Entercom has done a great job with processing it's stations in the Portland market. However, I've been hearing some MP3-like song files on 105.1 and 97.1. R.E.M. "Stand" and Stone Temple Pilots "Interstate Love Song" are some examples. It's only the older songs that have quality issues.

Entercom's former VP/Engineering Marty Hadfield banned ANY compression for songs stored in automation systems. Now that may have changed with Marty's departure but I doubt that's what causing the reported "quality issues".

You would be surprised how many people in radio don't understand the concept of "Broadcast Quality."

Actually, the "broadcast quality" of a 128K MP3 is lots better than the "broadcast quality" of a 48K Starguide satellite feed.

I was shocked at the artifacts I heard on XM Sirius's talk formats. Must be the codec is optimized for music...

Rule #1 in the digital age: Bandwidth is a valuable commodity not to be squandered on quality under any circumstances.

With Terabyte drives available for under $200 there is no excuse for data compression of audio files.

XM really pumps up the data compression on their talk formats. I would suspect less than 32kbps. Add to that the fact that the talk programs are being recompressed after arriving via satellite to XM for distribution.

Ever try recompressing a 128k MP3 down to 56?

Most satellite programming (Premiere, ABC, WW1) has migrated to new digital receivers which use AAC coding. Starguide is now obsolete... and was not worse than a 128K MP3 (unless someone recompresssed it) Starguide used MPEG Layer 2.

> Rule #1 in the digital age: Bandwidth is a valuable commodity not to be squandered on quality
> under any circumstances.

Preach it, brother!! That is the true business meaning of the "digital revolution." Bandwidth is now a commodity, and quality tradeoffs with financial consequences can easily be made and subsequently revised with ease.

Well hopefully this has been brought to the attention of Dan Persigal because mp3 song files are absolutely uncalled for!

I love the term most certainly born in a marketing department somewhere..."digital quality." BlueRay is digital quality. So is YouTube.

" BlueRay is digital quality. So is YouTube."
So is a light switch.

But remember, MPEG audio files, like MPEG 1 layer 3, can be made such that they can sound pretty much indistinguishable from the original and thus "broadcast worthy".

You need a high-quality codec like Lame for the best results. Not the Freunhofer system or Apple I-Tunes (the latter of which, from my experience, should *never* be used to compile MPEG 1 Layer 3 files of anything but speech!)

Next, you want to upsample the source WAV file to 48000 Hz (the highest sample rate in which a standard layer 3 file can be coded; reportedly layer 2 can go higher yet but I haven't tried it.) And last but assuredly not least, NEVER compile music, especially if you intend to broadcast it, at anything lower than 160 KBPS! You should code *everything* at 320 KBPS, constant bit rate!!

Conclusion: It would be possible to produce a "broadcast quality" MPEG audio file and it certainly is. Whether or not people would actually want to put any work or effort into doing so is another question! Yes, it is far easier to just drop a CD into the system, fire up Apple I-Toonz, rip it all to M1L3 at 128K, stick it all in a directory somewhere on the server and call it done. But you sacrifice a huge degree of quality using this method, and since hardly any effort went into making it, I don't know how anyone could be proud of it. Generally speaking, the main rule of thumb when coding MPEG 1 audio files of any layer, that (by my personal observation) seem to be all too often forgotten, is: "Bigger is better".

Just my $2.00 worth........

>>You need a high-quality codec like Lame for the best results. Not the Freunhofer system or Apple I-Tunes<<

That's especially hilarious considering Fraunhofer (the correct spelling) invented the MP3 format. But even with that crack, the post lost all its cred with that "upsampling to 48kHz" suggestion. That also gave me a good laugh.

>>With Terabyte drives available for under $200 there is no excuse for data compression of audio files. <<

But new drives are only part of the cost of "upgrading" a music library from compressed to linear.

There's the labor involved in re-recording/re-ripping of the source material, setting trim/sec tones and migrating the new library to the automation system and music scheduler database. Like stations trying to slash costs to the bone are going to drop coin for those hours. Uh-huh.

Just converted an entire station from MPEG1 Layer 2 to Uncompressed .WAV and it did not take as long as you would think.

Had a source of known good quality .wav files, put them on a portable drive and file transferred them directly into the automation system overwriting the original files. The overwriting took about 3 hours for the full library.

In most cases the songs were the same version and the old trim/tone data was able to be used on the new file.

Sound quality is noticably cleaner.

