixes and effects, which is why they know just how far above 0 the signal would be. But they can't actually save that in the output file, and most likely the sound card clips it even during preview/monitoring.

I tend to make sure my peaks never go above -0.5.

i guess what i dont understand is i imported a 2 track file that was recorded in pro tools without clipping and the waveform also reveals it does not have high peaks but it plays back at up to 2 over

That is odd - have you checked in the Mixing Console that your channel and Master faders are set to 0?

all at zero


here is an image


http://tinypic.com/m/fbx0s2/1

OK, there's an explanation for that.

You have Mono tracks which are Panned hard Left or Right.

The Vegas default setting for Pan is "Add Channels (0dB Centre)" which, as you now find, progressively adds up to a +6dB boost to the mono signal as it is panned away from Centre.

That is because when a signal is panned Centre it therefore plays through 2 speakers ... you hear both and your ears add the signals again. But as you pan away from Centre, say to hard L or R, then only 1 speaker is heard, which would result in a quieter perceived level (without any boost).

The Vegas default "Add Channels (0dB Centre)" +6dB boost is rather savage (although it is electrically correct); most find a +3 to +4.5dB adequate in typical acoustic environments.

Most modern hardware mixing consoles also do this - its commonly called "Equal Power Panning".

There are 2 remedies to your situation:

Decrease the Level fader in each track by -6dB (which will now give you exactly Unity Gain).
or
RightClick on the channel or Pan Knob and select another "Pan Type" from the menu. (I recommend "Constant Power" or "Film")

BTW if you always prefer that pan type you can rightclick on a channel and select "Set Default Track Properties..." and next time you start a new project that setting will be used in all channels.

Thanks for the great explanation . i used to do that but a while back i think they moved where you set the pan power so i spaced out

i think i used to set it to Balance 0dB

are you saying Constant Power + hard pan L & R = 0dB ?

for that matter, do you suggest the +3 on each L and R for compensation? I have never heard of that and I just always assumed (in any DAW) hard pan L & R of mono tracks was what you are supposed to do.

Yes both "Constant Power" and "Film" give you unity gain (0dB) when panned hard L or R, and -3dB for Centre; intermediate settings give good approximation of "Equal or Constant Power Panning".

"Balance 0dB Center" gives unity gain (0dB) when panned hard L or R, but also send 0dB to both channels when Centre, therefore does not 'compensate' as in my description.... not a good choice in my opinion.

The other 2 options, "Balance -3dB Centre" & "Balance -6dB Centre" also give unity gain (0dB) when panned hard L or R, but also send -3dB or -6dB to both channels when Centre, which is good behaviour in a sense but behave different in intermediate settings of the panner; neither are true "Equal or Constant Power Panning".

A desirable panner is one where you can pan from L >>> R slowly and the relative level presented to the listener remains constant - this is most noticeable when panning some element of a mix -the 'best' curve for this remains one of audio's conundrums, because acoustic addition is not as precise as electrical addition.