Check out this waveform

I recently got an email from a fellow King's X nut who mentioned that the first CD of the new Dream Theater album was too compressed and lacking dynamic range. Coincidentally, I was extracting the first CD to disk today and noticed that the first song, The Glass Prison, looks like this:



Yeeeeouch!

I know that's showing a near 14 minute tune in one window, but it doesn't look much better if you look at the .wav in two minute increments. Look at those levels!

OK, so this track probably wouldn't have much range no matter how it was done. But this is a bit crazy, no?

BTW, when extracting the tracks in EAC, I noticed every song on the first CD peaks at 100%. I haven't checked disc 2 yet.

Mixed by Kevin Shirley, mastered by George Marino.

 

Wow...

I'm curious.. if you zoom in, do you see any spots where it's clearly been clipped?

 

The CD is probably not clipped but this is typical of most CD these days, with boosting levels by as much as +6db or more. The playback meters pretty much stick at -1 or 0db. No dynamic range at all! All for the sake of trying to please the kids.

 

Can you stretch it out enough to show us a recognizable waveform for a few seconds? TIA!

 

I've seen similar commercially-sold CDs and, when zooming in, see that there is clipping. The clipping tends to be infrequent and of very short duration, but it's still a lousy way to do things.

 

It may not be concidered "compression" persay, but it's still trying to shove 10lbs of sound in a 8lb bag. Ah, sonic gluttony.

Yep, you hit it on the head, Patman. Most any commercially available title that's been recently done (not just remasters, new titles, period).

If he zooms in to a 5 second clip, it's gonna look like a crue cut. Yep, clipped. Is that why most recent Cds sound like an electric razor?

"Who needs dynamic range if we just want it loud? Besides, our survey says it sounds best this way!"

 

Looks like a bad case of "These go to eleven"

 

Wow, exactly. All things good and bad can be blamed on Spinal Tap.

 

Simply boosting levels won't change the dynamic range. A CD can max out at 0 db without a problem. It's just that has to be an unclipped peak.

It's when those peaks are clipped/compressed that dynamic range is lost.

 

Ahhh, but compression and limiting reduces dynamic range.

I also maintain that most CDs that are maximized are NOT clipped. The meters may read them as such but they are not.

and, Steve is right. This is not good 'ol creative compression/limiting, it's all about the volume game.

 

Thank you Patrick for illustrating so clearly what is wrong with most new CDs today! I can't imagine that's what the waveform of the original master tape looks like. All I see is peaks and no valleys! This is why we need labels like S&P.

Breath of life, indeed! I'm afraid that CD has flatlined!

 

Patrick...

Please just confirm, because far be it from me to assume... Is that the CRAPPIEST sounding recording of all time, or what?

 

I found an example of this about a year ago.

I noticed on the Van Halen HDCD remasters that came out last year the "peak level" was pegged at about 98-99% for most songs. I use a program called audiograbber (not EAC) to rip my CD's to hard disk.

If you play back the created .wav file in winamp +3db at just about any part of the frequency range, they will actually distort pretty badly.

 

I agree with you Steve. The graphic shows me why I listen to the music from the good ol' days rather than my "new country" CD's.

 

Umm, yeah. Who ever said otherwise?

Once again, simply moving the volume up isn't a problem. Some people here seem to think "oh, if the peaks are at 100%, my CD is compressed beyond belief." Not true (necessarily). If the *average* volume is up around 95% or so, then you've got a big problem. However, a peak going to 100% is not a problem at all.

 

Does this look more like it?

 

Luke, I know what i'm talking about, we both know about this.

I'm not speaking to you specifically, but if anyone doesn't know, compression is simply making softer sounds louder and louder sounds softer. This can be done to isolated segments of the frequency spectrum. Compression can have the effect of softening the sound. Overuse can make the music sound dull and edgy.

The digital ceiling (100db fs) can be hit with no problem as long as there are no "overs", or clipping. The usual method for boosting levels as high as the ones illustrated on this thread is to either compress the sound to increase the average RMS, then boost it even more with limiting/compressing the peaks to prevent clipping; or just boost the levels and compress/limit. Either way, if the parameters are set there should be no clipping. A good practice, one that isn't always followed, is to limit the peaks to around -.2db. Some converters cannot negotiate FS digital level and allowing some headroom may prevent clicks and other distortions from being produced by the D/A converter in question.

If you work on a DAW you can boost levels and limit peaks to a certain degree without adversly affecting the sound. But make no mistake, when this is done, the dynamics are being manipulated. But these days, it's just going too far! Then, there is the good case of not doing anything at all except boosting the peaks to 100%. This will not change the dynamics. It's just that you really don't want to do this in 16-bit because you can degrade your sound. Whether this is audiable to the listener is another issue.

If I am preparing a CD-R, the question of whether to boost the RMS depends on a few factors, like the overall level, the type of music, the quality of the recording, and the intended end user. And then, I do extensive experimenting to ensure that the musical content is not adversly affected by playing with the release time and amount of boost.

I won't even get into the problem of frequency response and EQing.

 

Bradley, the image you presented shows a file that is still limited/compressed but not as severly as the first one. Unless the file itself is clipped, the file you show has sufficient headroom for most converters to negotiate the file without doing clipping on it's own.

 

Thanks Grant.

 

One of my favorite examples of "bad compression" used to the extreme is Santana's "Supernatural" album. Although a huge commercial & critical success, this is a textbook study in the shortcomings of today's recording techniques. This album should have been entitled "Unnatural" because the use of compression to punch-up the loudness removed any natural subtleties in the dynamic range.

 

I can't even listen to that CD! I get fatiqued after just a couple of minutes of it.