The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Radio Broadcast Levels? Dynamic Microphones
Old 9th October 2011
  #31
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by JohnOmix View Post
I mix promos for tv. From what I've heard there is some limiting to make sure things don't spike and are below highest spec. But please tell me they don't use these multiband radio-type processors for broadcast?
They most certainly do.

Alistair
Old 9th October 2011
  #32
Motown legend
 
Bob Olhsson's Avatar
 

Quote:
Originally Posted by UnderTow View Post
I'm guessing that might be the work of the phase rotators but I could be wrong...
More likely the pre-emphasis reacts to the distortion.
Old 9th October 2011
  #33
I work in production at a radio station. Every commercial we air has to be recorded into the playback/automation software in real time. Because of this, it is part of my job to manually normalize volume differences between radio spots we receive or produce. This is all done before the audio undergoes any further processing/normalization. I try my best to match rms levels between spots and leave a healthy amount of headroom so the optimod will process the files in a similar way.
Old 9th October 2011
  #34
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by Bob Olhsson View Post
More likely the pre-emphasis reacts to the distortion.
How is it reacting to the distortion? It is just a high frequency boost. It isn't signal dependant. Of course anything above a certain level gets limited/clipped but if the two voice clips sounded subjectively the same, there shouldn't be much difference in the high frequency content.

On the other hand, if EQing and other processing has already introduced phase shifts to some of the voice clips but not to others, they will react differently to the phase rotators in the radio processing. In the most extreme case (where the one clip benefits maximally from the shift and the other loses maximally from the shift) there could be a subjective level difference of 6 to 8 dB due to the phase rotators.

But maybe I'm missing something about the pre-emphasis. Care to explain?

Alistair
Old 9th October 2011
  #35
Motown legend
 
Bob Olhsson's Avatar
 

Quote:
Originally Posted by UnderTow View Post
... maybe I'm missing something about the pre-emphasis. Care to explain?
The pre-emphasis is applied before the broadcast processing.
Old 9th October 2011
  #36
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by Bob Olhsson View Post
The pre-emphasis is applied before the broadcast processing.
Not necessarily but even if it is, that still doesn't clarify anything.

Alistair
Old 10th October 2011
  #37
Lives for gear
 

Quote:
Originally Posted by GoldenOne View Post
I work in production at a radio station. Every commercial we air has to be recorded into the playback/automation software in real time. Because of this, it is part of my job to manually normalize volume differences between radio spots we receive or produce. This is all done before the audio undergoes any further processing/normalization. I try my best to match rms levels between spots and leave a healthy amount of headroom so the optimod will process the files in a similar way.
What rms level do you aim for?
Old 10th October 2011
  #38
Gear Guru
 
Brent Hahn's Avatar
 

In my world, it's all politics.

Our clients (ad agencies) want to go into approval presentations with loud spots. They also insist that once a spot is approved, it goes on the air unchanged. So that's what they get.

We don't smash the crap out of stuff, though -- our waveforms still look more like waveforms than boxcars.

And for what it's worth, I've never heard any of our stuff get horribly mangled by Optimods or the like. Some stations in our market (LA) do odd things to the m/s proportions, but nothing a lay person would ever notice.
Old 11th October 2011
  #39
Lives for gear
 

....They also insist that once a spot is approved, it goes on the air unchanged....

Which "air" are they going on? If it's commercial AM or FM "air", I've got news for you: they -are- being changed by the on-air processing.
Old 11th October 2011
  #40
Gear Guru
 
Brent Hahn's Avatar
 

Quote:
Originally Posted by tpad View Post
....They also insist that once a spot is approved, it goes on the air unchanged....

