• Get involved.
    We want your input!
    Apply for Membership and join the conversations about everything related to broadcasting.

    After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
    If you use a disposable or false email address, your registration will be rejected.

    After your membership is approved, please take a minute to tell us a little bit about yourself.
    https://www.radiodiscussions.com/forums/introduce-yourself.1088/

    Thanks in advance and have fun!
    RadioDiscussions Administrators

the sound performance?

i need your opinion about which you thing is the best way to have better perfomance in music befor the processor
i have a pc automation which means that i am playing mp3s must the mp3s cut it whith mp3gain before send the sound to the processor
and if yes what gain(db)?do i connect the processor directly from the console or i must put anything else in frond?
 
DITTO Studio 1. Uncompressed Wav is the way to go.Hard drives are cheap today.don't convert the mp3 to wav files.
Won't change a thing.Just another conversion.
 
to much songs where i will find it again to do it wav?
this is very difficult thats why i am talking about mp3s
 
you can find the original music on jonestm.com...you can order them as wav files or on CD, whatever...you can ALSO rip them off the original CD's. Quality doesn't come cheap...and doing it right the first or second time is the way to do it.
 
you can come out of the console to the processor or the sound card.If you do mp3 the dsp-x manual states mp3 256,if for some reason you can't do linear.what is your format?
 
A "Red Book" CD--the kind you have at home to play music off of, holds around 700~800 mb of digital information; which is around 72 minutes of stereo music. In creating this CD, the original music from the master recording was sampled at a rate of 44,000 Hz (cycles per second for us old folks); in other words, to audio frequencies up to 22,000 Hz. Typically, human hearing drops off fast past 15,000 Hz., a piano tops out at around 5,500 Hz., some chimes at around 8,000 hz. However there are overtones and harmonics that reach this high, and create the "natural" sound of the instruments.

The digital encoding method used to create a CD uses a small amount of data compression. That is, a certain amount of repetitive information is "thrown away," in order to extend the amount of information that can be put on the CD. But this compression is very minimal

A .WAV file is a type of digital file used in most audio editors. The "wake-up" sound that windows plays when it boots up is a .wav file. WAV files can be encoded at different sample rates and different bit rates, depending upon their purpose. They are considered linear because there is no data compression used when the audio is converted to a digital file; one determines how big or small the file will be for a given length of audio by changing the sample rate, and bit rate--with a trade-off in playback quality--lower frequency response as the sample rate goes lower. Most WAV files are 16 bit, though 8 bit can be used for low quality audio. Fewer bits==less information converted from the original audio.

If you were to record a CD cut real time as a 44 KHZ. 16 bit file, you would have audio quality almost identical to the original CD. This file would take about 10 mb per minute of stereo audio. MPEG-3 is a standardized data compression scheme. The theoretical frequency response of the music piece is somewhat preserved, but "duplicate" information is thrown away in order to shrink the size of the resulting file. A lot of information. That same 1 minute stereo recording in MPEG-3 would be less than a megabyte per minute in hard disk space. But there is no such thing as a free lunch. Hence, if you recorded that same CD cut real time as an MPEG-3 file, and compared it with the 44 khz. WAV file, there would be an apparent difference in quality. Not necessarily in frequency response, but in the sense that certain sounds/instruments would not sound quite right.

Not a problem if you are listening on ear buds off an IPOD, or through tiny little plastic computer speakers. However, on an FM station, programing from MPEG-3 will never sound as good as the programming of the same music on another station using WAV files.

Many automation systems will handle both MPEG-3 and WAV files, although they may not be able to overlap them on playback. As mentioned, hard drives are cheap, a 500 gb drive could easily handle the typical 500~700 song library used in many radio formats. So a prudent thing to do, especially if you have the original CD's (or can borrow them again) is to re-do your library as 44 KHZ 16 bit .WAV files.
 
By linear, I mean clean wav files, 44.1khz, 16bit...ripped straight from the CD itself...or gathered from some other source. Mp3's and digital bit reduction clouds audio, and gives you less to work with when it hits the processor.
 
I went with mp3 because the price was right. I record to CD, rip that at 192 k, which sounds fine on AM.
If I have to use digital editing, I get a 128k mp3 file , which is still sounds good on the AM, but listening to the speakers right on
the server sound card is distressing.
Always keeping levels high seems to be the key in minimizing artifacts.
It was always the same in tape recording. Never just go by the meters, FIND OUT how much signal you can run into the tape, and
you'll never have an issue with hiss (almost). But the recording always sounded way more "alive" than one where the tape was not "pushed".

When I started running an automated music server, I was given a few mp3s of some 60's hits, and I was appalled at how dead-souning these were compared to any I had dubbed from 45 or wherever. Boosting the levels helps them some but they still seem flat.
Even with eq, etc, I wonder who recorded some of these and from what original.
I have a few Beatles songs in rotation from the early years where I've used original Vee-Jay and Capitol 45s.
The clarity of the early mixes and production is still amazing as is the dynamic peak clarity.
This still reads as clearly when I listen to these songs on the air, (in the car out front as a test).

