i ripped some music as explained here using eac+mpc but when i check the foobar console all tracks say:

INFO (CORE) : Clipping detected.


i never hear anything wrong with any of my music and i did rip correctly, so why does foobar always say this to nrealy all my music? mp3 as well?

can i do anything to stop it (whatever it does cause I dont hear nothing wrong)

Did you use --xlevel switch in mppenc command line? If there were any clippings it should have beeped during the ripping+encoding process.

Even then you can use Foobar's "Advanced Limiter" DSP plug-in. Please use the search feature .

BTW I hope you like fb2k. I am a fb2k n00b too. Did you check fb2k format string site?

Use ReplayGain.

No, it would not!
It would have beeped if there were "internal clippings" in the encoder, but that is a totally different thing.

Germanjulian: The best way to avoid those messages (and the clippings) is to check the "Use peak info to scale down tracks that still clip..." option in playback preferences in foobar.
You need to replaygain your tracks for this to work correctly though.

Using Wave Gain before encoding will also solve this problem.

Alternately, you can use Advanced Limiter or Soft Clipping Limiter DSPs to prevent clipping.

I take it you listen to a lot of loud music?

Is clipping bad? In my opinion, most of the time, clipping would result in noises that falls below the masking threshold.. It may not be that audible..

From what I experienced, clipping over short and isolated areas (e.g. 10 samples) is not audible at all ... but this can dramatically change if this happens repeatedly (like this is the case with modern pop compilations) ...

You mean the clippings very often, resulting in noises that exceeds the masking threshold of the decoded clips?

Hmm ... "loud" music with low dynamics (which is the kind of music that will most likely show clipping, even when encoding with -xlevel) will not suffer from noise over the masking threshold (that noise would have to be veery loud then) but the "DC" signal amount in clipped areas can be audible due to a totally messed-up impulse response ... that's my experience, at least.

If you take a discrete 1 kHz sine as an example, clipping becomes audible very fast ... then you will hear some ugly high-frequency noise besides the original tone ...

A single sine tone might have been a over simplification of the masking effect of clipping noises in a complex pop music clip.. Naturally, you will hear high frequency clipping noise for a single sine tone.. But what if you have many sine tones spread across the spectrum and each of them clipped? (I have not try this out yet!! )

i mostly listen to metal and classical music (yeah yeah i know)

i did use the correct mpc switch with the --xlevel

i never hear anything wrong with any music. ok lie! i do hear stuff. i ripped the jazz album i bought yesterday and listened to it tonight. sounded perfect just like the cd. excepth at song 3 at the 1.48 time mark when there suddenly was a short clickish drop in music for a millisecond. but wait its not mpc its acctually the cd and some sound engineer not doing his job its very very short but its there.

and clipping i have never heard. and i do have evry good ear and equipment, and i can listen very well....

1. Do not worry about clipping too much ... if it occurs over a short area of samples (the peaks are max. 10 samples long), it won't be audible

2. If clipping heavily occurs in the original wav, you could run some clip restauration algorithm (some wav editors offer this) but these algorithms will alter the wav files ... just like normalization does

Did you use MPC's ReplayGain, germanjulian? It's also explained in the sticky thread. Ask if you need more information than you can find there.

There is no point in using wavegain for mpc encoding. Use --xlevel instead (wavegain will add raise the noise floor of the original file, at least in theory, - not claiming that this will be audible but what the heck

The better alternative to WaveGain is MPC's Replaygain, not --xlevel. --xlevel can only get rid of format-dependent internal clipping, not the clipping that foobar complains about.

Actually, how does clipping in time domain effects the frequency domain? It would generate noise (no doubt) spread across the entire spectrum.. But would it cause some tone components to lose some of its signal strength? Would it change the phase of some tone components? This form of distortion if exists would be serious!

I get clipping, that I can hear... and It's never happened before. I think maybe the mp3's are damaged.