Which "air" are they going on? If it's commercial AM or FM "air", I've got news for you: they -are- being changed by the on-air processing.
Not news. Like I said, it's politics.
Old 12th October 2011
  #41
Lives for gear
 
Airon's Avatar
 

Does ABC still play with Orban knobs ? Some stations in the Netherlands use processors to stay within EBU R128 specs.
Old 13th October 2011
  #42
Lives for gear
 

My understanding is that the major US TV networks prefer to send the audio out the way it comes in. That's why they are imposing on the content providers to adhere to the ATSC specs, so the outgoing levels are legal.
Old 16th October 2011
  #43
Here for the gear
 
Jose Raposo's Avatar
 

Hello everyone,
I am new here,
it seems that this world out there is still no a standardization of the delivery of audio files for radio broadcast.
In Portugal a few years already we deliver to 320 kbps mp3, 0 db
Most of the automation systems for broadcast does not have to convert to another format, although previously had to be converted to MPG2. As the sampling rate I think that both 44.1 and 48 kHz, accept. Normally I sen at 48 khz.
All Broacasters use limiters or other digital devices, like Orban and of course often are not calibrated the best way, but what happens to our file also happens to other studios and it happens to all the music and program.
Let us say that our studio all is well, without distortion and with good quality, then the work was well maintained.

Best Regards,

José Raposo
Sound Station, Lisbon, Portugal
Old 18th October 2011
  #44
Yikes, conflicting answers here. So how do we find out for sure about broadcast and cable?
Old 21st October 2011
  #45
Gear Head
 

I supply radio programmes as wavs. Usually peak at around -3dB, but once the radio station receives them, they immediately normalise all the files, so it really doesn't matter.
Old 13th March 2012
  #46
Gear Nut
 
MrTinkle's Avatar
 

Quote:
Originally Posted by jazzdog View Post
I supply radio programmes as wavs. Usually peak at around -3dB, but once the radio station receives them, they immediately normalise all the files, so it really doesn't matter.
We'll since my first post I can sure tell you they love normalising to 0db! From my point of view it's blasphemy because every thing gets normalised then compressed some more before going out. So yes, definitely stick to -3 or even -4db.
Old 13th March 2012
  #47
Lives for gear
 
nott's Avatar
 

well, the most files are around -3 peak. rarely -1 and sometimes -5.
go with -3db!!!
Old 14th March 2012
  #48
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by MrTinkle View Post
We'll since my first post I can sure tell you they love normalising to 0db! From my point of view it's blasphemy because every thing gets normalised then compressed some more before going out. So yes, definitely stick to -3 or even -4db.
If it gets normalized anyway, why does it matter if you deliver a file that peaks at 0 dB or -3 dB?

Alistair
Old 14th March 2012
  #49
Gear Nut
 
MrTinkle's Avatar
 

Quote:
Originally Posted by UnderTow View Post
If it gets normalized anyway, why does it matter if you deliver a file that peaks at 0 dB or -3 dB?

Alistair
Well I said that too, but at first I was sending them files that were mixed as I would for music and mastered to be at 0 db. Then they started asking me to send them at -3db.

They seem to batch normalise things instead of doing single files and the normaliser they use (on the play out software) does not measure the difference between loud and quiet tracks and adjust them correctly.

So instead of louder tracks getting reduced from 0db to -3db and quiet tracks getting pushed up to -3db. It just makes everything as loud as it can (like a heavy set limiter) then brings it all down to -3db. So tracks that are at 0db get crushed and start to sound fuzzy.

It seems backwards and does confuse me ...They also use a lot of heavy compression on the output so it makes sense to leave headroom for things. That's from my previous experience anyway.
Old 20th May 2012
  #50
Broadcast RMS/Peak

Quote:
Originally Posted by seanmccoy View Post
Stations use different chains, but virtually all of them use an Optimod or something similar that will make the average levels about the same no matter what. Bob Katz's book includes a section on this, which also quotes the referenced article. Based on all that, it would seem the safest method (i.e., the one that is likely to do the least damage) is to mix with the RMS levels down a bit to avoid hypercompression at either the mixing or broadcast stage, but it's okay to allow peaks to hit near full scale. I use K-12 for all broadcast mixes now, and try to keep the average levels in the upper yellow as much as possible.
I have Bob Katz book so I checked it out. It says -10dB for broadcast.
My question is would I master my track as normal (for example limiting to -0.1dB and having an RMS of around -10dB) Then normalize to -10dB peak?
Old 20th May 2012
  #51
Gear Guru
 
Brent Hahn's Avatar
 

Quote:
Originally Posted by Ecktronic View Post
I have Bob Katz book so I checked it out. It says -10dB for broadcast.
He means for TV, not radio.

