[visanj]

Hi

I was reading about BBC Radio 3 and their live concert programs. I have a question though

If a FM Radio station broadcasts live performance/concert and if we listen using a good FM Radio then is the whole chain analog? I mean I don't see the need for ADC and DAC conversion

In that case, this should sound more natural right? or am I missing something?

 

[theveterans]

The chain you're listening is all analog. Most likely the source is digital and is run through a digital broadcast mixer (DAC) then broadcasted as FM radio waves

 

[visanj]

The chain you're listening is all analog. Most likely the source is digital and is run through a digital broadcast mixer (DAC) then broadcasted as FM radio waves

But if its a live performance then why the source should be digital? Can't they record analog form and broadcast directly?

 

[old tech]

Hi

I was reading about BBC Radio 3 and their live concert programs. I have a question though

If a FM Radio station broadcasts live performance/concert and if we listen using a good FM Radio then is the whole chain analog? I mean I don't see the need for ADC and DAC conversion

In that case, this should sound more natural right? or am I missing something?

Why would it sound more natural? Do you think your TV had a more natural picture or sound when it was analog broadcast?

 

[visanj]

I don't remember end-to-end analog tv which mostly be either live match or news. Of course picture was grainy until we tune it perfectly

anyways just a thought

 

[pinnahertz]

If a FM Radio station broadcasts live performance/concert and if we listen using a good FM Radio then is the whole chain analog? I mean I don't see the need for ADC and DAC conversion

Sorry, it's not going to be analog the whole chain. In modern FM transmission chains it's far more likely there's been a conversion to digits at some point early on. Audio processing, which every FM station must have, has been all digital for quite some time. The stereo transmission signal is digitally generated, and in modern FM transmitters, the entire modulated carrier is generated digitally. Going backwards in the signal chain, the program is most likely being mixed on a digital mixing desk, and if the concert takes place in a concert hall, the final mix will be transmitted to the studio digitally, where it will enter an digital audio infrastructure for switching, routing, and mixing, then head off to the transmitter as a digital signal.

In that case, this should sound more natural right? or am I missing something?

You're missing the fact that handling a signal digitally from mixing to transmission is far, far less subject to damage than the old analog method. There's no reason an entire analog path would sound more natural, never has been.

The chain you're listening is all analog. Most likely the source is digital and is run through a digital broadcast mixer (DAC) then broadcasted as FM radio waves

Most of the FM transmission chain is now digital up to and including the final modulated carrier. It's been this way for some time now, with benefits both in performance and stability.

But if its a live performance then why the source should be digital? Can't they record analog form and broadcast directly?

If it's a live performance, then it's broadcast without recording of any kind. Microphone signals are most likely mixed together using a digital mixing desk, and after the desk there's no reason not to handle the signal in the digital domain.

Digital mixing desks have many advantages in live broadcasting, not the least of which are stored and recallable settings and automation. Live concerts are rehearsed, and broadcasters will attend and mix the rehearsal, making notes of settings, and storing them for recall during a live performance. Digital mixing is done at 32 bit floating point internal processing resolution, if not 64 bit FP.

I believe the original question is rooted in the concept that analog signals are somehow more pure than digitally handled ones. Probably, the OP thinks that digital signals are made up of tiny digital "steps" forming a "staircase" type of waveform. This is simply not true, and never has been true. There are no steps in the output waveforms from a DAC, none at all. Because an analog signal can (and does) pick up analog noises and distortions along the way, it's been more difficult to handle a total analog signal from microphone to transmitter without damage. This is particularly true for live concert broadcasts were the mixed signal must be sent some distance to a radio studio or broadcast facility before being transmitted over the air. Before digital technology, the method was to use copper wire pairs belonging to the telephone system. Wire pairs longer than a thousand feet or so required equalization to maintain flat frequency response up to 15kHz (forget 20kHz). The longer the run, the more equalization required to restore the high end. The more equalization, the greater the noise, and the longer the run, the greater the noise pickup. It was kind of a loosing battle until good high speed digital transmission became possible. Now we can send perfectly flat 24 bit digital audio over an Internet connection anywhere on Earth that has adequate network speed, with no change in frequency response, distortion, channel separation, or noise. The audible difference is not small.

Also be aware that the transmission of a stereo signal on FM involves a few processes that are far more difficult to pull off well in the analog domain. FM Stereo handles Left and Right by transmitting the sum of the two (L+R) as a mono signal with response up to 15kHz, then transmits the difference of the two (L-R) using a modified AM system (double-sideband suppressed carrier method) with the carrier centered at 38kHz, but suppressed. The carrier is re-injected in the receiver. The L-R subchannel is far more noise-prone than the L+R signal, resulting in noise, distortion, and high frequency separation loss. Modern transmitters (actually, audio processors) develop the entire FM baseband signal comprised of L+R, 38kHz L-R, and 19kHz "pilot" signal digitally, which results in much higher separation and lower distortion. Today, all of the transmission quality limitations exist within the receiver, which is most likely a digitally controlled analog demodulator.

