Kung Fu 2: Return of the dB.

*please hum the theme to 'Enter the Dragon' for this post*

Ahhh.....Young grasshopper has studied 'Lame' style with the '--alt-preset extreme' technique. My kung fu is getting better. But a new beast has shown himself. Volume difference from album to album. I have begun training with 'MP3gain'. Am I still on the right path to Enlightenment? I fear my Husker Du MP3's will not sound as strong as my King Crimson MP3's. Enlighten me. I will remain in praying mantis stance awaiting your lessons.

One of the best parts of the FAQ explains this, starting right here and ending at the bottom of the page.

In short, don't worry about normalazation that much, and use VolAdj, part of the reason Hijack exists in the first place.

I prefer to just use the volume knob. While Tony's new VolAdj settings are much better than the original presets (and likely ideal), they still pump way too much for it to be enjoyable to me when I'm really listening to the music (as opposed to having it be background noise).

What I am trying to defeat is the old 'SST' recordings (Husker Du, Black Flag, 'early' Sonic Youth). If you ever heard a CD from the 'SST' label you would understand what I am talking about. It's just weird how low the levels are on these albums.

I know exactly what you're talking about. I have many on my empeg. I don't think that there's going to be any way to get those to sound more like the more professionally produced tracks beyond having massively different EQ settings for those songs.

True, true, they were are done in a horrible studio, but thats not really what I mean. The volume is very low on these recordings, very low.....

One of the best parts of the FAQ explains this, starting right here and ending at the bottom of the page.

In short, don't worry about normalazation that much, and use VolAdj, part of the reason Hijack exists in the first place.

Perhaps it's a US/UK thing, but I find unnormalised CDs are much more common than the FAQ makes out. I appreciate the point of the FAQ entry that linear normalisation doesn't always solve the problem completely, but anything you can do to mitigate it at rip time rather than playback time is a good idea, as it will provide improvements when (sacrilege!) you use those MP3s not in a car player -- for instance, in your Receiver or your Ipod.

Peter

I agree that if you do have some albums that weren't normalized to begin with, then normalization can help a little bit. But even if you normalized all of your MP3s completely, you would still have the exact same problem.

use those MP3s not in a car player -- for instance, in your Receiver

Or you could use VolAdj on the Receiver as well...

http://www.norman.cx/receiver/

(I must get around to adding a UI to change the VolAdj levels though, or at the very least recompiling with Tony's new low settings)

Ah, cool, Andy. I was meaning to ask you about that. I'd certainly like to be able to use similar parameters to the latest hijacks. A user interface is not completely required but would be nice.

One option you might consider for these is to use Cool Edit or a simliar tool to amplify the tracks that are recorded at a low level. It works fairly well without to much of an adverse affect on sound quality.

If you save the original files there is no down side. You can always revert back to the unmodified files.

use Cool Edit or a simliar tool to amplify the tracks that are recorded at a low level.

As stated in the FAQ, this will not work because if you amplify an already-normalized track, you will clip the peaks.

If the track has not already been normalized, then normalization will amplify the track just below the peak-clipping point, that's the whole reason for normalization.

If you're using cool edit you may as well apply a compressor and then normalise that. Compression is an art of it's own. be careful how you use it.

I stand corrected...

I have used this for my own digital recordings but, they were not normalized.

I have not actually tried this for the MP3 music. I just turn the volume knob listen and repeat if necessary.