Broadcast Quality Normalization

Wow that was cool :O

Well, maybe you can put a sample CD together,
maybe remix any material that might not be
’Broadcast’ perfect, (category 5), and drop
off some sample CD’s to college radio stations.
Also ‘there are some digital stations’ on 'satellite’
radio that you can get some air-play on.
And if all else fails - try Walmart !! :)

I’d prefer WALMART, more outreach, wider market, more sales, more fame, more subsities, . :)

I think they are correct - most Cd burning programs will run a check first and if the level is above 0 you are asked if you want to fix that to conform to “Red Book” - there is a series of books that set standards for recodings; radio may have additional standards but I don’t really know, AM stations used to have a lower “sound limit” for broadcast, I don’t think it is a problem for FM and may be handled in re-recording for broadcast.
As was mentioned, most Cd are actually at slightly less than 0 in case the player can’t handdle or is mis- calabrated.
I think the best idea would be to run the songs through a mastering program (iZotope makes one called Ozone that is pretty good and not terribly expensive). Most stations are very restricted in what they put on the air, and if there is a problem getting it to play they are just going to throw it out, not try to correct it.
At this point, you don’t have to remix, and you could just run the songs through a Compresser, but something like Ozone will make them sound better.
Bax

In digital audio… there is no such thing as “above 0” in storage. Can’t happen. Therefore… your mastered wave forms may be AT 0db and miserably clipped, but they will never go passed 0… ever.

Loan him your “Kaboomifier” Bubba… :laugh:

D

That is proprietary technology there, D…

BubbaTosh, BubbaMac, BubbaSoft or BubbApple?
BubbleGumpIntosh?
iBubba
®

Quote: (Bubbagump @ Sep. 04 2008, 9:54 AM)

That is proprietary technology there, D...

Whoops! Sorry... I'll send a copy of my new plugin; "Slick Mix 66". (Just as soon as I finish coding, beta testing and inventing cold fusion.) :whistle:

D

I Bubba, you Poppa.

Diogenes, and anyone else in the know -For discussion’s sake, because I can only repeat what I “think” I know:
Well, you are correct about 0 db in digital - but unfortuntely, you can create a sound file with energy that is trying to cross above the technical limit of 0, they are called "clips"
unless,
You are using and analog recording system when it is called "headroom"
Some VU meters are set up to register how far the sound energy is above 0 when recording. I don’t see how such a meter could work in playback, but he said he had peaks above 0, so I figured he was getting that when he recorded or burned. Forinstance, I use Sound Forge CD Architech to burn and it will tell me I have a clip and let me burn anyhow if I chose.
If the number of effected samples is low, I think less than 3, it can be hard to hear the clip on many playback systems, but the clip is there and a more critical listening system may make it evident. My point, I would never put out something for sale that had clips (Sound energy that has tried to go above 0).

I’ve been reading every reply since I opened this topic. Thank you all - I really have learned so much. The peaks are noticed when I mixdown into a single wave file and pick the option to replace the chosen tracks with the mixdown wave. I made sure the settings are best returned to thier origional position but there [I thought this through] is a strong propensity that the settings did’nt return to the default state and therefore [perhaps the bass-tones, or eq] the signal is slightly boosted without my notice. I imported the same tracks I was sure caused the state of 0.02 signal after mixdown into a new project, and as all of you said the signal stays solid at 0. Sorry to stump you with my untechnical- perhaps imature statement. :cool: and/but I’ve learned so much. I’m also making prints for other OCALA musicians and keeping them on file like using compression and peaks in songs for radio submittions. I hope you all forgive me for the goose chase. :love:
:;):
:p
:O I can’t wait for v6.0.1 where the soft clipping feature will become adjustable. It’ll work well in some of Crystaleans treble mandolin stuff as a limiter. I like to boost the signal real high and play some of the sounds in the soft-clippin feature; sounds crazy I know but it is the perfect percusion amist a pounding bass tone… seriously - I like playing the clip. Anybody else?

You are talking about intersample clipping perhaps… this is not clipping in the traditional sense. Traditional clipping is where the electrical input to a ADC exceeds its maximum allowed input voltage and you have digital overs. Those are the nasty crackles we usually attribute to clipping (a better ADC will simply limit the values it puts out avoiding this being so bad… but you still have a crappy sounding wave form.) Past that… you have intersample clipping and plain old squared off wave clipping. As you mentioned… most clip indicators consider three samples in a row with a value of 0dbFS to be clipping. The latter is where things are hanging at 0db for too long chopping off a wave form. The former is where the slew rate is very fast between samples and in calculating the wave form, the DAC calculates a value higher than 0. In any event (and yes I am splitting hairs) the value on the media will never pass 0db. It is all in the converters that this happens.