Please try to get your head around the idea that digital audio signals are the "pure" ones, analog audio is the type that suffers loss and distortion along the way.

 

[bigshot]

FM radio is lower fidelity than CD. That's all that really matters.

 

[pinnahertz]

FM radio is lower fidelity than CD. That's all that really matters.

Much lower. FM suffers a lot from noise, distortion, and crosstalk. It's not hard to understand why. The spectrum generated by frequency modulating a carrier is theoretically infinite with sidebands above and below the carrier. In practical terms we can ditch everything above and below the carrier beyond 100kHz away from it and not take much of a hit. When to reduce the received spectrum bandwidth, you increase distortion and throw away stereo separation. It's just physics.

If you have flat RF bandwidth 200kHz wide, you're golden. But that's not what happens in practical reception. Receivers are bandwidth limited of necessity, and bandwidth filtering alters the phase (group delay) in the pass band introducing distortions. To make things worse, the received FM signal has delayed echos of itself mixed in from reflections off buildings and terrain. These reflections introduce incidental AM on the signal which can be up to and over 100% in some conditions. Delayed signal echos (multipath) radically increase distortion, and huge amounts of incidental AM cause FM demodulators to lose lock on the carrier. It all gets pretty nasty.

FM can be pretty good under ideal reception conditions, but with that wire antenna tossed behind the receiver, that's not going to be very likely. You need a real FM antenna above ground, out in the open, and directional. Digital broadcast signals are time-stamped so that delayed reflections are easy to identify and ignore (up to a point). Separation, noise, and distortion are not a variable function, but rather the system tolerates a measure of signal corruption without audible impact until the corruption is beyond a certain point, then it just stops...and falls back to analog FM (at least that's the USA system). No, analog FM is not better than anything except analog AM. I guess it can beat cassette tape, perhaps 1/4" reel tape under excellent reception conditions.

But then, all analog transmission and recording methods are inferior to the CD. 38 years later people still don't get it. But just ask any of us who were intimately involved with analog (and in my case, FM) before digital recording arrived. We're happy to discuss the improvements.

 

[pinnahertz]

Did you know that when the CD was introduced it initially caused problems in FM transmission? Yup. FM uses a crude noise reduction method called "pre-emphasis" where highs are boosted during transmission (17.5dB at 15kHz re: 400Hz), then cut with a complimentary filter at the receiver. The system works pretty well, but it means that high volume, high frequency sounds can cause a transmitter to over-modulate, or at very least, make the signal quite hard to processes. Vinyl and tape have a rather soft high end at full output, sort of a level-dependent HF rolloff. CDs have flat response at full output, and therefore can have significantly higher HF content and high volumes. The difficulty was severe enough that one solution was to deliberately roll off the CD player response with a passive filter starting at 10kHz, down 6dB by 20kHz. Modern broadcast audio processors manage high frequencies better, so this is not the problem today that it was in 1982, but it did give us a few fits back then. I still have a couple of the little RC networks I made to cut the CD high end and make it broadcastable. The little buggers do make CD players sound more analog-ish, if you think that's a good thing.

The other problem was excessive dynamic range. Radio of all kinds is a limited DR medium by nature of reception conditions, including the environment the listener is in, such as cars. Early classical CDs were so widely dynamic that some were not broadcastable without quite a bit of effort. An extreme example, the Telarc recording of the 1812 overture couldn't really be broadcast even from the vinyl, but the CD was impossible. Real cannon shots peaking 20dB above average? How do you manage than when the listener is sitting in a room with a noise floor only about 35dB below radio volume?

Fond memories of the old days.

 

[old tech]

Good stuff @pinnahertz a fascinating history.

I read somewhere that despite the technical improvements, the sound quality of FM radio has deteriorated in the past couple decades due to less bandwidth being made available. Is there any truth in this? I remember FM radio sounding a lot better in my youth than today, but of course it could just be rose coloured memories.

In Australia, commercial FM broadcasting didn't appear until the very early 1980s. It was a big step up in sound quality over the AM radio but certainly not as good as CD quality - indeed to my ears, many of my cassette recordings (from sophisticated decks on high bias tapes) sounded better.