"That's especially hilarious considering Fraunhofer (the correct spelling) invented the MP3 format. But even with that crack, the post lost all its cred with that 'upsampling to 48kHz' suggestion. That also gave me a good laugh."

Why? How is that hilarious? I didn't mean it as such.

1lossir, have you even heard/of the LAME system? It is practically all I ever use any more, especially for music. Ask some of the other users on this board; they'll probably tell you pretty much the same thing. Hell, ask Alfredo_t or Missing_kskd!

Probably the only thing I ever use the FHG system for nowdays is constructing recordings where quality is not as high of a concern: (1) speech and (2) on-hold music for playback over the telephone.

Instant enlightenment: http://en.wikipedia.org/wiki/LAME
More instant enlightenment: http://lame.sourceforge.net/

Okay, you don't like the upsampling idea? Don't do it then. Keep it at 44100; according to my tin-can ears they both sound the same. That is only the procedure/technique that I use if I need to program some of my older material, made waaaaaaaaaaaaaaay back when I was still a budding ignoramus mastering everything at 44100 (I've since learned to record everything at 48000 and keep it up there.) Your technique may be different. But being insulting about it really serves no purpose.

My post is merely a suggestion, based on what I have read in above posts in this thread (and previous posts based on similar topics in previous threads) to those who insist on using, or have to use MPEG for any number of reasons: programmer is stuck in his ways, corporate mandate states that MPEG must be used, technical limitations, etc. Probably not everybody is going to upgrade their system right now because of the economy.

I am not arguing PCM's superiority to MPEG--by my experience it is, and keeps proving itself as such. The one and only reason that I, personally, am stuck using MPEG 1 layer 3 is because I've yet to encounter a DVD disc player or pocket CD player that will run WAV files off a CD- or DVD-ROM disc. (I make background music for people, mastered in linear PCM, then coded in 48000/320 CBR stereo M1L3 files constructed in the LAME system. Then I put four hours of it on CD-ROM discs, which the end user plays on a commercially-available DVD Video player connected to his stereo or PA amp.) If I could find one such player--not a Blue Ray machine, I can't afford one of those right now--this situation may be different.

Uh-huh.

I don't understand why up-sampling 44.1 kHz source material to 48 kHz prior to compression would produce better results. However, if you are making a recording, the higher the sampling rate, the better. 44.1 kHz should be reserved for circumstances where you are burning audio CDs.

The LAME encoder is the only one that I use today, although I am not an expert on all of the options that are available. When encoding music, I turn on the variable bitrate encoding and set the quality to the maximum (or near-maximum) value available. I am not at my Linux computers right now, so I can't consult the command-line help.

One thing that I have noticed is that even "professionally" produced MP3s sometimes have room for improvement. Case in point: my brother gave me a mp3 talking book. Listening carefully, I noticed that some of the compression artifacts were in stereo. It seems that these MP3 files were ripped directly from a CD source, as the sampling rate is 44.1 kHz, and they are encoded in stereo. The original sources were analog recordings of several reading sessions, as the tape print-through and a few editing errors revealed. If I had been given the task of encoding those MP3s, I would have ripped the CDs to WAV format and then converted the WAVs to mono. I might have even experimented with downsampling the WAVs to 22050 kHz to see if any perceptible loss of quality occurred, given that these were speech recordings without a lot of high frequency content.

>>Why? How is that hilarious? I didn't mean it as such.<<

It's hilarious because you're saying a a ripoff of Fraunhofer's technology is the "best MP3 encoder". That comment along with your "theories" about how to best use FHG codecs tells the board you haven't a clue about how perceptual audio coding works.

As for "ever hearing of LAME", I started working with audio compression algorithms probably around the time you starting soiling your diaper. So don't try to school me, son.

>>I don't understand why up-sampling 44.1 kHz source material to 48 kHz prior to compression would produce better results.<<

That's easy.
It won't.

It's like dubbing a 7-1/2 ips tape to a 15ips tape because you think it will sound better at the higher speed. In reality it's no better than when the original recording was made - or in the digital realm - when the original recording was sampled.

........Author: Notalent
Tuesday, April 07, 2009 - 7:52 am


You would be surprised how many people in radio don't understand the concept of "Broadcast Quality."........

No I wouldn't.

LAME is a very, very good encoder. I've been through a number of CD mastering projects and when we exchange sound checks with mp3, LAME is the encoder that does not cause problems. (final sound checks are raw of course, but for the first coupla passes at processing, mp3 is just great!)

