Hi

First, I have to apologize for this a little bit long post, but the following is very important for me and one reason for joining Hydrogenaudio. I know there are already several discussions according to this, but I consider this problem to be the far biggest problem of modern music.

I'm copying music since now about ten years and I guess I have copied and analyzed at least several hundreds of CDs so far. There were examples from the beginning of commercial CDs (1982) until today. The problem I discovered is that since about the year 1992 CDs get lauder and lauder, I mean the year the CDs have been mastered. This comes along with clipping and dynamical limiting. They are used more and more often and also more and more strictly. They can appear alone or in combination. The newest Pop-CD I know that really seems neither to be clipped nor limited is Suraya - On nights like this (1997).

Note: I consider a song as being "clipped" or "limited" if only one sample of this song seems to be "clipped" or "limited", if you look at it in a wav editor. Then I call the sound quality of this song "bad", although one clipped or limited sample is never hearable. But once you have started with that, you'll soon end up with heavy clipping and hearable distortions. Limiting creates no distortions, but also kills the natural dynamics.

Are there any ways out of this sh...? I use declipping functions of some programs, but most of them make sound quality even worse. Even if you're an experienced user (I hope I am) and offer much time (I do), you need great effort to improve sound quality a little bit. So this can't be a general solution. I also know no program that is especially designed to repair dynamical limiting.

I want to name an example: I like many songs of Celine Dion, I think this art is comparable to classic. But unfortunately I know no CDs from her that aren't heavily clipped (especially All the way...a decade of song).

The result is that I buy no more any new-mastered CDs (and also no remastered CDs!). Sound quality is best on CDs that were mastered from about 1985 to about 1990 - no longer analogue recordings, but still properly mastered. Fortunately, this is my favorite time period for pop music.

My main question is: Why? Why are all modern Pop-CDs so badly mastered? I say Pop-CDs, because classical music (at least sometimes) seems to be better - how unfair. But you can also consider this to be a general discussion about bad sound quality on modern CDs.

Cheers

hasnt this allready been covered that new cd's since the 90is sound worse.

then you need to provider evidence of any statement you make on this forum. see the warning bar on my one. thats what happens.

then music from

om about 19XX to about 1990 - no longer analogue recordings, but still properly mastered. Fortunately, this is my favorite time period for pop music.

is the only good pop music anyway nearly

Nobody but you, and maybe mods, can see your warning meter.




It's louder
But you are right, anyway

Yes, we all wonder why newer CDs are mastered so poorly.

The answer is that producers want the music to be louder.



Now, as to why that is, I'm baffled. A false perception of the ear is that the louder a recording, the better the quality. If you compare two samples, a quiet and a loud one, you may be tricked into thinking the loud one sounds better. (I read this somewhere here, I think I'll be backed up on this.) But this is obviously not true.

Everytime this topic comes up, somone posts a link to this article. Today...it's my turn.

http://www.prorec.com/prorec/articles.nsf/...6256C2E005DAF1C

a very nice article indeed ... i was looking for it

Spinal Tap covered this best:

"This is a top to a, you know, what we use on stage, but it's very, very special because if you can see, the numbers all go to eleven. Look, right across the board."
"And most of these amps go up to ten. Does that mean it's... louder? Is it any louder?"
"Well, it's one louder, isn't it? It's not ten. You see, most blokes, you know, will be playing at ten. You're on ten here... all the way up... all the way up... all the way up. You're on ten on your guitar...where can you go from there? Where?"
"I don't know."
"Nowhere. Exactly. What we do is if we need that extra...push over the cliff...you know what we do?"
"Put it up to eleven."
"Eleven. Exactly. One louder."
"Why don't you just make ten louder and make ten be the top number, and make that a little louder?"
"[long silence] ...These go to eleven!"
–This is Spinal Tap


That about covers it.

The issue of LOUDER IS BETTER (LIB) is only one problem with most (almost all) popular music (rock, pop, country, etc.). And most problems are purposeful as has been said of the LIB syndrome...although in my experience these recording techniques have been in use for a lot longer than is claimed about LIB.

Recording engineers have always produced popular music so that it sounds reasonably good on the lowest-common-denominator equipment (radios, car audio regardless of cost, boom-boxes, and just about any consumer home stereo from Sony, Pioneer, Bose, etc.). If you choose your music based solely on what you like in one of the pop genres, your likely to be sorely dissapointed in the sonic quality of the recording when listening on truly mid- to high-end gear.