Listen to the CDs themselves on air and consider if they're OK or if you need other sources.
I am just now listening to a song I dubbed from a commercial compilation CD running on my part 15, and it's not sounding....correct
compared to what I know my viynl dubs sound like. I even hear something like mistracking tape high-frequency changes.

I have an easier time accepting 128k streams and files if I do not get to A/B compare with the clean source.
Or if I don't already know what the music "should" sound like.
I do NOT hear 128k artifacts when I listen critically to my AM1620.

I'd kind of like to, in order to have something to worry and fuss about.

I enjoy WFMU's 128k stream and only rarely am annoyed by the sample rate.

96k is fatiguing to listen to.

I find myself more satisfied with 128k and 192k than I would have thought. But then I'm feeding an AM.

The least you can do is edit every file and boost volume levels to fully use the range available, but don't clip.
If they range too far from "loud files" to "quiet files", one or the other's gonna suffer.
Processing agressive enough to deal with some songs hot, others weak will probably also crush the dynamics of the best sources.
Unless of course you can afford upper-crust processing.
 
menotti1 said:
you can come out of the console to the processor or the sound card.If you do mp3 the dsp-x manual states mp3 256,if for some reason you can't do linear.what is your format?

The DSPX manual, to be precise, states:

We strongly advise against the use of MP3's and other compressed audio formats for audio storage. If you must
use compressed audio we advise rates of 256 Kbps and higher. Linear formats are always to be preferred.
Compressed audio formats employ frequency masking data reduction techniques to reduce the bit-rate. Through
re-equalisation the DSPX-FM can violate the frequency masking characteristics of the bit reduction process, creating
distortion that was inaudible prior to the DSPX-FM processing.


Like it says in there, we do not recommend using compressed sources. Not only is the compressed audio of lower audio quality itself (depending on the bitrate used) but even if it sounds good being played back on a regular audio system, it gets worse when being played through a broadcast processor. The masking that the codec relies on to "hide" artifacts during coding is disrupted by the multiband dynamic equalization plus you add up to 17dB of boost to the high-frequencies - like said earlier, the area where codecs exhibit most of their artifacts. If you play a typical 128 kbps MP3 on-air back-to-back with the same track of a CD, most of the time there will be a noticeable difference in sound quality, depth and cleanliness.

[Personal opinion:]
Coding audio to MP3 perhaps made sense when storage space was small and expensive. This has been no longer case for quite a while and it's time broadcasters get back to proper quality uncompressed audio sources such as WAV (or lossless formats like FLAC). The other media such as TV is doing a whole infrastructure change to move to higher quality, higher definition picture (and sound) quality and market seems to want to buy higher quality. We as radio broadcasters keep shooting ourselves in the foot, using subpar audio sources when we could be delivering higher audio quality to listeners and this is completely within our domain and depends only on our effort (time and money).

Just my opinion...


Regards,
Goran Tomas
 
While most of your conclusions are correct, you frankly don't know what your are talking about:

1) There is no audio or bit rate compression on a Red Book CD - this is linear PCM

2) The sample rate of a Red Book CD is 44100 kHz/CPS

3) If you were to properly rip a CD to a wave file the results would be identical, not very close.

With all due respect (and as I say, your conclusions are correct)

TomT said:
A "Red Book" CD--the kind you have at home to play music off of, holds around 700~800 mb of digital information; which is around 72 minutes of stereo music. In creating this CD, the original music from the master recording was sampled at a rate of 44,000 Hz (cycles per second for us old folks); in other words, to audio frequencies up to 22,000 Hz. Typically, human hearing drops off fast past 15,000 Hz., a piano tops out at around 5,500 Hz., some chimes at around 8,000 hz. However there are overtones and harmonics that reach this high, and create the "natural" sound of the instruments.

The digital encoding method used to create a CD uses a small amount of data compression. That is, a certain amount of repetitive information is "thrown away," in order to extend the amount of information that can be put on the CD. But this compression is very minimal

A .WAV file is a type of digital file used in most audio editors. The "wake-up" sound that windows plays when it boots up is a .wav file. WAV files can be encoded at different sample rates and different bit rates, depending upon their purpose. They are considered linear because there is no data compression used when the audio is converted to a digital file; one determines how big or small the file will be for a given length of audio by changing the sample rate, and bit rate--with a trade-off in playback quality--lower frequency response as the sample rate goes lower. Most WAV files are 16 bit, though 8 bit can be used for low quality audio. Fewer bits==less information converted from the original audio.