The Fraunhofer codec suffers from lack of spectral accuracy at the higher end > 10Khz, and temporal accuracy all over the place, particularly at the lower bitrates < 160Kbps. This is easily heard, and would not be something anybody but a very casual listener wouldn't be able to distinguish.

Lack of spectral accuracy manifests as the sound being "different" but otherwise clear. Temporal accuracy manifests as the slurring or "watery" sound often heard on higher frequency instruments (cymbals, for example) and on vocals, where subtle harmonic changes discolor the sound.

I don't agree with the up-sampling as it will introduce artifacts. There is a case for up-sampling to an even multiple of the original sampling, if effects are to be applied to a source file, but that's it, and only if the effects require the precision inherent in the higher sampling range.

The tuning of the LAME codec is more extensive than that done with the Fraunhofer one. Because of this, it does not always perform the best at the lower bitrates < 128Kbps. Below this bitrate, you are talking crap anyway, so who cares really?

It shines best at 160, 256 and is debatable at 320. Once you hit the 320, there is a clear point of diminishing returns and it's also my opinion, if that level of quality is desired, go for the raw audio and use a loss-less format to handle it. FLAC, or something.

Encoding above 320Kbps is just foolish, unless it is being done for multi-channel sound. Better again to just use lossless audio.

(we do 256Kbps most of the time --it's the quality / compression sweet spot for LAME)

IMHO, well encoded mp3 files can sound broadcast quality, or better given they are not trans-coded as part of the audio chain. Since there are essentially NO raw audio chains, extra negative artifacts are bound to be present in the final "to air" stream.

I do wonder about that. HD Radio uses spectral replication at the receiver end for audio above somewhere > 5khz. If the mp3 file is well encoded, from raw audio, perhaps the second gen artifacts wouldn't be significant. Don't know. Anyone here care to comment? The spectral replication bit is different enough to warrant direct experience, which most of us just don't have

(and that's my bitch about Ibiquity as it's basically licensing radio... and I've been awful good about that as of late, so it's time to plug that just because... If that codec were open, it would see the same kind of ongoing improvement and peer review LAME does and would very likely perform better than it does today. At the least, people could have some direct experience with it, and that is helpful on a lot of levels.)

As for converting audio. If one is compressing, that's just a batch file, or some goofy utility, given said utility allows one to specify the codec engine to be used directly. Easy cheezy. Write it, run it and step away and wait.

I can't imagine replacing audio files would be so onerous as to not be practical. Besides, it can be an incremental thing. Generate a file list, generate a type list, sort, and that's your to-do list.

One a shift, and a year later, you've nailed all the bad files. Big deal. This is the kind of thing that absolutely can be done. Radio isn't the only industry with tight time constraints. The good stuff happens incrementally, and it's how a business remains strong in these crappy times. Welcome to the club. Nobody has it any better in this respect.

Another thing about mp3 is the playback engine. Some are noticeably better at accuracy. Even with poor quality files, they exhibit less slurring on the higher frequencies than other ones do. That's an ongoing PITA as basically a given playback system really needs to be vetted with a few files before committing to it. This is important when uncompressing files to raw audio format, like when burning to CD. The built in encoders in a lot of software really kind of suck at this. Don't use them. Better to just rip the audio, then run your audio CD program directly against the uncompressed audio files.

The LAME project was a rewrite of the original mp3, and is supposed to be delivered in source code form, which the user then compiles. In reality, it's available as a binary easily enough, and that's a bending of the rules. (oh well!)

It's not a rip-off. Many people who characterize it as such, simply don't understand the software motivations related to it's creation. The idea was to insure that mp3 was OPEN, and to make it ROBUST as it could be, not just stop advancement once licensing and a standard was agreed upon.

The project wouldn't exist, if those of us who want to compute on open hardware and software were able to do so without having to basically reverse engineer significant codecs over and over... (that really sucks, and if I could change anything, I would mandate RAND licensing for all this stuff, so that open computing isn't such a battle.)

And that's the state of things today. This isn't about free as in free beer. It's about free as in, "my computer does what I tell it to and is not my personal media policeman". Remember where you read that when your shiny new Vista or Mac machine tells you you can't manipulate some media file --even potentially your own damn file! KSKD will be happy at home, not worrying about this one bit. Never have, never, ever will.