In your opinion, how does FM sound quality stack up against DAB+, like we have in Australia and the UK? DAB+ should sound better as it is technically in a different league but the regulator's choice of very low bit rates to save bandwidth and enable multichannel broadcasts detracts from sound quality. To my ears DAB+ generally does sound better than FM but falls short of its potential.

 

[pinnahertz]

Good stuff @pinnahertz a fascinating history.

I read somewhere that despite the technical improvements, the sound quality of FM radio has deteriorated in the past couple decades due to less bandwidth being made available. Is there any truth in this? I remember FM radio sounding a lot better in my youth than today, but of course it could just be rose coloured memories.

In all countries the FM band was initially channelized based on the bandwidth required for high quality audio given the maximum carrier deviation, typically +/- 75kHz from center. That has not, and cannot be changed. What has changed in the US is the allowable coverage contour spacing between co-channel, and adjacent channel stations. The rules have relaxed somewhat, and with the application of direction transmit antennas, it's a tighter jigsaw puzzle than ever. But reception would only be degraded in fringe areas where reception was already challenged. Basic FM audio performance capability has gotten better with advancements in the equipment that creates the signal (all digital now), and with advancements in tuners.

But what has degraded over time is the quality of the broadcast signal as a result of a raging loudness war. Radio has been in a furious loudness war far longer than recorded music has, and the tools are quite aggressive, employing quite a bit of deliberate clipping and high speed peak limiting as well as multi-band processing. Stations are now all very loud, but with less transparency than ever. Sad, but true. The FM system is capable of far better quality that you hear on the air, but it's because of a choice to fight a war that, in the end, is decided by the listener and his own volume control.

In Australia, commercial FM broadcasting didn't appear until the very early 1980s. It was a big step up in sound quality over the AM radio but certainly not as good as CD quality - indeed to my ears, many of my cassette recordings (from sophisticated decks on high bias tapes) sounded better.

In the earlier days a lot depended on the analog chain. A lot of stations tried to improve things by adding a lot of devices to the chain. In the end, simpler was better. Also, a lot was learned about how to transmit a quality FM signal. Older transmitters actually were quite band limited. It was possible to tune a transmitter to broaden it's pass band and improve the received signal, but that concept didn't appear well understood until the late 1970s, early 1980s, and wasn't always possible with older transmitters. New ones were inherently broad band, and today are really good.

In your opinion, how does FM sound quality stack up against DAB+, like we have in Australia and the UK? DAB+ should sound better as it is technically in a different league but the regulator's choice of very low bit rates to save bandwidth and enable multichannel broadcasts detracts from sound quality. To my ears DAB+ generally does sound better than FM but falls short of its potential.

I have not heard DAB+, but here in the US we have "HD Radio", a tragic misnomer, and it suffers from low bit rate as well. However, if you get a reasonably solid bit stream, the noise floor is well below average FM, the frequency response is flat without special high frequency processing (no FM pre-emphasis to deal with), and separation is nearly perfect. In those aspects, our HD1 signals usually beat analog FM. However, when you reduce the individual channel bit rate by slicing up the total bit budget into HD2, and HD3 channels, we suffer from lossy compression artifacts. Codecs have gotten a bit better since HD Radio was introduced, but it's still a problem. Stations concerned with highest quality don't bother with anything beyond HD1, and there's even a provision for dropping analog FM completely and increasing the total digital bit rate as a result. Nobody's got the guts to try it so far. HD Radio never has penetrated the market very well, most listeners are still on analog FM. The problem is paying for another radio to get the same stations you already get. Not a good seller.

DAB has one significant advantage over the IBOC system used here: it's not trying to put the digital signal in the same bandwidth and channel as the existing analog modulation. Here that's what they stupidly did here, because spectrum=money, and we're all greedy. The HD signal raises the analog noise floor, and makes mulitipath reception worse. And since the HD carrier is a fraction of the power of analog FM, digital reception has its own problems. There's a move to increase the HD transmit power, but it doesn't always work out because it causes reception problems for adjacent stations. It's just a badly done system that was pushed forward by big business without concern for the end result. Oh well.

 

[aukhawk]

In the UK at least, FM radio routing (from studio to transmitter) was the EARLIEST practical implementation of digital processing of high-quality audio (such as music) - pre-dating CDs or digital mastering in recording studios by a full 10 years. In 1972 (first year of operation of the BBC's 'PCM' system) an audio critic wrote this:

The PCM system is now giving a ratio from the output of the continuity desk [at Broadcasting House] to my own receiver of 65dB unweighted. Thus showing an improvement of 7dB over average conditions of the past. The improvement is noticeable mainly at the lowest frequencies with an almost complete elimination of all forms of hum... Although the crosstalk has usually been satisfactory the capability of the new PCM link is so amazing that at 900 Hz the transmitted crosstalk was measured in the early hours of Sunday 17th September and found to be 46dB (which includes the crosstalk of my tuner)... All these improvements have been noticed quite dramatically on relays of the last three Prom concerts...
Those capable of receiving the BBC PCM [fed] stereo radio are almost certainly hearing the finest sound quality available in the world.

