The thread discussing poorly mastered discs has got me thinking...can anything be done by the end user to remaster a disc? And I mean without professional studeo equipment. The new Dylan disc is pretty good but, like everything else, its recorded too loud and has occasional clips. The clipping doesn't annoy me as much as the overly loud bass on every song. And I can't seem to use EQ (on my car stereo) to make it much better.

ReplayGain prevents clipping and fixes the volume, AFAIK.

The only clipping ReplayGain prevents is clipping due to lossy encoding. If there is clipping present in the source material (CD), there's nothing you can do. And you'll still have the dynamic range compression.

Depends... you will most likely not be able to fix a bad mastering.
You can't do a thing against clipping on the CD or compression and so on but you might be able to compensate things like the bass a little.
Anyway, it's most likely not a good idea and the results will likely not be as good as a re-master would be (or could turn out worse then the original).

One of the main reasons for mastering is to make it sound good on as many systems as possible, an experienced mastering engineer will likely know how to achieve that. You may be able to make it a bit better sounding for your current system, but that's it basically.
I would simply keep the fingers away from it and do little adjustments on my system (if it doesn't work in your car, well.. you could screw around, but I doubt it's worth the effort).

No, once the damage is done, it's done. Since you don't have the 1 to 64 pre-mixing tracks and/or the two pre-mastered tracks, you're left with only the ability to make minor adjustments: band/parametric EQ, digital effects, heavy-handed noise reduction, de-clicking, additional artificial compression/expansion, etc. Generally you can only "fix" minor problems at that point.

For years I have wondered if at some point audio and storage technology would get to the point where record releases would include not only the final stereo mix, but all of the separate pre-mixed tracks as well. It might not even contain the final-mix data, just instructions (software) telling the playback devices how to create the final 2/5.1/etc channel mixes from the original tracks. Perhaps even the digital effects would be applied on the fly. I suspect that only a minority of artists and record companies would take advantage of such technology, but it would open up an entirely new remixing/remastering/mashup/etc universe.

-brendan

I've always wondered how effectively expansion could be used to undo compression. It would sure be nice to see some of the curves that are used.

yes, seems "impossible" at the first sign... in this "loudness war" we,the listeners, are loosing.
for example, i like megadeth but i hate how loud all cds was done.
regards greynol and all!

@ Railroadrum

trying to help in something:
if you want,host the music with "worse" quality from your new B.Dylan(as .wav,no matter how big) and pm me the link.
i can try some "miracle" without cosmetics.
(i have big patience with audio and after i hear...who knows?? )
your choice and taste.
regards!

postfish seemed to have clipping restoration. I haven't tried the feature myself, so I'm not sure if it's just a misunderstanding or if it works well.

http://www.hydrogenaudio.org/forums/index....showtopic=22051

OT: speaking of Dylan and cosmetics...heh heh...

http://www.rollingstone.com/rockdaily/inde...blanchett-baby/

-brendan

See also: http://www.hydrogenaudio.org/forums/index....showtopic=28450

By lowering levels, Replaygain definitely can prevent one form of clipping: that which occurs in D/A converters and output stages that cannot properly handle signals at 0dB -- which turned out to be most of the CD players that were examined when the issue was first raised (in the late 1990's IIRC). I don't know how today's CD players perform in this regard.

Obviously whenever Replaygain *raises* levels, this 'prevention' does not apply -- in fact the track may actually now clip on playback, because in an attempt to make all tracks play back at a similar level, the peak level of this particular track now exceeds 0 dB. Foobar2K has an option to look ahead and prevent this from happening.

However, the clipping actually 'mastered in' to the recording due to compression and limiting during production -- i.e., what shows up in waveform views as flattened peaks more than a few samples wide, generally at or near 0 dB -- will still sound like clipping*, in any of these situations. For those there's nothing you can do other than attempt to 'rebuild' a rounded peak from a flat one, as 'clip restoration' algorithms do. It's a guess, not a resurrection of the original unclipped recording (if that even ever existed!)


(*Realizing, though, that not all visible clipping is necessarily audible. It matters how 'wide' the flattops are, and how closely together they are spaced. Sometimes waveforms look worse than they sound.)

Some peaks appear flattened, but when zoomed in are not flat.

This is why I asked about using expansion in an attempt to undo compression. Of course I realize there are other factors such as attack and release times and band limiting that make undoing compression virtually impossible.

But I am talking about ones that really are.

Also, some files that are clipped do not show perfectly flat peaks...e.g., if some postprocessing has been done to the already-clipped .wav

General pupose expanders intended to improve the sound of artificially compressed music on playback used to be sold as consumer items - none were exactly best sellers, but probably the best known were made by dbx (model numbers 1bx, 2bx, 3bx etc.) These boxes are still around on the secondhand market, and it might be fun to play with one and see if it would make a modern hypercompressed recording sound more like music.

The solution for the Dylan album which began the thread is probably to wait a month for the vinyl version <g>.