So, we have LAME and many other projects that exist because of that. And since they exist, the work required to realize them means it's not too much more work to make them perform to their absolute top potential. Typically, the same peer review process used to build and vet code is applied to the performance of the project and the result is very, very good. If it were not, somebody else would fork the open project and proceed to then make it very, very good. This is what lots of eyes and open code delivers consistently. It's simple competition and innovation. --just the kind of thing that does not happen when formats are patent encumbered as they so often are.

The Fraunhofer codec engine you license, was not built like that. Sorry. Development and tuning of the perceptual encoding engine has basically stopped, with Fraunhofer perusing other licenseable technologies. Their model is to secure the licensing. Once that is done, they have very little incentive to actually build on that license, so they don't.

LAME will see updates as needed until it's no longer useful. BIG difference.

Say what you want about the ethics of the project, know that I've turned quite a few people on to that codec and they DON'T go back.

Heck, want to feel good about it? Do what a lot of people do. Pay for their license, then use LAME! Chances are you've paid as part of your PC hardware anyway, so it's not like Fraunhofer is getting ripped off. They got paid. So it all works. Open computing is possible, and quality audio encoding is possible, and Fraunhofer collects the money. Everybody is happy.

This is also why they really don't care all that much! If it was a problem, they would care, but it just isn't. Funny how that all works.

You know what it boils down to? 99.9% of listeners wouldn't notice the difference between a 128K MP3 and something of a higher quality.

I think you are wrong about that.

They don't care, but a whole lot of them notice.

Nah... it would have to be really lo-fi for Joe Schmo to really notice. Not everyone is an audiophile radio-geek.

I see practical problems associated with having a radio station music library in compressed form. If your compressed library is your only source for the material, what do you do if you find a problem with a track? Say something as simple as a tape dropout you want to fix, or too much leader at the head, or a low level intro. You'd load the problem file into your editor, fix the problem and then re-save the file to your lossy format. So now it's been through the laundry twice. If the library is uncompressed to start with, you don't have to worry about any of that.

All major radio stations should be required to replace all compressed music source material with linear files. The NAB should get this problem taken care of!

128k mp3, particularly done with a below par codec IS low-fi.

Really noticing is caring. Same thing.

Besides, that's not broadcast quality is it? Unless we want to redefine "broadcast quality" to be the lowest common denominator.

Seems kind of foolish to me. It's easy to lower expectations. Very, very difficult to raise them once lowered.

If that's going to happen, it's generally a bad idea to do so, without getting something in return. Seems to me that's a cop-out. The average person might not notice or, if they do, care, but they are not setting the bar are they? When good quality output doesn't cost anything (and having quality source audio files really doesn't), then why lower expectations?

Save a few pennies? That doesn't have a real return. What else?

Good point on the editing Scott. Each pass through the cycle just makes it worse... There are some direct mp3 editors out there. I've not tried any of them though. Sure seems like those would have limitations with block sizes and such. Or they have content limitations as only so much information can exist at any one point in time. How are effects done then?

[START OF RANT]

From 1lossir's insistence that FHG compression is somehow superior to LAME, 1lossir gives me the impression that s/he's a stubborn old pook whose idea of "broadcast quality" is playing recordings of worn-down aluminum 78's copied from about six or seven successive generations of eight-track carts before finally being duped to cassette. The records, of course, played from one of the cheap-ass old BSR "turntables" like you used to find in those old stereo/phono/tape units in the '70s, with a hubcap platter, 10-ton stainless steel tonearm and a Tetrad cart, no less.

BY MY EXPERIENCE (undoubtedly by the experiences of others as well), LAME *IS* SUPERIOR TO FHG MP3 for precisely the reasons Missing describes above, especially with regard to spatial accuracy and its openware nature. I really don't care how long you claim to have been working with digital audio compression 1lossir, that non-descript statistic you mentioned doesn't mean much to me anyways. If you can't tell how much clearer the sound quality of a "virgin" M1L3 file constructed from a first-gen PCM source is compared to same thing constructed with FHG codecs, you've probably either got ears of tin or you probably just flat out don't know. Yeah, I said it. Consider yourself told!

My "theory" as you warmly dismiss it, about how to best utilise the FHG system, I couldn't possibly dumb down any more than as follows: just don't even bother with it at all if you are coding music.