As for putting something for sale with clips… well, that is a personal thing as half the new CDs out today are miserably clipped. technically… you can have awful clipping at any level. Taking any wave and forcing a set of consecutive samples to identical bit values is clipping be it 50 samples at 0db or 50 samples at -10db. (then you have square waves which are purposely clipped, but due to their even periodic nature sort of dodge the negative effects of what we call clipping.) Go grab a Nickleback record and do some analysis. Tis ugly. But again, this doesn’t mean that the DAC will actually clip with a digital over… it may just sound terrible from the squared off wave… or it may indeed clip as it is not built to limit itself to 0dbFS (ala intersample clipping) and will create a digital over.

What he is likely referring to is the digital over value on the playback peak meter. This is an analogy to passing 0db in analog. However, you still don’t pass 0db. Being that things are processed at 32 or 64 bit float internally, it is simply calculating “were I to truncate this to 24 or 16 bit… what would the bit over flow be?” and then it converts that to a db scale value. That then in turn is a clue to you to turn down so many db so as in a conversion to lower bit depths clipping is avoided. So he really has clipped files topping out at 0db… but in the processing of said files, he had roughly 2db too much gain which caused clipping. Again, the stored value is never more than 0db.

Thinking about it… for some real fun with intersample clipping… generate a sine wave at 22.05 khz a 0dbFS and in 16 bit. Play it back… though its value never passes 0 anywhere in the file… the meters will often freak out due to the slew rate. 22.05khz is the fastest slew rate of a wave form that a 44.1 khz file can store.

Bubbagump,
Thanks for the more detailed explaination. Now, I have a question. I know that you can get “internal” clipping in a mic when the sound pressure is just too high and it doesn’t necessarily show up as a digital clip (at least, that is what I thnk happens.)
Yesterday I had something happen that I have never had come up before. I did some recording of a female singer and the wave forms showed normal peaks above the center line and squared off form below the line. From your explaination, even though this was considerable below a 0 db clip, does the wave form indicate that clipping has occured? The sound is unusually “strident”. Have I got a probelm somewhere, maybe the mic?
Thanks,
Bax

OK… first off that is analog clipping. You are creating a signal going into the mic that is higher than the voltage it can cleanly handle therefore causing distortion or clipping. (Or you may simply be extending the diaphragm to a point where it can’t move any further) Distortion/clipping can happen anywhere along the line where the input of something is higher than what the output can handle. That is what gain staging is all about. That is how a guitar amp works. Create a ton of input gain to clip the signal… but in that case it is desirable. Though using the guitar amp analogy… you can crank the pregain creating distortion/clipping while keeping the output gain low… a distorted guitar sound at a low volume. At the end of the day… to say clipping is intrinsically bad is folly. There are times and places where clipping is a good thing. The key is to know what clipping is and isn’t, how it behaves, and where it is desirable or to be avoided. Some mastering engineers on high end converters purposely clip as they like the character it imparts.

Geez Bubba… you ARE a smart feller! :p Next thing you know, you’ll writing stuff about IMD, IUD’s and OU812’s! :laugh:

Don’t tell me… you’re like me, just can’t get enough “nerdy info” about something I’m interested in…

You KNOW my stance on the “loudness wars”… After tossing several otherwise excellent CD’s in the circular file, I went on a quest to find out WHY they got on my nerves so much. The MUSIC was fine. LISTENING TO THEM was NOT fine. All this digital distortion “art” is not for me. I can’t stand that noise… it just ain’t natural for my natural ears to enjoy.

D

PS Wow… just thinking about this makes the skin on my scalp crawl… just like those darn CD’s do… (did)

Heh, I just got a project to master where they use digital distortion on purpose. I sent it back saying hey, you have digital clipping and it makes your recording sound “broken” (like the speakers are bad) so fix it. They told me… that’s what we want it to sound like.

Quote:

I did some recording of a female singer and the wave forms showed normal peaks above the center line and squared off form below the line.


I can't say for sure if it's the case Bax, but often this can appear from a bad mic capsule, bad mic cable, bad pre-amp or even a poor A-D converter. My first guess would be the mic. If the diaphragm can't move freely enough, you'll get a "lopsided" signal. In extreme cases, yeah it sounds "funny"... maybe even "strident".

D

PS A couple of my cheapo small diaphragm condenser mic's exhibit this behavior. The higher the SPL's, the WORSE it gets.
Quote: (Bubbagump @ Sep. 04 2008, 1:26 PM)

Heh, I just got a project to master where they use digital distortion on purpose. I sent it back saying hey, you have digital clipping and it makes your recording sound "broken" (like the speakers are bad) so fix it. They told me... that's what we want it to sound like.

**Looking at floor, shaking head disgustedly... **

Wait a sec... **pulls on flame-proof leotard. (pink; don't tell nobody!)**

*Sigh* What is this world comin' too? Any dweeb with a laptop, a few CD's of samples and a VSTi or two can make "music" with no knowledge of MUSIC and recording whatsoever...

You should slap an L2 on it, push it WAAAAYYYY past the RED and send it back to 'em with a hefty bill. Don't answer your phone though, when they expect you to pay for their ruined speakers... I can see it now; "Whoa! Man! That RAW...*BZZT! PFFFT!* Ksss...." :laugh:

D