Short of clipping, make it whatever peak level you want.
Old 30th April 2017
  #52
Lives for gear
Rez for 2017.

I hit stations with -3dB peak and -20LUFS 320 MP3. If I send them too loud/too compressed they tend to sound like total crap when aired by most FM stations. AM seems to be more forgiving but the whole format sounds like garbage anyway.
Old 30th April 2017
  #53
Lives for gear
FWIW, two of the radio products I worked on just won Best of Show at NAB:

Stations process. Some of them (or, some of their CEs and PDs) say "we don't let processing interfere with quality." But when the GMs see a difference in processing result in an increase in profits, they have to go for the bucks... and CEs or even PDs can't change that.

Instantaneous loudness attracts listeners. Lack of distortion keeps them. The goal, among all of us who face ourselves honestly in the morning, is to come up with processing that attracts listeners, doesn't seriously screw the quality of the sound, and lets station owners turn the knobs down from "11" towards settings that are more benign. We believe some stations still do that.

With each new hardware/software generation, we're getting better at making sounds that jump out of the radio but don't fatigue the ear. No guarantee stations are using that hardware, however: that's all over the map and there are strong competing pressures. (FWIW look up "Voltair": one of the radio products I worked on.)

Bottom line: In the US producers have no control over what the station will do to your spot. Unlike TV, where there's some semblance of a standard, radio is raw capitalism.

Heck... a few years ago I did a voice-only spot for a national chain that, between the script, performance, editing and processing had amazing impact. One of our local stations ran random music underneath, killing the spot but following a PD's dictate that "there shall be music at all times". We got a make-good, but only because we spotted it and complained.

--

Another thought, though: I learned in a career of agency spots that distortion, timing, and spectral distribution have a lot more to say about perceived loudness than VU (or LKFS) meters. There are still things you can do to sound clean, have an impact, and don't get too damaged by aggressive processing.
Old 30th April 2017
  #54
Gear Addict
Lads this is a 5 year old thread.
Old 1st May 2017
  #55
Gear Head
 

This thread is indeed 5 years old, but it is fascinating to look back, what has happened during the last 5 years:

CD has died.
DAB+ networks are phasing out FM.
End of loudness wars in music = less extreme processing of radio spots.
Ubiquity of 3G/4G networks enabling mass consumption of music streaming services (major streaming networks are loudness normalized to -16LUFS), and pretty much everyone setting their own comfortable levels.
Most importantly - the "loudness processing" is shifting to integrated loudness normalization & metadata based processing, with minimal actual damage to your audio, compared to "Cranked to 11" Orbans & Omnias.

I think, I cut & mixed my last radio spot some 4 years ago. The last delivery specifications I received, wanted me to have at least 8 dB of dynamic range! Talk about a major shift, after years of making squared pancakes, with 8db RMS average. It's like some playout tech guy @ some broadcast network decided to go on a solo crusade against horrible sounding promos in his station
Old 1st May 2017
  #56
Lives for gear
Zanmixr, there have been a lot of changes. But broadcast radio, as far as some GMs and consultants are concerned, is still a loudness war. I work on one of the major processor lines, and we have to constantly remind users that the knobs can be turned down as well as up.

On the one hand, we're learning ways to increase perceived loudness and actual mod percentage without horrible distortion and flat-topping. But there are still a lot of GMs who come from an era where a little more distortion was the price of better ratings, so they push our cleaner tools even harder. [Most GMs still seem to have come from sales, rather than from programming or engineering.]