Ever wonder why audiophiles listen mostly to jazz and classical? It's not entirely because of disdain for rock music. It's because so few rock recordings can hold up under critical listening. Sure...play it in the car on road trips or listen to it on computer speakers or on a boom-box...maybe even the cheap home stereo as background music...but keep it off the "gear". A much higher percentage of jazz and classical recordings are excellent.

I'd love to be able to listen to some of my favorite rock bands on the "good stuff", but I just can't stand the sonic quality of most CDs in the popular music categories. When I do run across a good pop recording, I feel like I've found a diamond in a pile of sh*t. It's a shame, but that's how it is and has been for a very, very long time.

actually, limiting does create distortion. that is, it doesn't prevent distortion. in fact in many cases hard limiting makes matters worse. i have an album which is hard limited to just under fullscale yet there segments where as many as 200 consecutive samples are clipped. the distortion sounds exactly the same as if the waveform is clipped over fullscale. so, just throwing a look-ahead limiter into the mix doesn't eliminate clipping and it certainly doesn't prevent distortion.

I always thought limiting was distortion because the output is not the same as the input. But I am not an audio professional, so I might have missunderstood the definition.

Edit: everything

yes... i certainly don't appreciate excessive hard limiting.

the thing that gets me is people who don't notice it don't care, and people who do notice it abhor it. so why do it?

i've never heard anyone go "wow! that CD is soooo cool, just listen to how loud it is!" in my experience there's a dial provided on even low-end gear for making a CD louder.

unfortunately there's no way to regain the original dynamics that i know of. limiting is basically compressing at infinity to one above the threshold. if it was 4:1 or even 30:1 there might be a chance that hitting the right settings on a compressor could expand out something approaching the original, but when infinities are involved, compressors choke.

i really feel like punching the people that decide to excessively hard limit. perhaps it's an attempt to sell the overpriced vinyl releases that sometimes come out with the CD. it's a sad situation when the vinyl actually sounds better than the CD. it's like digital never happened.

Unfortunately, vinyls are now cut so loud that it's a miracle that the needle can stay in the groove. The sound is completely distorded too, but rather like a woofer that would be driven too loud.
I'll post examples later, when I have the time to record, edit and compress some samples (I mean into MP3).

what's even more annoying is that hardware man's are probably taking those hot CDs as standard loudness level for the headphone output of their mp3 players (hello Euro iPod), and replaygained stuff sounds way too silent on them even with volume cranked up

Well, a stronger output signal would decrease the life of the battery, so there's more than industry-standard-volume-levels that governs portables' output loudness.

But, quite frankly, I do think the Euro iPod's volume limit is absurd. It's like saying "we shouldn't allow sharp knives, because they're not needed for typical day-to-day cutting & are only going to hurt people." Yeah, good work, let's not take into account anything outside of the norm.

I guess some folks forgot that music even has the ability to be quiet? (Is it law in some European countries to not allow devices above a certain volume amplification? I think I remember hearing so.)

I'm so bookmarking that article!

Does he get a warning for that? Mods!

edit: 'tis just a joke ok, lighten up

As has been mentioned before, this "loudness" issue need not apply only to newly-cut recordings, but remastered ones as well. I listen to mostly classic rock and have noticed that newer compilations are much louder, feature a higher level of dynamic compression, and even clip on occasion in comparison to older CD pressings and my vinyl. And this is OLD music we are talking about here, as in 1960-1985, which is sad! Will future generations even know what properly-mastered recordings sound like?

Only too true.

Bob Dylan "Bringing It All Back Home" remaster, Album gain: -8.48 dB

sigh

You want a true example of this? Check out St. Anger from Metallica, talk about pushed to the limit, but I suppose the quality wouldn't have been hindered that much because of the purposely bad mix.

That's so damn true....
Try to listen Beatles original sound album songs and hear the same songs in the "red and blue" double albums remastered collection....
The Paul's bass lines are pumped to the limit....
Classick tracks like Come together are not the same thing of a time...

This is a topic that I care deeply about. I am actually about to write an e-mail to Incubus, even though it might not get to the actual band, and probably nothing will be done about it, about this very subject. I am plain sick of listening to CDs on my dad's $20000.00 hifi system, just for it to spit noise back at me because the damn record company demanded that the record be very very loud. Audioslave's album literally hurts my ears.

Plus, I was also looking forward to buying Vapour Trails, because my friend sent me a track off of it the other day. I'll still buy it, but it sucks that I won't be able to do some serious listening to it.

Do not buy it!

Not buying is your strongest and only vote.