That was a 13/32 system (compare with the later CD standard of 16/44).
See this link: BBC FM History

 

[Outaphase]

Might be of interest - I listen to analog FM (very selectively, like KUSC classical, or KKJZ 88.1), but these days do it with a dsp-based receiver, instead of analog componentry. No "hd" or "dab" here, just pure analog with a dinky little chip. Mind blown.

My tuners consist mostly of radios that have the Silicon Labs chips inside:

https://www.silabs.com/audio-and-radio/fm-radios

Basically all the filtering, de-emphasis, tuning and whatnot is controlled by software, converting the analog rf-spectrum to digital, processing it, and of course last mile d/a to an amplifier. Seems very quiet and pleasing - provided you are listening to a station that knows how to set their Orban's right, good source material, and so forth.

Buried somewhere in the dedicated source hardware threads is my mention of use of - get this - a multiband radio. CCrane Skywave. Or the "ssb" model, chosen because it has a slightly newer SiLabs chip. I freaked out on the performance, but would love to see Schiit or someone take this to another level, for those of us lucky enough to have GOOD analog fm still available.

 

[flibottf]

That was a 13/32 system (compare with the later CD standard of 16/44).
See this link: BBC FM History

That was quite an interresting read, thx!

 

[gregorio]

I was reading about BBC Radio 3 and their live concert programs. I have a question though
[1] If a FM Radio station broadcasts live performance/concert and if we listen using a good FM Radio then is the whole chain analog? I mean I don't see the need for ADC and DAC conversion
[2] In that case, this should sound more natural right? or am I missing something?

1. As others have covered well, today and for many years, pretty much the "whole chain" is digital, with only the transmitted FM signal itself being analogue. However, prior to around 20 years ago more of the chain was analogue and going back in time, more and more of the chain was analogue until we get back to around the mid/late 1970's where the whole chain was analogue with the exception of the signal distributed from Broadcasting House to the various transmitters. Live Radio 3 concerts were mixed on analogue mixing desks up until around 20 or so years ago, when large format digital desks (capable of handling a large number of mic inputs) matured, although some of the outboard units were digital, reverb being an obvious example (going back to the early 1980's).

2. It's really the opposite; you're not missing something, you're adding something! As digital audio technology was rolled out to the public, there was pretty much universal consensus: The general public certainly preferred it, the engineers and objective measurements all confirmed it and even the aficionados and Hi-Fi press sang it's praises. Without question digital was better, a lot better! Even to the point that the word/logo "Digital" became a dominant advertising moniker and inventions such as the SPARS Code were employed to imply more digital processing than was usually the case.

Analogue didn't get appreciably better and digital didn't get appreciably worse, so what changed? Others will remember better than me but over the years more and more of the aficionado and Hi-Fi press diverged from the engineering/scientific facts, as highlighted by Carver in the 1980's (with regard to amps). Audibly perfect digital audio conversion/reproduction was available even in the 1980's but during the 1990's even cheap consumer units started to able to achieve this feat. That left audiophile manufacturers of digital equipment with no justification for their far higher price points. So they didn't really have a choice if they wanted their companies to say in business, they had to invent something new that was supposedly better (than CD). As you obviously cannot get anything audibly better than audibly perfect, the only solution was advertising designed to make audiophiles believe that CD audio was not in fact audibly perfect. The big leap I recall was at the turn of the millennium with the introduction of Sony's SACD, which advertised a "more natural" analogue sound and was based on the false assertion of the "stair-step" output of PCM digital audio (and Sony was well aware it was a false). Eventually CD audio was rebranded (in the audiophile world) as standard or 'low-res", while the new, supposedly better formats were branded as "hi-res" and ALL audiophile digital equipment manufacturers (and content distributors) again had a justification for their far higher price points!

The something that you're adding is therefore 20 odd years worth of FALSE audiophile marketing! After 20 years of shills and reviewers paid to promote this false marketing and of course the target audiophiles themselves being biased by it (and posting their biased "impressions" on sites like this), it has become so ingrained that many/most audiophiles simply take it for granted, without question. Or to put it another way, what you're missing is what even the audiophile world knew 20+ years ago (plus the proven objective facts), digital is far more accurate (and therefore more "transparent" and "natural") than analogue!

G