On the other hand: for a lot of formats, squashing -- and even distortion -- can help the Nielsen PPM system load more watermark IDs into the program stream. So it continues to be a battle of quality vs crunch. [This applies to bigger markets and some countries only. PPM isn't universal.]

Is lack of quality what's killing radio overall? Hard to say. Radio was a more profitable business back in the days of top-40 squeeze, but a lot of things have changed.
Old 3rd May 2017
  #57
Lives for gear
Also, with Spotify and Pandora ads being as big or bigger than terrestrial there a sweet spot for compression seeing as FM will squash the crap out of your spots where internet might not even touch them.

I engineer a couple of the highest profile ad campaigns in the US, that you all might have heard. The spots go out to everything and it's a one shot delivery so I can't send different versions for different formats
Old 28th April 2018
  #58
Here for the gear
 

Quote:
Originally Posted by MrTinkle View Post
Wow, lots of info

I was worried that if I mixed everything to -0db and they limited the hell out of it after it would sound crumby. I think the play-out software stations use is called Myriad? Not too sue though...

Anyway, cheers for the responses fellas. I'll just set the bar to -2db just to be on the safe side, if they don't moan then I'm happy
Yeah, you can mix it relatively soft and dynamic - the station's processor chain will steamroll it anyway. The less processing you do, the cleaner it will sound out of the transmitter, even post-Optimod.
Old 29th April 2018
  #59
Lives for gear
The less processing you do, the cleaner it will sound out of the transmitter, even post-Optimod.

I don't quite agree. In my experience, the cleaner it sounds in your studio, the cleaner it'll sound after Omnia*.

Think of the station's processor as a medium speed multiband AGC followed by multiband limiting. You want to control the faster things in your mix -- like voice-over envelopes -- so the AGC has a good chance to control the overall level. You definitely don't want peaks that are so far above the average that they hit the limiter, except in very rare instances.

Clarity, however, is important because any station processing will tend to emphasize any distortion in your product. So it's a constant battle to find the processing recipe that makes your spot the loudest thing in the break, but doesn't break itself when it gets to the airchain. The most important settings in your studio, after the EQ is correct, are the separate threshold / ratio / time constants for each of your compressor's bands.

*Telos' Omnia is one of the major competitors to Orban's Optimod. I worked for Orban until 2000, when they changed ownership. Now I work for Telos. The third player is Wheatstone. All these products have different 'secret sauce' that may include predistortion, stereo enhancement, stereo un-enhancement or some SSB to protect against multipath for FM, asymmetrical clipping for AM, and a bunch of other things that if I tell you, I have to kill myself.
Old 29th April 2018
  #60
Here for the gear
 

Quote:
Originally Posted by Jay Rose View Post
The less processing you do, the cleaner it will sound out of the transmitter, even post-Optimod.

I don't quite agree. In my experience, the cleaner it sounds in your studio, the cleaner it'll sound after Omnia*.

Think of the station's processor as a medium speed multiband AGC followed by multiband limiting. You want to control the faster things in your mix -- like voice-over envelopes -- so the AGC has a good chance to control the overall level. You definitely don't want peaks that are so far above the average that they hit the limiter, except in very rare instances.

Clarity, however, is important because any station processing will tend to emphasize any distortion in your product. So it's a constant battle to find the processing recipe that makes your spot the loudest thing in the break, but doesn't break itself when it gets to the airchain. The most important settings in your studio, after the EQ is correct, are the separate threshold / ratio / time constants for each of your compressor's bands.

*Telos' Omnia is one of the major competitors to Orban's Optimod. I worked for Orban until 2000, when they changed ownership. Now I work for Telos. The third player is Wheatstone. All these products have different 'secret sauce' that may include predistortion, stereo enhancement, stereo un-enhancement or some SSB to protect against multipath for FM, asymmetrical clipping for AM, and a bunch of other things that if I tell you, I have to kill myself.

I think we are saying the same thing, just in different ways. Of course in production you want clean sound and decent(not slammed) levels, before the product gets anwywhere near the TX chain.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump