Radio Processing Reference?

[Peakly]

Originally Posted by joeaudio

I've been trying to talk people
into not smashing music for years.

Sweet. Man that sounds so much better!

[OldSonic]

Originally Posted by huub

Well off topic as in it's not about radio processing, this was a tv gig..

Only difference is that there no pre-empasis an stereocoding on tv, all multiband processing are the same for fm-radio and tv.

[Radioman]

Originally Posted by OldSonic

Only difference is that there no pre-empasis an stereocoding on tv, all multiband processing are the same for fm-radio and tv.

TV audio has pre-emphasis too (50 uS Europe, 75 uS NTSC) but a smaller frequency deviation (+/- 25 or 50 khz) respect to the +/- 75 khz of broadcast FM, and don't use stereocoding because stereo audio comes from 2 separate audio carriers.
For the others aspects the TV and Radio Optimod versions are very similar in audio processing, the main difference is that TV Optimod has a separate setting for the "commercial breaks" audio respect to the "normal program" audio.

Generally 90% of FM broadcasters sets their audio processor with a sort of loudness equalization (low and high extremes boosted) and act first an AGC levelling and after the multiband processing. If the mix is well balanced the correction could be minimal, if the mix is unbalanced with a bass or high excess (or lacking) the Optimod acts to re-equalize it and put all the frequency ranges into the "spectrum mask" setted by the broadcaster.

[Junction]

Have only just jumped on this thread (and new registered member today), but as an ex radio guy of 21 years, and recently having gone back to recording, I find this a very interesting and important issue. I was personally responsible for screwing up millions of hours of broadcast music for years. But you know I see recording and broadcast audio processing coming closer together, the amount of compression used in the recording/mastering studio can be incredible, reducing average dynamic range down to 2 or 3 dB for much of the song. The task of the Omnia or Optimod processor is now really simple, we just have to knock maybe another 1 or 2 dB out of it and we have ultimate loudness!

But seriously, like we must be sensitive to the devices our music is being consumed, so must we be sensitive to how it is handled on radio. I applaud Masterer and others who are taking interest in the affect of radio processing on music and making minute adjustments to compensate a little.

I don't think it was mentioned here, but inversely to the pre-emphasis 50uSec or whatever is applied to the transmission chain, the opposite de-emphasis is applied at the receiver, so the end result should be flat response to 15KHz, excluding affects of processing.

BTW, on FM there a hugely sharp roll off after 15KHz, nothing gets thru at 16K, so good ridance to those sparkly and airey highs.

Keep up the interesting contributions.

Cheers
Michael

[Radioman]

Originally Posted by Junction

I don't think it was mentioned here, but inversely to the pre-emphasis 50uSec or whatever is applied to the transmission chain, the opposite de-emphasis is applied at the receiver, so the end result should be flat response to 15KHz, excluding affects of processing.

BTW, on FM there a hugely sharp roll off after 15KHz, nothing gets thru at 16K, so good ridance to those sparkly and airey highs.

Yes, the 50 or 75 uS pre-emphasis (3,18 or 2,12 khz cutoff frequency) is compensated by the de-emphasis in the receiver, but it means that in the TX the "0 dB" apparent level decreases in a 6dB/octave mode upper these frequencies, while in the cd the 0 dB level is equal from 20 to 20 khz.
So it isn't possible to transmit a cd without a frequency dependent compression sidechain network. This is the reason because the broadcast limiter have a band splitting, because the HF can't be processed linearly but we must compensate the 6dB/octave gain upper the cutoff frequency, and if we don't want to loss loudness in LF area due a HF excess we must split the 2 areas and process them separately. It's a situation like the CCIR or NAB tape eq, after a complete record + play cycle we have a linear frequency response too, but the HF maximum input level of a analog tape recorder is lower in respect to the LF maximum input level, and this "2 way" treatment should be considered in mastering phase too.

This is the main reason of FM HF lack, because in a FM receiver the maximum output level in the 15 khz area (post 75 uS de-emphasis) is 17 dB under the low frequencies, and 3,5 dB better (13,5 dB) in a 50 uS standard receiver.

[Bob Olhsson]

Simply comparing '80s music videos with contemporary ones tells the story. The old stuff frequently comes out of the TV set significantly louder than the new stuff.

The challenge is making it loud enough that it doesn't wind up in the wastebasket at a programming meeting and yet clean enough that it doesn't sound wimpy on the air.