Hello Everyone,

I've encoded several thousand normalized wave files with --alt-preset extreme -Z. I have analyzed them with MP3Gain and have no problems with that. Now I want to Gain Standardize the wave files with DiamondCut Live 5, a program I trust with my wave files. But it has no analyzer.

My question: Can I not use the mp3Gain analyses? Would the analyses of extreme -Z files be an accurate reflection of the gain change needed to standardize the Wave files, since the Waves are the parent files of the extreme -Z mp3's? Or, perhaps, would it close enough not to worry about it?

I'm fine with MP3Gain's 1.5dB increments. I can hear some difference in the increments but it's OK even for the highest quality MP3's, though I do have to do quick listening tests on the files, in order to to adjust a small percentage of files (so far, almost always down) into another increment bracket. But since the Waves are the Masters, and will be archived in Ape, I want to be more picky with them.

My software handles Normalized Gain Scaling beautifully, in increments of 1/10 dB., and it is scaled 1 - 100, like the MP3Gain analyses (I think).

What do you guys think? Can I trust the MP3Gain analyses, and use them as a benchmark for the Wave files?

Thanks.

Why don't you try WaveGain? And it's not limited to 1.5 dB steps. I'd recommend album gain, medium noise shaping, logfile.

BTW, your terminology is a bit confusing. Normalizing is usually not associated with the ReplayGain process, but rather with the "dumb" peak-normalizing of your average audio editor.

I have nothing against WaveGain, but according to the readme it is not completely lossless, and though it may still be OK, I already know what the results will be with DiamondCut as I helped develope the new version, and I trust the math behind it.

But the biggest reason is I don't want to go through the laborious process of analyzing the waves when I've spent this time with MP3Gain and have run the analyses, etc., including a comparison listening test on each file to adjust the MP3Gain level to the CD volume using the highest quality CD files (I know which ones since I've both heard them and seen them), and don't want to have to re-edit the Wave files. It seems to me I could use the results of the MP3Gain analyses as a benchmark and "convert" the results to the Normalized Gain Scaling filter in DiamondCut. (BTW, by 'gain normalization', I mean the peak waveform in the file is made to be 0dB, the rest of the file < 0dB, accordingly. The Normalized Gain Scaling filter is somewhat different than gain normalization, more complex, and not just a gain changer. And FWIW, if I had to do it again (please, No) I wouldn't alter the gain in any way until I knew the target peak level of each file first. ).

But what I don't know is how --alt-preset extreme may affect the file, and if that somehow would throw the gain proportions off as opposed to a wave file that has been losslessly unaffected. My guess is if there is any alteration in the gain through encoding it is too little to worry about. I know I have never audibly detected any difference. And MP3Gain is supposed to be lossless. BTW, Dibrom wasn't sure either, though I doubt my question was clear. And I know you are very knowledgable, sooo.....

Sooo.., now I'm wondering whether to just start over and get the Wave files right, then just encode new MP3's, MPC's, etc. By far, the most important thing is to get the Waves right and archive them. But that would require hundreds of hours more work. I just don't want to rip all these songs and have to edit (silent lead ins & fades, noise reduction operations, etc.) them again. That's where most of the time is.

So CiTay, if you or anyone has any ideas or suggestions, I'd appreciate it. Thanks for the reply.

IveyLeaguer

Wavgain is good, and mathematically sound.

It's only lossy in rounding the very last bit and applying dither to avoid audible harmonic distortions caused by truncation. This is what you want.

Foobar2000 audio player will calculate ReplayGain values and apply them as a tag (or even just store it in the fb2k database) without modifying the audio (this lets you choose album or track gain at time of playback). If you want to output scaled WAVs, its diskwriter can apply gain and can add dither, just like Wavgain. Perhaps this is what you want - don't apply any changes, just do the calculations. Fb2k can also apply replaygain tags to lossless files, like Monkey's Audio or FLAC.

ReplayGain is psychoacoustic, in that it uses the equal loudness Fletcher Munson curves (simplified). MP3 encoding is also psychoacoustic - it's not an exact representation, but inaudibly different.

The replaygain value will (in my experience of foobar) be within about 0.01 dB from lossless to lossy, and usually exact.

I haven't made a final decision and maybe you can help. The background:

My purpose is to long term archive about 2000 original Waves. Though they have all been edited to some extent, at minimum every file has had 1 second+ of silence added to the beginning & adjusted to ~3 seconds of silence at the end, and the gain has been normalized to peak @ -0.20dB. Many noise reduction operations were performed, mostly on older recordings. Many more "softening" of first notes or audible sounds, improved fades, and other results of sloppy studio production have been corrected. But the integrity of the original recordings has not been compromised in any way. No enhancements were done, that is to say no equalizations, filters, etc. were used to improve the sound. It's just that I have better reproductions of the originals due to the hundreds of hours I have spent on this project. So, as you can see, it's not like I can use the original CD's for master backups, they're there for an emergency or catastrophy, I suppose, but the original master recordings will be the files we are talking about and are superior to the original CD's. From these masters mp3's, mpc's, etc. will be cranked out.

How would you go about balancing the gain on these files? Or do you think I should just leave it as is? Do you have any other suggestions?

If I had it to do again I may not have gain normalized the files, I may have waited and lastly balanced the gains. But I have found that listening is required anyway (with gain adjustment), and I'm in a totally lossless environment.

Right now I'm leaning to 'balancing' the gain on these files so it's done once and for all, and there'll be no need to replay gain in the future. But I want to be sure to do it right, and I prefer to eliminate the task (and expense) of making copies as they are now, just in case. I'd rather just use WaveGain, or whatever is best, and do it right just once. What do you think?



If I understand what you mean, this is fine and I have no problems with that. So it is, for all practical purposes, lossless and will not affect the integrity of the originals? I just saw the lossless disclaimer in the readme and it scared me off. Perhaps I didn't read it thoroughly.

Why do you want to replay gain the 'originals' if you can replay gain the mp3s, mpcs, or Flacs you create from the same waves?. (If you use flac for storing and add replay gain the flacs the process is lossless)

Thanks for the reply. Please see new topic "Archiving and WaveGain" (in General) as my original question and purpose for this thread has changed entirely.