(I can say the same about the bletcherous excuse for an M1L3 codec Apple put into its I-Tunes software. In fact, I imagine I could possibly cause a DoS event on this forum just listing all the things Apple got WRONG with the entire bagbiting suites known as "I-Tunes", as well as "I-Pod" in general, but I am not going to right now. My lunch break's almost over and I gotta' get back to work.)

Shit, even with a dumbed-down point-and-click interface like WinLAME, even some of the stupidest n00bz, luz3rz and script kiddiez that the Gods ever put on the face of this Earth could produce M1L3 files with quality better than FHG!!

One other thing, my name's not Son. It's Seth. Son is the name of the dude who does my Grandparents' landscaping.

[END OF RANT]

LAME does sound better. MP3's are for the consumer. Old-school thinkers still live in MP2/3 world for broadcasting. Overcompressed processing just makes these MP2/3's turn into a mess on the air.

Big time. Processors employ wavelet technology to get what they get done.

If they are directly manipulating a compressed audio file, that's one thing. If they are manipulating the uncompressed product of a compressed audio file, artifacts can and do occur.

Muddy, tiring sound are the two that come to mind most often when I hear it. The Scott "through the laundry twice" deal.

mp2 can be really good, BTW. The problem with it, from a modern day perspective is simply that it's not all that much compression. (which is why it can be really good!) When mp2 saw greater use, that minor amount of compression made good sense. Bet it still does in some scenarios.

However, as our throughput increases, mp2 is a net value loss over raw loss-less audio compression. Frankly, mp3 is a give away, freebie, teaser for the real thing.

That could change over a fairly short period of time. Been reading some info about teens preferring that sound. It's the same effect we have imprinted on us --or those of us, who got exposed to a lot of analog signals.

Radio should just deal with this. Do it incrementally too. Somebody some where publish the goods, recommended settings, schedule and ease of encoding software and processes and just let it happen.

Could be completely gone within a year, keeping expectations where they are and not losing ground.

Save that crap for the streams, where the freebie effect has value. If they want it clean, they can buy the production, pay for a subscription and commercial free download (podcast), or listen on air.

Playing the freebies on air just devalues the medium. There is enough of that going on as it is.

For my productions MP3 is what I use for distribution. Pretty much everything DVD-related (it seems) supports MP3. My own personal listening is most often done in PCM. If I have to use a compressed format for that purpose (the CD player in the truck!) it's usually layer 2 files. The only thing that's *really* MP3 about it is the extension. ;o)

TwoLAME r00lz.

(No joke--none of my CD players will recognise the files if they have an XXXXYYYY.MP2 extension. Sometimes the machine will just spit the disc right out. So I swap it to an XXXXYYYY.MP3 extension. Ba da bing: it plays! That's another interesting point to bring up: MANY (if not most) MP3 PLAYERS CAN PLAY MPEG 1 LAYER 2 AUDIO!)

Lame is definitely not lame! They've made great advances in that mp3 codec over the years to where it sounds as good as wav to my ears (at 320k encoding). The well respected open source project is at http://lame.sourceforge.net/.

Make sure you're running the latest lame codec file (I think lame_enc.dll in windows once it's compiled). Some software installs will overwrite it with an inferior older version (Microsoft's media player, perhaps?)

You can roll up your sleeves and read all about it at that site and compile your own (or get one already made). As mentioned by all above, there are worthy alternatives to getting great sounding files without having to encode in space hogging wav's. AAC and FLAC are also coming on strong. It just takes knowledge, care and time to make sure the result has a high Q (quality) factor.

You guys can debate the merits of one compression scheme over another all day long. They were all invented to make digital files smaller at a time when systems were slower, storage was expensive, and no one wanted to wait a week for a wave file to download.

All compression methodologies take something out of the original recording. It may be musical overtones in some frequency ranges that aren't easily apparent to the human hearing system, but something is always removed.

The bottom line is that you cannot make a copy of anything and have the copy quality be better than the original. The only way to maintain original quality is to make all digital copies using the same or better sampling than the original using a high quality piece of equipment. A higher sampling rate will not improve what you started out with.

The better the original, the more...or less...you can process to achieve the on-air sound you want. Linear copying of material for broadcast is best.

Unfortunately we have a lot of commercial content that comes to us over the internet as mp3 and when music is involved, that doesn't help.

The NAB is not in the business of policing quality. The SBE should be. After all, they're the Engineers.

All but the loss-less, simple digital compression that removes redundancy in the bit stream. Agreed otherwise.