If you were to record a CD cut real time as a 44 KHZ. 16 bit file, you would have audio quality almost identical to the original CD. This file would take about 10 mb per minute of stereo audio. MPEG-3 is a standardized data compression scheme. The theoretical frequency response of the music piece is somewhat preserved, but "duplicate" information is thrown away in order to shrink the size of the resulting file. A lot of information. That same 1 minute stereo recording in MPEG-3 would be less than a megabyte per minute in hard disk space. But there is no such thing as a free lunch. Hence, if you recorded that same CD cut real time as an MPEG-3 file, and compared it with the 44 khz. WAV file, there would be an apparent difference in quality. Not necessarily in frequency response, but in the sense that certain sounds/instruments would not sound quite right.

Not a problem if you are listening on ear buds off an IPOD, or through tiny little plastic computer speakers. However, on an FM station, programing from MPEG-3 will never sound as good as the programming of the same music on another station using WAV files.

Many automation systems will handle both MPEG-3 and WAV files, although they may not be able to overlap them on playback. As mentioned, hard drives are cheap, a 500 gb drive could easily handle the typical 500~700 song library used in many radio formats. So a prudent thing to do, especially if you have the original CD's (or can borrow them again) is to re-do your library as 44 KHZ 16 bit .WAV files.
 
OK, but understand you have chosen to give your listeners an inferior product. And no amount of processor fiddling will undo the fact that at 128 kbps only 10% of the information remains in the MP3 version.

Just as well you didn't spend a lot of money on your processor.

Today, start with your highest rotation cuts and find linear versions

bil34 said:
no i work mp3s and it is very difficult now to collect the cds again to wright again the wav files... :(
 
There are some steps that can help SOME of the recordings, maybe.
Finding an amp with certain flaws can "re-smear" out some of the artifacting.
Leaky capacitors in audio coupling circuits, while not welcome in hi-fi reproduction, can, like soft focus in portraiture, hide smaller flaws.

Similarly, small or just plain overloaded audio coupling transformers in low-signal circuits can use magnetic hysteresis to partly
strip the zizz off mp3s. This loss of resolution, again not welcome as detrimental to hi-fi recordings, can give a better result
than your original file, when the original has this kind of artifacting. Expect tradeoffs but this is worth trying.

Both inductive and capacitive coupling attempt to return the wave to sinusoidal, preferrable in the case of mp3s where
the bitrate is lower than one might like and the original is no longer available.


Maybe there's a market for such a processor, with auto-variable blur-n-smear.
 
Here's something pretty funny as a side-note to all this discussion. Last year (not this year) at the NAB in LV the BW guys were using audio out of a computer that was mp3 for their demo. Appearently several of the people that listened to the demo of their processor crinkled their noses at the coding sound. I think more of us noticed than they thought. Case and point: Don't use mpeg anything on your radio station if at all possible, especially if you have other compressions downline like HD or links. I can hear the difference in the high-end between mp2s or mp3s and something straight off cd or linear. There IS a difference.
 
I am pleased to see Goran (and the proc amp manufacturers) using their terms properly. Data reduction is what happens in most CODECs... you can't get it back. Whereas compression (reed - solomon in CDs if memory serves also various .zip and .rar schemes) allows recovery of the original bitstream. Let's try to use the two properly, so we don't confuse people, whattaya say?
 
littlejohn said:
I am pleased to see Goran (and the proc amp manufacturers) using their terms properly. Data reduction is what happens in most CODECs... you can't get it back. Whereas compression (reed - solomon in CDs if memory serves also various .zip and .rar schemes) allows recovery of the original bitstream. Let's try to use the two properly, so we don't confuse people, whattaya say?

I've been fighting this one for years... But people are so used to refer to coding as compression, that when you talk about data reduction they don't understand what are you talking about. That term does not compute in their heads. When you mention compression, they are immediately back with you, so I mostly try to use "lossy compression" and "lossless compression" to meet them halfway and still be technically correct. "Psychoacoustic coding" also seem to scare people away.

The problem with word compression alone (apart from being incorrect or at least incomplete) is that it is very easily confused with audio dynamics compression. So when you're talking about dynamics processing for coded audio and interaction between dynamics processing and coding, some people are having a hard time understanding that you are talking about two completely different things...


Regards,
Goran Tomas
 
You have to be, to a degree, politic; in that you're selling boxes. I ain't selling anything, I can be as big of a a$$ho@# as needed :):)

Really, as long as the user understands what's happening, the semantics take care of themselves. After all, its there problem, no?
 
littlejohn said:
You have to be, to a degree, politic; in that you're selling boxes. I ain't selling anything, I can be as big of a a$$ho@# as needed :):)

Really, as long as the user understands what's happening, the semantics take care of themselves. After all, its there problem, no?

In a way... But you'd want people to genuinely understand things and understand the difference. I've been trying to explain this to colleagues while I was still working for radio stations (just until very recently) and some of my friends who are into IT and telecommunications and who all have engineering degrees. But like I said, most of the time it's compression=MP3...


Regards,
Goran Tomas
 
Status
This thread has been closed due to inactivity. You can create a new thread to discuss this topic.


Back
Top Bottom