• 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

Frustrated At Audio levels

C

calavan888

I know that this has been discussed on the boards before. So here goes again, I finally get around to putting audio into Nexgen. Thought it sounded good on the trusty headphones. I used the internal normalization. Set it for the standard -10dbfs. Used PCM Wave first then tried MPEG 1 Layer II (3.7:1). For 2 days I tested it out at the station. First Day Wave ,second day MPEG 1 Layer II. Wave sounded Good ,as usual ,Compressed sounded really muddy, which as I understand at the ratio I used it should be really bright.As for levels ,I am surprised the board, Processor, and or the transmitter didn't blow up ,from overload,I mean those machines worked their circuits off !!! LOL! The levels were all over the place. Which is surprising because Nexgen is supposed to somewhat correct the levels during extraction. I have tried the RCS boards (It has become like a graveyard) nobody is home or for that matter even responding anymore.I am hoping someone here can help me. I am not, before I forget, using any expensive audio cards. Just the standard audio out.But that should not matter as the audio captured from the cd ,during extraction , does not go through the card. ::)
 
I'm the LAST guy that ought to respond to your question since I am not operating a broadcast automation system. I do a lot of audio editing which becomes available as "Podcast" material on a website.

When you say you normalize to 10dbfs are you saying that your peak levels are 10 dB below full scale? You refer to that as "the standard".

I measure the average RMS value of various segments that I assemble and adjust the level of each segment to be of equal value, RMS. The peaks of each segment will vary depending on who voiced the segment. Once I have assembled a podcast I will then normalize to -0.667dbfs. That is my own personal arbitrary standard, not any one's recommended standard.

I have always assumed that if I were loading material into a broadcast automation machine I would adjust the level of all components to some standard "Average RMS Value".... maybe -22db?

Am I totally off-base with that assumption?
 
I just pulled up the last one I produced. -17.96 db average rms (sine) as measured by Adobe Adition 2.0
 
The way I understand the process, is that the normalization technique used in Nexgen tries to normalize audio to -10 dbfs. I think it it is supposed to try and calculate average RMS without clipping the audio. That is the vague answer I got a while back. Unless I misunderstood. Which has happened....from time to time.
 
I'm going to really expose my stupidity on the subject here. ;D

I've never been in the same room with a NexGen so I don't know their "terminology".

There should be two times or two circumstances where automation might be expected to do something about controlling levels.

First would be when you are inputing material to the hard drive. You are ripping a CD track onto the hard drive. You've just come in from the car dealer with a fresh recording of a customer-voice commercial and you are going to stuff that onto the hard drive for later broadcast.

Is this the place where you are saying NexGen will attempt to normailize to -10?

The second opportunity that the automation-beast would have an opportunity to adjust gain levels is when it is playing back the various segments (music, commercial, weather) for output to your audio chain to either a transmitter or a web-stream. Are you saying the NexGen will take the content of the hard drive, no matter what level it was recorded onto the drive, and send the output on to a board input, or your audio processor with a line output level of -10dbfs?
 
Hi!

If your board is using regular 'ol analog dial VU meters, then what you *really* are seeing from them is a reaction to the average power of the audio.

Newer albums (and the "remastered" older ones) are taken through a process where they basically apply techniques similar in concept to what we do in broadcasting to make the newer albums sound louder than older ones, or even ones from a "competing" label or band. The electrical levels on a CD made 20 years ago vs. one published yesterday are the same, what's different is the greatly increased power (RMS) level of the new CD.

This new CD will cause VU meters to read a much higher level even though the electrical level is no different. So, to an audio processor, what it sees is no big deal to work with. What does happen, however, is this....depending on how the processor is designed, you could hear the newer music sound more "smashed" on the air vs. the older stuff as the processor is squashing a song that is already squashed -- double processing it.

The older the audio processor, the more noticeable this issue could be. Again, a lot depends on the way the processor was designed. Newer ones are generally designed with this issue in mind. In this instance, a normalization process at the ingest point that takes this power level into account can help even things out as these squashed recordings may end up being at an electrical -20 or so dB FS (FS = Full Scale [zero]) vs an older one which might hit -5dB FS. On a VU meter both will look close in level because the RMS based normalizer matched the loudness of the two cuts to make them sound even to our ears.

A normalizer that puts everything at, say -10dB will not help this issue.

Maybe this post has shed some light on what you are seeing (I hope!)

-Cornelius
 
It's my opinion that the dynamic range capability of even the most basic digital audio (16-bit) is adequate for all but the most demanding consumer listening purposes. Even allowing for peaks of 20 dB, RMS levels versus noise floor can be far superior to that of unprocessed FM, for example.

Unfortunately, that has not registered with a lot of folks who, rather than use the available dynamics, would push things up to the top continuously and indiscriminately.

This is a problem with many complex facets, and Cornelius covered it pretty well, but suffice to say, we are dealing with a demand for highly processed program signals and a supply of material that is mostly already highly processed, mixed with quite a bit of archival material that is not. For the most part, yesterday's processors aren't up to the task. Not elegantly, anyway. :D

It's my opinion that processing for such a mixture involves using intelligence (either human or artificial) to discern what processing the material already has, versus what it needs, and processing only to reconcile the difference and no more. But, hey, that's just me. ;) I'm a minimalist in a maximalist world. LOL!

Further, to equalize source-to-source loudness levels, in actuality only RMS (average power) processing should be needed, but peak processing is now de rigeur and all but unavoidable. This is for no other reason than 'everyone else it doing it' because any program that is not heavily processed seems weak in comparison. Never mind that the audio quality suffers... :'( IOW, there is very little need for peak processing, but a huge amount of desire. A lot of people don't know the difference, or are afraid to "under-process" because of what they perceive to be competitive pressures. Or they simply don't care. ;)

As another aside, RMS processing alone does not tend to aggravate data compression algorithms by generating sonic artifacts the way dense peak processing does, but, well, what are you gonna do.

So we have every logical reason to avoid excessive peak processing, but loudness trumps logic, every time.

BTW, the idea of "intelligence" in processing was recognized as early as the 1960s, [CBS Audimax] but is only now becoming recognized as no longer being a luxury, but a requirement, for the modern processor.


Kind Regards,
David
 
You should use an audioscience card with dedicated outputs on the console(usually 4) .The card assigns those outputs .It will fix the levels using the faders on the board.Never saw a Nexgen without the audioscience card being used.I think a large part of the problem is the card you're using.The audioscience has adjustable mixers.Once i ripped cd's using their cd extractor.loaded files into nexgen,set the faders,Never a problem...Ebay has the older 4300 series that will work fine.PM me, i have one i'll sell cheap to help you .currently using the 6114 card,nice....
 
Wow :eek:!!!! Thanks guys for all your input. I love learning!!! I am using TM Studio GoldDisc .All of the music I was trying to put in, at this point, are all from the 70's to 1990. As for the audio card, I have to get PCI Express. And they are expensive!! :'(
 
As for the PCI Express slot, my computer has only one of those. No others. It was the only one I could find with Windows XP .I have noticed a few songs that sound bad,not all just a few. Actually it is Player101, sorry my bad! And yes PCM.
 
now we're getting down to the real nitty gritty.You can get a dell optiplex 745 with xp off ebay cheap..That way you could run the cheaper audiosience cards listed on ebay.i run a 4332 in the prod/VT machine and the AS6114 in the dedicated on air machine(on air only)...is there not a card that will give you an extra pci slot?i stole the 6114 off ebay for like $499.00,not way i'm paying 1900.00 for one.
 
Wow! You are right on both counts! :D This computer actually has two audio outs, independent of one another. I wonder, I should try that first. If it sounds bad a computer (dell optiplex 745 with xp) off ebay. And get an audioscience card.As for some expansion or converter unit,which one would think would be out there, I can't seem to find. I will google it...
 
just an expansion card.check with tiger direct.mind describing your computer.not sure what you means by 2 outs independent of each other.how does the audio show up in device manager. please share what the audio device is.if you can get it to work as a 2nd card in player 101,fine.
 
I found one on buy.com only $32.00. Sure beats buying a computer! As far as the 2 independent audio outs , they show up as 2 different sound cards. Doable for right this second, But I would love to have better quality in the sound.After I get this adapter then I will be looking into getting a sound card. You said you had a spare on hand? Which type was it ? How much would you like to get for it?
 
Why are you compressing your music mp2? Just curious. Is your hard drive that small?

I would never trust a piece of software to normalize my levels for me. Perhaps on a short piece of production. I'm a bit on the anal side, I look at every track...spot check a few spots in it...then determine what final level I'll use. 99% spot on.

I've never been a fan of the TM libraries. YMMV. Especially anything that's had "no noise" applied to it. I find them muddy and cloudy sounding. I can point you in a couple different directions to get clean, linear audio for less than $2 a track, real radio versions, remixes, album versions, edits, mono, whatever you want...if you're interested, PM me.
 
oldiesstation said:
Watch some of those cuts on that gold library.Some of them sound like HELL.
Click to expand...
Listen to Cowsills - Hair from the TMC Gold Disc...mine has a slight pop/scratch on the RECORD they took it from.
 
Sgeirk , I was just trying to figure out what all hu-bub was about. I hear good things and bad things about encoding audio in MPEG, I listen to what people say on either side and test it for myself. Then I can come to my own conclusion. I try to be fair and non biased until I evaluate what ever might be in dispute. ;) As for TM GoldDisc ,Yes there are some,in my case not many, cuts that sound bad due to mono to stereo conversion. All the radio stations I have worked for since 1989 have used TM Discs.Although when I was first exposed to them ,I think, they were called Century 21 Productions. So I went with what I was familiar with. I don't really find them to be a major concern or problem. The only thing is the levels. They way they market them, it sounds like they are all perfectly mastered for perfect audio levels. But they do appear top be way off. :'( Which in turn, is a disadvantage , of course this is just my opinion.
 
Status
This thread has been closed due to inactivity. You can create a new thread to discuss this topic.


Back
Top Bottom