BTW: I'm not an audiophile. Most of the time, I listen to stuff on rather ordinary gear. My best playback devices are a good set of headphones I got by opening all the packages, playing a few well produced tracks to see if I could still hear why they are well produced, then picking from there. Spent maybe $50 tops. My car rig is just simple, plain and not well balanced at all. And I've a nice set of 80's era tower speakers at home to listen on once in a while.

Lots of people are like this. We notice well produced stuff, but we don't obsess over it. I've got great ears, which is why I'm part of some projects. I'm also quite happy to enjoy the art and meaning of the production on it's merits, ignoring completely the technical things. The two are different.

Well produced media has value in that when I'm in the mood to experience it at it's peak, it's there for me. And that's kind of enduring over time too. Most times, it's not worth fighting the battle for perfection, just to enjoy some art.

This is why the bar is important. Ordinary people (and in this respect, I'm quite ordinary) have lives and things and need simple entertainment. Where they notice or when they have time to notice, the value perception for a job well done matters.

It's completely true that airing mediocre audio isn't going to impact the numbers significantly, particularly now when the expectations of mediocre is on the table. It's true for the reasons I just gave. However, when it comes down to one of those benefit of the doubt moments, the media production that shows some and effort and passion was invested gets the nod. That could happen the one time somebody is in a car, hears that song that just resonates with who they are and what they are feeling and they turn it up! If the quality is there, you have a radio bonding moment. If not, it's frustration and that devalues everybody trying to get the moment.

That is why those that care about it do what they do. And again, that's true no matter what your discipline is. Failure to grok this is a problem that everybody deals with, not just the people who don't understand. Look around at what people think of radio, on average, and that's the impact of "ordinary people don't notice". The bar has been lowered, despite plenty of people still executing at the higher level. That sucks.

From a sales perspective that has implications also, and this is why it matters at a business level.

When it (the production of whatever it is) matters, who are people going to call? Will it be the crap master, or the one that has demonstrated solid mastery of the craft? Who will they call when their job and reputation is on the line? (it's not the crap master)

Who will endure?

Those people that have that mastery and can apply it as a second thought, competing well over time. Good production values (and that applies to simple production, not just audio production), service and support are the holy trinity of business. When you do all three, the question of whether or not the work will meet expectations is off the table, leaving simple time and cost as factors.

This is really easy for the sales person, and that ease translates into dollars and stability for everyone else. Good sales can make up for a lot of other things. This is part of why we have sales people. It's better to let them do the selling without also having to do the diplomacy as well.

For the marketing people, it's a slam dunk too. Branding something like that is just cake backed by real street cred. Branding mediocre is diplomacy too. Dilutes the core message and just reduces your return on investment.

That kind of street cred gets lots of leads as people have solid expectations. Expectations matter over longer periods of time. Word gets around and that is a built in lead generator that the crap master won't have. Where sales dollars are concerned, that's worth basically being able to ask for the solid rate and get it more often, instead of cutting corners, racing to the bottom.

Why leave money on the table?

The crap master is just there for the quick buck and everybody knows it. Big difference; namely, one that determines who endures and who does not.

The crap master does not endure because they do leave that money on the table, and over time their business is weaker because of that. Typically what happens is they leave that money on the table, then take it out of their employees ass, further exacerbating the crap cycle! Over time, that adds up to a whole lot and they cannot compete, unless everybody else buys in and also races to the bottom.

(welcome to most of corporate America, where they teach people how to do this! Ugh...)

All of that comes down to two words:

Manage Expectations.

Everybody in the chain has a burden to do this. Everybody from the President down to the guy just doing stuff every day.

Focusing on that brings all the other stuff into play as an artifact of managing expectations. Think about it: To set expectations, you do what? Communicate. To meet them you do what? Execute properly. To execute you need what? The right people and the right tools. Who are the right people? THOSE THAT HAVE MASTERY OF THEIR CRAFT. The right tools are built by those, WHO HAVE MASTERY OF THEIR CRAFT.

And it goes on and on. Ask your sales person what he wants to sell. That sales person will say they want to sell good value. Why? Because that means good money. Simple.

I have the happy pleasure of working with the person who first said those two words many, many years ago. It took me a good five years to really grok what they meant. I've spent the rest trying every day to leverage that understanding.

Here we are nearly 20 years later, and I'm still watching and learning how those two words play out. BTW: I asked once, "How do you know how to sort all this stuff out and do what you do well?" Those two words were the answer I got.

