[splicestudios]

I'm involved in a project that has deliverables to both UK and USA (co-production, hence the double deliverables): the UK master uses -18dBFS reference tone, peaks no more than 8dB above reference (i.e. -10dBFS), the US master uses -20dBFS reference tone and peaks at -10dBFS. So both peak at the same level. I've seen DCI specs before where the LM100 (yes that dreaded box) has different readings for PAL (UK) and NTSC (US), giving the PAL range as 2dB higher (-24 to -26, versus -26 to -28), which I assume mean that PAL is louder than NTSC by 2dB.

So am I right to assume that the UK specs allow for a narrower dynamic range (slightly more compressed sound), that would also average 2dB louder than its US counterpart? I've read DCI specs discussing RMS voice for their shows should be -32 to -20dbFS but overall RMS / VU should never pass -17dBFS.

I always think of British shows as having a more 'open' and dynamic sound, and US shows as having a far more in-your-face sound where everything is loud (although not necessarily through just applying heavy compression / limiting).

In making the 2 versions, it's not just a case of me dropping the overall level of the mixer (who is PAL and uses a -18dBFS reference) by 2 dB to compensate for the US master?

It wouldn't work the other way - a US show at -20dBFS ref with peaks at -10dBFS can't be bumped 2dB cos it would make for peaks of -8dBFS.

Or... have I got this entirely ass-backwards?

Even though I am in a PAL country, most of my broadcast clients go with US-style conventions of -20dBFS ref and -10dBFS peaks.

Regards
Kenn Delbridge, MPSE
SPLiCE Studios
Singapore