For purposes:
Metallica - St.Anger
replaygain_album_gain = -9.93 dB
replaygain_album_peak = 1.269472

Track 3 and 10 excluded from Replaygain test because I dont have them.

Mat

And let us not forget Audioslave.

replaygain_track_gain = -9.67 dB
replaygain_track_peak = 1.099151
replaygain_album_gain = -9.84 dB
replaygain_album_peak = 1.190704

Not only is it way over the limit, but it also is just plain distorted. I can not even listen to this CD, and I love the music.

I'm still going to buy it man, it's great music, just not great quality. My way of "voting" is going up to the record company's building and egging it. And instead of listening to it on the $20000.00 system, i'm just going to listen to it on $90 comp speakers

Yeah, as I said, it hurts my ears just to listen to that CD.

Rush's Vapor Trails has got to be one of the worst CD's I have ever owned in terms of shoddy mastering. I can't remember the exact replay-gain values, but they were all close to -9 or -10dB, IIRC. Geddy Lee even brought this up in an interview shortly after the album was released, claiming that distortion occurred when they (the engineers) tried to get the effect that they were aiming for. (What effect, a jet engine taking off? ) Sadly, although Rush is one of my very favorite bands (as if one could not tell by my alias and sig. ), I just cannot bear to listen to VT. Thus, it's probably the only Rush release that I am not all that familiar with. Unless a properly-mastered version is ever released, I doubt I ever will listen to it very much.

The Sisters of Mercy - Some Girls Wander By Mistake ©1992
3. Phantom
replaygain_track_gain = +0.62 dB

There are also two short interlude type songs, one which goes as high as +14.89 dB. Granted these songs were recorded much earlier but having been mastered in '92 I would like people know there are exceptions and not to make all those blanket statements about music going to crap. I would have to say though that most recent albums of the type I listen to have clipping problems. It is an unfortunate trend in the music industry that quality gets backburner to appeal.

Another example:
Synaesthesia - Desideratum ©1995
Disc 2 - 2. Subversion
replaygain_track_gain = +0.40 dB

[edit: spelling]

I love it too - I just listen to it in my car with two 12in subs - I don't think it sounds that distorted to tell the truth. IMO, the most compressed album I have ever come across is "Red Hot Chile Peppers - Californication" - that just sounds bad.....

AHA!!!! so its not only me


why does it sound so bad? is it because of distortion??

Yeah....open up the song "Parallel Universe" in a wav editor....95% of the song is at 100% volume.... the entire 3-4min section of the song is at 100% volume...just plain horrible

heh, replaygain values for that song:

replaygain_track_gain = -14.29 dB
replaygain_track_peak = 1.370208
replaygain_album_gain = -12.98 dB
replaygain_album_peak = 1.387573

That's just...sad.

That actually manages to beat Animositisomina.

8. Impossible
replaygain_track_gain = -12.35 dB
replaygain_track_peak = 1.321746
replaygain_album_gain = -11.94 dB
replaygain_album_peak = -1.502868

Although it does lose to the album peak above.

The Rush in Rio live CD might help with that. That is, if you can overlook the the fact that the audience volume level is a bit too high.

Is it just me, or does anyone else think that the last two songs on Rush In Rio (the soundboard extras) actually sound better than the rest of the cd?

and if you think californication was bad enough, listen to what they did to the tunes on Greatest Hits.

Hi again,
unfortunately, this thread has'nt become what I really wanted.




As I already said, the topic is important enough.



Proove something that is so obvious and accepted by most HA members ? But what should I do ? Offer some graphs or portions of audio for downloading ? This would be not so easy for me because of the circumstances of my internet access. But would that be a real proof of "all new Pop-CDs (or nearly all) have bad sound quality"? By the way, my warning bar is still 0%.



The answer can't be just so simple. Can you imagine that nearly all mastering engineers have the intention to make pop music sound really bad ? Or is it just sloppiness ? My personal suggestion is: They use superior formats (like 32-bit float or so) with samples above 0dB and then just convert down and clipping occurs.

Does anyone know which format record companies use for storing records for future ? Possibly in a clipped form, so they are destroyed forever.




So... But I can still notice digital distortions when listening to a so called boom-box. Every boom-box and low-end equipment offers better quality than today's Pop-CDs. What do you consider to be "real" mid-gear, FooForThought ? The difference between a common single audio-CD player from Sony, like you said, for example, to high-end seems to be really small; not worth the costs.




Outscape, are you really sure of that? The basic difference between clipping and limiting/compression, it doesn't matter if hard or soft, is that clipping only affects samples above the clipping level. But the dynamical processes reduce the loudest peaks this way, that even the lowest samples of the peak (near 0) are affected. The lower samples are reduced in relation to the higher. So in your example, there seems to be both limiting and clipping. I know an overcompressed song sounds awful, but not distorted.




OK, my statement is only partly true. The song "It's all coming back to me now" on the album I've mentioned was written by Jim Steinman, he's a musical composer.

Cheers

I thought the exact same thing, but didn't bother to reply.

Given a level matched comparison, just how bad does the equipment have to be for the compressed/squashed mix to sound better?

The answer is: bad enough that it doens't have enough power to handle the peaks.

There is some equipment out there like this. But so what - it's relatively rare* - and owners of this stuff should buy better equipment - not cause the rest of us to listen to poor quality audio all the time!

* most equipment - even cheap stuff - can cope with short loud peaks, but not continuous sounds at the same level. It varies though.

Cheers,
David.

Most known for his work as the behind the scenes guy for Meatloaf's Bat Out of Hell. Also for an aborted attempt at co-songwriting and producing Def Leppard's Hysteria

clipping and dynamic range aside, my other beef with popular music is how it's mixed in such a way that nothing really goes into the low bass or high treble. Basslines become humming midranges, bassdrums are reduced to unexciting pulses, and trebles just become wishy washy 5khz-8khz sounds. As several have mentioned, play it on a great stereo and it sounds as unexciting as a cheap boombox :/
Just one of the several reasons I don't bother with RIAA music anymore.

This refers to my previous post:



I've done some limiting & compression tests; with the programs that I use I actually receive no distortions:
Convert a clear (neither clipped nor limited or compressed) *.wav to 32bit float.
Normalize it to, say, +30 dB. Now it sounds hopelessly clipped, but it isn't.
Apply a very hard limiter to, say, 0 dB.
No distortions are audible or even visible.
So you receive the same reduction in dynamics, but no distortions. (Compared with the one that sounds clipped; it sounds exactly the same as if you would convert it down to 16bit.)

I have to vote for a new worst album - "Jay-Z - The Black Album" . The whole album sounds like it was pushed 10db over what it should have been - the bass is totally ruined thoughout the entire album. Heres a pic of one song....



Track 4 - "Encore"

replaygain_track_gain = -12.46 dB
replaygain_track_peak = 0.999969
replaygain_album_gain = -11.10 dB
replaygain_album_peak = 0.999969

ouch

http://members.cox.net/sppimley/Badly%20Cl...0Waveform-3.jpg

Here is a good example. Now I forgot to take a snapshot of the beggining, but when zoomed the song has marginal clipping which I can stand. For whatever reason that I cannot fathom there is an obviously stepped effect as the song has been limited more and more towards the end. By the time you get to the end...it looks like this:
http://members.cox.net/sppimley/Badly%20Cl...0Waveform-2.jpg

As it is a really good song otherwise I just cannot understand why someone would butcher it during the recording process like that. Notice the peak level is 93.8%, yet entire peaks of the song have been sliced flat well before they reach that.

ARTIST=Judas Priest
TITLE=Tyrant
ALBUM=Sad Wings Of Destiny
TRACKNUMBER=06
DATE=1976
GENRE=Metal
REPLAYGAIN_TRACK_PEAK=1.54149842
REPLAYGAIN_TRACK_GAIN=-7.53 dB
REPLAYGAIN_ALBUM_PEAK=1.54149842
REPLAYGAIN_ALBUM_GAIN=-8.97 dB


Yeah... I love Repertoire Records....

"We guarantee that our releases represent the best value for money in classic Rock & Pop Re-issues without any compromising on quality and information."

edit: keep in mind this file was compressed lossy - this may have introduced additional clipping.

I just bought a polish compilation called "In Goth We Trust", and surprisingly, few tracks are overcompressed, though most are from 2000 - 2002.
It seems that Poland still escapes the loudness war.

For example, this is one minute of audio from Artrosis - Kolej Rzeczy, 2002



EDIT : ...and save for the electronic sounds, this could be classified in "metal rock" !

Sorry, I realize that the above track is a bit misleading, since it is actually taken from the DVD Video (AC3 2.0, HeadAC3he, no compression, normalized to 100 %), not the audio CD.

Here are four tracks from the CD, which are still correct, exept the number 01, that is overcompressed (and clipped, I had to set the scale below 0.96 in MPC in order to avoid clipping warnings).

From left to right and up to down :
Released in 2000, 2002, 2000, 2003.