Many of my peers are gone because of tough times. I, and others who understand this are not.

That's not a boast. Times are tough. Stuff happens. There are some people who do understand this stuff gone, or having to move. But, all the crap masters are gone now. That's the point.

This is a personal testimonial as to the power inherent in those two words. And it's relevant to this thread.

Don't lower expectations for simple cost reasons. Broadcast quality is the bar set by those who know how it's done right. Meet those and exceed when you can, and do it efficiently, and you will endure.

It's just that simple, no matter what discipline you practice.

Which is why it's silly to just air crap audio, "because nobody will notice". And now I've come full circle. Sorry for the long way around, but sometimes stuff just needs to be said, and I would have said this on any forum, regarding any discipline, not just radio.

In this, radio is no different from any other business.

Why say it? Because I really like good radio, and I really like dealing with people who manage expectations.

The OMNIA is a superb unit used by many stations.
In fact, it is so good it manufactures squelch like picket-fencing noise trying to find the elements missing when playing back those MP3s.
It can actually sound like you're listening to the FM in a fringe reception area.

Replace the audio with a WAV file, and all that noise/perceived "reception problem" goes away.

I wish that all stations were managed by real broadcasters like Hwidsten and Robin_mitchell. They know that (in 2009) there is no acceptable reason to ever compress audio files. Broadcast quality means linear, period.

When I was at Entercom we went to great lengths to make sure that no mp3 music made it on the air on any of the stations.

On 97.1 Charle FM, Huey Lewis & The News "I Want A New Drug" and Pat Benatar "Invincible" sound like complete audio garbage! How did these nasty MP3's get in the system, and why haven't they been replaced?

There have been incredible improvements in audio processing lead by Bob Orban, Frank Foti (Omnia) and Greg Oganoski (sp) who is from the Northwest, and as Robin says, the gear is so good that it goes looking for what it needs to make your station sound good.

Orban is accused of building processors with a "sweet spot" that is a group of settings that makes his unit sound best....or at least they way Bob thinks it should sound. I don't believe that is the case with the Omnia, as you can make it sound as bad or as good as your expertise allows.

But, in the final accounting, the quality of the original material you're feeding down your program line, and the quality of the equipment you're using BEFORE the processing determines how good or bad you're going to sound.

Most of you guys are too young to remember the audio research that was done to determine the differences between what men and women in different age goups liked and didn't like about music that was equalized in different ways. That hasn't changed, and ought to be considered when processing is set up based on a station's demographic target.

I found this website while looking around for old audio processors. Audio here is quite impressive, brings back memories of how good AM radio could sound if done right.

The loudest AM radio station on the internet. Really.
Original 1970s aggressive on-air sound using real electron tube amplifiers, 11kHz preemphasis, Gates Sta-Level audio compressor, CBS VOLUMAX 4300 audio peak limiter. It doesn't matter how many watts you have, it's the peak to average ratio of the waveform and the time constants used that make the sound!
http://www.standardbroadcast.com/590CKEY.html

One of the people responsible for KGON sounding so good is Bob Brooks. He will not "rip" any music into the system (or at least he didn't when I worked in the building). It was all done in realtime to insure the best playback. I'm not sure if Bob dubs any of the other station's music.

Like Hal and Robin said...garbage in, garbage out.

Very cool.

Here's a direct link to the audio feed (in case you want to load it into your XMMS or Winamp directly):

http://www.standardbroadcast.com:8000

Wouldnt'cha know? It's an M1L3 feed!

Tip: Listen via a DSL line on a computer connected to a stereo receiver in order to fully appreciate it. If you just listen through headphones connected straight into the sound card your ears'll get fatigued *really* fast.

EDIT ADD~

Lookit this:
http://www.standardbroadcast.com/Pictures/Things/Radios/Broadcasting/44bx01.jpg
My Grampa acquired one of these mikes in the 1970s when he was in Viet Nam, and when I was a little kid he said that would become mine when I was older. Somehow over the years the mike got mislaid and he never did give it to me before he died......hopefully Granma ("Hurricane Judy") didn't throw it away at some point over the last 20-some years since.........

I don't know how words can describe how cool this is:
http://www.standardbroadcast.com/Pictures/Things/Radios/Broadcasting/590CKEYstation.jpg

Listening to this feed, I am surprised that I do not hear any distortion, despite the compressors being driven very heavily. I am even more surprised, looking at the schematics for these processors, that they used diodes as the gain control element.