Help! Audio Processing for a Classical FM

dx7 · Mar 3, 2009

I am doing some programming consulting for a full-power, full-time, non-commercial classical FM station. I am not an engineer so any help from this Board would be much appreciated.

Here's the problem -- even when a symphony orchestra is playing over the air at full volume from a CD, the station sounds much quieter than the other stations on the FM dial. Listeners complain that when they punch their car radio button to our station from another station, they have to crank up the volume considerably even during loud symphonic musical passages.

We do NOT want to compress the sound to compete with the pop music stations, and we do appreciate the wide dynamic range of classical music. We understand that quiet musical passages should not sound as loud as the louder passages.

Still, the GM wants to know if there is a way to make the station sound louder overall without the heavy compression of pop music radio.

The station uses an Optimod 8500 and has a CE who is reluctant to change the compression scheme from the 2 band-AGC classical music preset that came with the Orban.

Thanks in advance.

dx7

whit979 · Mar 4, 2009

For what it's worth, the Classical 5-band+AGC setting in the 8400/8500 series sounds very smooth. It is one little notch in loudness above the 2-band settings. That doesn't really address your modulation issue, but it might be something worth hearing anyway to consider as a 'happy compromise' setting.

gtomas · Mar 4, 2009

First consider that what you want to do is valid and the right way to go. Talk to the CE, hear his reasoning. On every few listeners that called in, consider that there might be a dozen others that actually prefer the wide dynamic range you are providing your listeners at the moment. The classical music audience is a very peculiar and sensitive one.

Having said that, if you do decide to provide more levelling/compression follow Whit's advice. Give it a go for a week, see how you all feel about it.

Regards,
Goran Tomas

rrsounds · Mar 4, 2009

dx7 said:
I am doing some programming consulting for a full-power, full-time, non-commercial classical FM station. I am not an engineer so any help from this Board would be much appreciated.

Here's the problem -- even when a symphony orchestra is playing over the air at full volume from a CD, the station sounds much quieter than the other stations on the FM dial. Listeners complain that when they punch their car radio button to our station from another station, they have to crank up the volume considerably even during loud symphonic musical passages.

We do NOT want to compress the sound to compete with the pop music stations, and we do appreciate the wide dynamic range of classical music. We understand that quiet musical passages should not sound as loud as the louder passages.

Still, the GM wants to know if there is a way to make the station sound louder overall without the heavy compression of pop music radio.
The station uses an Optimod 8500 and has a CE who is reluctant to change the compression scheme from the 2 band-AGC classical music preset that came with the Orban.

Thanks in advance.
dx7

Assuming that the Optimod's maximum modulation level is set properly, the processor is preserving peaks at the expense of loudness. Not that there's anything wrong with that, but some might see it as a competitive disadvantage. I believe this is the situation you find yourself in.

The occasional classical 'fortissimo' CAN be as loud as the CHR is all the time, if you drive the peak processing hard enough. The Optimod has controls to do exactly this, with little downside if you are judicious.

Source-to-source consistency and reducing average level variations can be considered as a separate issue from peak loudness. There are of course numerous choices, including being more aggressive with the Optimod's leveling. If there are board operators, you might also want to investigate their operating practices, which can have a huge influence on absolute levels and consistency.

A number of major market classical stations are using our AGC for 'artistic' loudness consistency, followed by a processor with sophisticated peak control (which you have already) to set the maximum peak loudness. I would say that the last thing you want to do is make the processing obvious. So that means sticking with relatively slow release times, careful setup of release gating, and using a sensible balance of limiting versus clipping.

This is not a casual matter, and can take some attention and close listening over a very long period of time. I'm sure most will agree that whatever tools you use, the key will be a make a concerted effort to balance the loudness/consistency advantages of more processing against the downside, which is to perceivably alter the dynamics of the material.

Good luck, and let us know how it works out!

-- David

fm-engineer · Mar 4, 2009

whitfm said:
For what it's worth, the Classical 5-band+AGC setting in the 8400/8500 series sounds very smooth. It is one little notch in loudness above the 2-band settings. That doesn't really address your modulation issue, but it might be something worth hearing anyway to consider as a 'happy compromise' setting.

I provide engineer services for a several NPR and university non-coms. The Classical 5B+AGC preset is a very common setting for NPR stations. The PD's often play with the less-more setting, but I would stick with the classical presets. Some purist PD's prefer the 2B over the 5. I also agree with Dave. There are some major market non-coms running the Ariane Sequel into the 8500 set up for limiting only classical 2B or 5B. You will find purists who do not like RMS detection or matrix processing. What you won't be able to argue is the sound which is quite amazing. If I chose an Ariane to do the AGC work, I would use the stereo purity setting and would stay away from the matrix processing. Of course, just the opposite would be suggested for other formats where I do often employ the Ariane in matrix mode.

OKCRadioGuy · Mar 4, 2009

I don't know about Classical purest types, but the Ariane we are using for a three hour show we send up to a local FM (in matrix mode) sure sweetens up the audio compared to when audio is played directly in THEIR studio. I love the sound! It was bought for leveling but honestly it gives a effect that makes the audio really stand out.

Sgeirk · Mar 4, 2009

I vividly recall when our NPR classical station replaced a McMartin single-band limiter with an Optimod 8100. It was 1985.

We had at least 5 or 6 complaint calls within a week. Hardcore classical listeners are very hardcore.

wgliradio · Mar 4, 2009

The 8100 still has the best 2 band structure. I never understood why Orban never properly ported it over to the digital domain, but an Ariane in front of a 2200 or 2300 is not the same as one in front of the 8100.

Avoid the mistake alot of people do with classical music... setting the gates for a high threshold. Just because it's classic doesn't mean everything should be gating all the time. The 5 band 8500 with strategic coupling and an Ariane out front in discreet mode with a slower IDR is where you want to be. If your 8500 is barefoot, play with the release on the AGC so that it is slow averaging, but don't make it gate "right away". Your best bet is probably no more than 5bd of gain reduction in the AGC on the loudest program material.

oldiesstation · Mar 4, 2009

Unless they're on an HD platform, a nice rebuilt 8100A/Ariane would be VERY nice.Can't get more purist than that.Listeners would love it.IMHO

LA_Guy · Mar 5, 2009

A friend of mine took a single Texar Prism, took out two of the compressor cards and went into the other two wideband as a stereo device. This was used on WCRB/Boston in front of an 8100 for years-during some of their highest ratings ever. The big advantage of this is the 'window' type of level detection the original Prism used.

Needless to say, it sounded GREAT! It acted like a gentle hand on the pot slowly turning things up and down as needed.

rrsounds · Mar 5, 2009

LA_Guy said:
A friend of mine took a single Texar Prism, took out two of the compressor cards and went into the other two wideband as a stereo device. This was used on WCRB/Boston in front of an 8100 for years-during some of their highest ratings ever. The big advantage of this is the 'window' type of level detection the original Prism used.

Needless to say, it sounded GREAT! It acted like a gentle hand on the pot slowly turning things up and down as needed.

I like this idea, and would have been interested to hear it. Though I suspect a two-band (with 'rubber-band' linking) version might be more to my liking.

At some point when I was designing the original analog Ariane I decided that, in the final production version, we would put each Ariane Processing Engine ("APE") on its own card. The idea was that we might re-purpose the cards in other configurations, such as the interesting thing your friend did with the Texar cards.

I was even going to call one such project the "Ariane Solo" since it had just one band.
This never came about mostly due to the realization that the basic box with support (I/O, power supply, chassis, and metering) was the bulk of our cost. Therefore in small quantities we couldn't sell a stripped-down Ariane for a whole lot less than a complete version.

While some people might find a one-band or two-band Ariane really useful, such is the fate of niche products. Unless you can sell LOTS of something, the investment-versus-return may be a bit too steep to be financially practical.
And of course, the viability of analog products these days is even sketchier. :'(

FWIW, the Ariane Sequel has the ability to link bands together that allows consolidation of bands' control to what is essentially a two-band version. However, you DO pay for all eight bands... ;D

I still have a bunch of APE cards....

--David

ChiefEngineer · Mar 5, 2009

We carried the WCRB network a few years. Unprocessed and clean from the Satellite. A pleasure to listen to in the shop or control room. This is a RADIO board however. Discussion of this has been about processing and not about the real world.

YOU ARE ALL MISSING THE POINT.

Problem is most Classical stations are fringe signals. And even good signals end up being fringe signals because of signal to noise and overall loudness. We had no loudness except at the peak of the 1812 overture. We tried everything and kept geting the same complaints. (In the City Grade contour.)

And don't trust radio people because they have been taught loudness. Radio people manage radio stations. It may be a classical station but these are still lifelong radio people. "Why aren't we as loud as the other station in town? Regardless of format.

100 percent is like the weather forecast. 100 percent chance of rain today - in the 250 mile by 250 mile area the NWS affiliate serves. This also means there is a 99 percent chance there won't be rain - in the area served.

You will have 100 percent loudness at least 1 percent of the time. Changing format was the best thing that ever happened to our loudness. And we changed to talk. Even voice peaks are more consistent that classical music.

The constant changes in volume are only acceptable in an area where the signal to noise is constant and unchanging. The subtle orchestra presentation is not a radio item. Listening at home is one thing, in a car another.

Ultimately the changes in volume were dedicated to the purist who also complained about the audio signal in an area where the signal was good. In the end the station changed format brecause the hoopy de doo classical listeners were so fickle and it wasn't worth the time and effort.

Convince your staff to do this now rather than challenge insanity. Car talk or all things considered or bob and tom beat classical music from the hell you face trying to please people who will never be pleased. based on the few classical stations left in the US this is the trend and not just a casual observation.

TO SUM - It doesn't matter what processing you use. It will be wrong to someone. Despite the best quality and lack of processing that pleases the audio purist, the manager won't like the loudness compromise. the regular classical listener expects the signal to be as loud in the car as station x is at home despite the noise from the road and other cars.

Then consider the age of people lsiteningto the music. is their hearing what it used to be and how does this affect the signal?

rrsounds · Mar 5, 2009

ChiefEngineer said:
We carried the WCRB network a few years. Unprocessed and clean from the Satellite. A pleasure to listen to in the shop or control room. This is a RADIO board however. Discussion of this has been about processing and not about the real world.

YOU ARE ALL MISSING THE POINT.

Problem is most Classical stations are fringe signals. And even good signals end up being fringe signals because of signal to noise and overall loudness. We had no loudness except at the peak of the 1812 overture. We tried everything and kept geting the same complaints. (In the City Grade contour.)

And don't trust radio people because they have been taught loudness. Radio people manage radio stations. It may be a classical station but these are still lifelong radio people. "Why aren't we as loud as the other station in town? Regardless of format.

100 percent is like the weather forecast. 100 percent chance of rain today - in the 250 mile by 250 mile area the NWS affiliate serves. This also means there is a 99 percent chance there won't be rain - in the area served.

You will have 100 percent loudness at least 1 percent of the time. Changing format was the best thing that ever happened to our loudness. And we changed to talk. Even voice peaks are more consistent that classical music.

The constant changes in volume are only acceptable in an area where the signal to noise is constant and unchanging. The subtle orchestra presentation is not a radio item. Listening at home is one thing, in a car another.

Ultimately the changes in volume were dedicated to the purist who also complained about the audio signal in an area where the signal was good. In the end the station changed format brecause the hoopy de doo classical listeners were so fickle and it wasn't worth the time and effort.

Convince your staff to do this now rather than challenge insanity. Car talk or all things considered or bob and tom beat classical music from the hell you face trying to please people who will never be pleased. based on the few classical stations left in the US this is the trend and not just a casual observation.

TO SUM - It doesn't matter what processing you use. It will be wrong to someone. Despite the best quality and lack of processing that pleases the audio purist, the manager won't like the loudness compromise. the regular classical listener expects the signal to be as loud in the car as station x is at home despite the noise from the road and other cars.

Then consider the age of people lsiteningto the music. is their hearing what it used to be and how does this affect the signal?

That's quite a rant, ChiefEngineer!

Personally, I'm not quite so cynical about Classical Radio as you appear to be.

1) In my opinion, it DOES matter what processing you use. While we can all agree that someone won't like it no matter HOW it is set up, does that mean we throw up our hands and say 'I give up!'? I don't think so! We're not paid to give up, we're paid to solve problems as best as we can. Compromise will be necessary. I can deal with that.

2) And the 'real world' IS processing. Most classical music stations have some kind of processing; the question is how audible it can/should be to meet the requirements of those who are responsible for its long-term success.

3) Not sure why radio people shouldn't be in charge of radio stations.

My wife Ariane is a violinist in an orchestra. I hear her 'band' live pretty often, and I hate it when people cough during a concert. HATE it! Kind of like radio interference, it distracts from the listening experience. I'm sure some people complain to the management about the coughing. Some patrons probably never come back. Nevertheless, I don't think we'll see Ari's orchestra going 'all-talk' anytime soon.

--David

PS: I think the reason the Classical format is dying is NOT because no one could figure out how to process it for Radio! It's because the main demographic of those who love classical music is simply getting old.