- This topic has 20 replies, 6 voices, and was last updated 14 years, 4 months ago by
.
-
Posts
-
[quote author=”timmywa”]Assuming your uncompressed tracks are .WAV, you can find a tool called wavegain ( http://www.rarewares.org/others.php ) and use it to bring all your audio to the same level. I’ve not used wavegain, but it’s partner, mp3gain has a default of -89db. Jesse recommended I lower that to -85db for what would be industry standard. If wavegain is the same, then I would recommend levels like I mentioned. You don’t need to be anywhere near 0db, as you wouldn’t have any headroom.[/quote]
That’s basically where I was headed with my next recommendation, so thanks for suggesting it. 🙂
the industry standard is 83 dB SPL, and ReplayGain does a decent enough job in getting those correct in relation to the original CD audio standard… but the default ReplayGain level is 89dB I’m assuming because they valued decreasing the dynamic range of hyper-processed stuff (which sort of seems like a battle already lost) to having enough headroom for stuff that already sounds great and in some cases gets turned up even louder causing the peaks to require limiting. :/
i recommended 85 dB because it’s between the two, but more on the side of 83dB. Tim, what percentage of songs would you say actually get turned up at 85dB?
Another benefit of what you’re doing here is that crossfades and mixing of things together you’ll generally already have decent enough headroom where nothing will have to limit the signal to prevent clipping. (or it just plain won’t be clipping anymore) For the most part. 🙂 The result of that is a cleaner more dynamic sound. Like when the jocks are talking. (unfortunately it won’t clean up the language, lol)
quote :Upon checking, our jingles are sitting at around 0 dB and so are some of our tracks (say Oasis) but we also have some tracks sitting closer to -10 dB (some Coldplay, Evanescence, etc.).My music tracks top out at about -6 with my jingles hitting -3 or so. I try to stay away from 0. Riding a song down at -10 can be difficult for any processor… especially if you randomize your selections.
Normalizing an entire library is a tough call. A few peaks above threshold can force a lot of songs down below where you’d rather have them. If you have to use a normalizer, then I guess that’s the way it is. I’m fortunate to have been able to build my library over enough time that each cut has been given its own "look" before it lands on the hard drive.
The Breakaway products really do a very good job of leveling disparate cuts. To look at the outputs, it’s amazing how consistently the level’s held at "0", while still sounding like there are some dynamics in there.
However, as Jesse says, density is an entirely different issue. You can use a multiband processor to squash and clip your audio as much as you want. Pulling that compression back out is another matter, especially when running a large library. I have everything from big band music of the 40’s & 50’s, up to current CD cuts that have enough peak limiting to glue the needle at ‘0’. Nothing can make that much range sound equal, though Breakaway does as good a job as anything I’ve heard.
The axiom applies, regardless… "Garbage in… Garbage out".
[quote author=”JesseG”]i recommended 85 dB because it’s between the two, but more on the side of 83dB. Tim, what percentage of songs would you say actually get turned up at 85dB?[/quote]
Hi Jesse,
Thanks for the recommendation. I am looking at this today, using dBpoweramp and its useful DSPs, including replaygain:
– Convert file to 32 float
– Apply Replaygain, though they offer two desired volume settings for "active" and "inactive" ReplayGain: between 0 and -40dB??? what settings would you recommend here?
– Convert back to 16 bitsHow does this sound? Do you think our uncompressed wav files will survive this? 🙂
Thanks,
Audio
[quote author=”Audio”]- Apply Replaygain, though they offer two desired volume settings for "active" and "inactive" ReplayGain: between 0 and -40dB??? what settings would you recommend here?[/quote]
I’m not sure that will do what WavGain does. WavGain already processes internally at 32bit float, and optionally dithers back to the output bit depth (same as input file by default).
Check here, abotu half-way down the page
http://www.rarewares.org/others.php
and start by reading the manual. =)[quote author=”RodeoJack”]My music tracks top out at about -6 with my jingles hitting -3 or so. I try to stay away from 0. Riding a song down at -10 can be difficult for any processor… especially if you randomize your selections.[/quote]
Rodeojack,
I have been wondering about your music tracks levels / jingles levels… How did you come to this conclusion as you mentioned it would be beneficial to have slightly higher levels at the start of a new cue… I am assuming that having equal levels on both the jingles and the music tracks would do the trick when applying a small fade prior to the segue point.
In any cases I have now managed to lower our jingles to around -4 dB using Wavegain and it seems to work much better this way! 🙂
We still have some problems on some music tracks being either too close to -1 dB or -10 dB. So we want to alter these via wavegain as well but are now wondering whether we should get at the same level as the jingles or replicate something along the line of what you’ve done.
Thanks,
Audio
- The forum ‘Breakaway Professional Products – [discontinued]’ is closed to new topics and replies.