Normalising

[goddamn rain]

Hi there

When I've finished harbalizing a track I set the output to zero. Fine! (These are mostly 32 bit float files.)

If I use 16 bit files and set the output to zero, am I effectively normalising them? If so, will this have a degregading effect on the quality?

Should I re-save my 16 bit files as 32 bit before I harbalize them or would this be pointless?

Am I just a worrier?

Cheers

David

[Har/Bal]

If in normalising your 16 bit files you introduce a positive gain it does make sense to first convert a 16 bit file to 24/32 bit provided that conversion includes dithering.

HarBal does not internally dither low bit depth data to higher bit depth so it is likely that by doing so externally you will have a minor improvement in perceived quality. How readily you can hear that difference will depend a lot on the source material.

Regards,


Paavo.

[mflorio]

Hmm... I've never heard of dithering to *higher* bit depths ?

Anyway, it absolutely does make sense to convert 16 to 24/32 before doing any processing in HarBal (or any other process). Then, at the final stage dither back down to 16 if you need to burn an audio cd.

[goddamn rain]

Thanks guys.

Reason I asked is because I want to tweak some finished mixes that I did before I bought Har-bal. These have already been processed and dithered so I figured i would probably have to re-save them as 24/32 bit files before har-bal did it's thing!

Thanks again

David

[bbdude]

I posted on the old forum on a topic related to this. I was planning to combine Har-Bal with Ozone, first har-balize and then do the final loudness maximization and dither in Ozone. If I do this, is there any reason to change the gain setting in Har-Bal?

Also, I'm still not sure if I understand what the normalising process is doing in Har-Bal. Is this just basically just turning up the output volume?