CD dynamic range

[Dan B]

Provided Link: http://en.wikipedia.org/wiki/Loudness_war


Is there an easy way to measure if or how much a CD is compressed?

[bangleiii]

It's a conspiracy to make a CD sound like
a tube amp with a record!

[AJ]

Provided Link: http://www.mindspring.com/~mrichter/dynamics/dynamics.htm


This place (check the link) showed it very well. I have a program that i downloaded from somewhere called Right Mark Audio Analyzer; does somone know if you can do it with this?

[Paul O]

> Is there an easy way to measure if or how
> much a CD is compressed?

Despite what many POP artists and record producers have done to recorded music of late, the medium is capable of delivering 90db+ dynamic range. Defies logic whay they feel it's necessary to cram everything into the last 1db that spans just under digital saturation and just over it.

[billfitzmaurice]

There is a difference between the dynamic range potential of a CD and the dynamic range of the material recorded upon it. One advantage to CDs is that they can be recorded with a very high dynamic range, ie., the difference between the lowest level and highest level, without hitting the noise floor, much more so than with analog media. But having that range available doesn't mean that the recording engineer/producer will take advantage of it. So while a CD has the capability of a 90dB differential between the lowest and highest levels recorded on it in most cases only some 20dB of that potential is actually utilized.

[marlboro]

And the reason for that would be?

I should think that most playing systems, meaning speakers, are not capable producing the dynamic range from very quiet to very loud. Most speaker systems that people use have a very limited playing dynamic range.

Speakers with sensitivities in the 80's are simply not sensitive enough to play the soft passages unless you turn up the volume, and when you do that the loud passages become intolerable.

Unless you are using an array design or a horn design this is your dilemma. For awhile Audax was selling a midrange speaker with a 101 SPL which was a great match for many tweeters, but good luck finding one now!

Marlboro

[Dan B]

How can I tell if the recorded material on a specific CD is limited or clipping other than listening.

I've tried this:
<A HREF="http://www.qsl.net/dl4yhf/spectra1.html">http://www.qsl.net/dl4yhf/spectra1.html</A>

But maybe something that displays the wave form would be more suitable for a blue collar guy.

> There is a difference between the dynamic
> range potential of a CD and the dynamic
> range of the material recorded upon it. One
> advantage to CDs is that they can be
> recorded with a very high dynamic range,
> ie., the difference between the lowest level
> and highest level, without hitting the noise
> floor, much more so than with analog media.
> But having that range available doesn't mean
> that the recording engineer/producer will
> take advantage of it. So while a CD has the
> capability of a 90dB differential between
> the lowest and highest levels recorded on it
> in most cases only some 20dB of that
> potential is actually utilized.

[AJ]

How do the speakers sensitivity make a differance? Whether the speaker is 80db 1w/1m or 90db 1w/1m shouldn't make a differance if the drivers have good dynamics.

[billfitzmaurice]

> How do the speakers sensitivity make a
> differance?
They don't. The dynamic range/compression ratio used by the producer is purely defined by his/her personal taste and the desired end result.

[bill]

> Is there an easy way to measure if or how
> much a CD is compressed?

If it's been overly compressed, it'll sound awful. The percussion will sound squashed, and the overall sound will pump up and down with kick drum hits, among other things.

Compression was originally developed for AM radio, where the amount of compression was directly related to how far the station would reach (the more you modulate AM, the more power you put out, up until you reach clipping of the carrier.) This was later extended to FM, because the louder stations tend to jump out at you as you scan the dial. The phenomenon got really bad in the 70s as FM started to supplant AM and canned formats replaced the original free-form album-oriented rock that started on FM with the FCC's simulcast prohibition of 1967.

If you want to hear the dynamic range that a CD is capable of, listen to Rickie Lee Jones' album Flying Cowboys, especially Ghetto of my Mind.

Best regards,
Bill

[Pete Schumacher ®]

> And the reason for that would be?

> I should think that most playing systems,
> meaning speakers, are not capable producing
> the dynamic range from very quiet to very
> loud. Most speaker systems that people use
> have a very limited playing dynamic range.

If we set the uncompressed upper limit of our point source speaker at 105dB, then most speakers we build will have the dynamic range available to fully reproduce the CD dynamic range, except perhaps for the lowest notes.

> Speakers with sensitivities in the 80's are
> simply not sensitive enough to play the soft
> passages unless you turn up the volume, and
> when you do that the loud passages become
> intolerable.

Based on what reasoning? Marlboro, if you have enough power and the speakers have enough displacement, then you'll have plenty of dynamic range.

> Unless you are using an array design or a
> horn design this is your dilemma. For awhile
> Audax was selling a midrange speaker with a
> 101 SPL which was a great match for many
> tweeters, but good luck finding one now!

I knew that was coming. It's purely a function of the ability of the driver to compress the air, not how sensitive it is. An 85dB/2.83V 6" woofer with 8mm Xmax will play much louder than a 100dB 6" woofer with 1mm Xmax at 200Hz.

> Marlboro

[Andy_G]

> It's a conspiracy to make a CD sound like
> a tube amp with a record!

not at all.

The reason that a lot of recording cram the sound into the top level is to make the recording sound "loud".
There are enough "gooses" in the world that think loudest is best, and if you put on a highly compressed "loud" recording, they will prefer it over a less compressed "softer" recording...purely because its "loud", and sounds better on their Bose system, (which in basically incapable of any dynamics anyway).

[Dan B]

http://upload.wikimedia.org/wikipedia/en/9/9e/Exo-politics.PNG


I spite of the fact that I don't have a line array or horn loaded speaker ;) .....

Can I load a song into Audacity like they did at Wikipedia to figure this out?

[brianp]

. . . with input level meters, put in a tape and set it on record/pause. Play the CD and watch the meter. Set the record level so the quietest passages just barely illuminate the first green LED (or just barely twitch the needle at zero, in the case of VU meters), and watch how high the loudest parts drive the meter. Highly compressed recordings won't drive it far at all, while something like Telarc's 1812 will spike it hard. The meter will be calibrated in dB, so you can subtract the lowest reading from the highest to estimate the range on the disc.

> Is there an easy way to measure if or how
> much a CD is compressed?

[marlboro]

It sounded good in my mind, but after posting it, I couldn't stack up enough evidence to support it.

But I left it there anyhow for the discussion value.

I knew you'd catch it Pete, if anybody did.

Marlboro

[kene]

> This place (check the link) showed it very
> well. I have a program that i downloaded
> from somewhere called Right Mark Audio
> Analyzer; does somone know if you can do it
> with this?

Good site there; I already knew different digital masters could have different degrees of distortion obviously but I never knew companies actually used different masters for different parts of the world! It's one more thing I really don't want to worry about.

[damkor]

> There is a difference between the dynamic
> range potential of a CD and the dynamic
> range of the material recorded upon it.

True, and if a band is playing all out, fortissimo the whole time, recording live, you will not notice a whole lot of dynamic range. So, any measurements have to be interpreted carefully.

But you can usually tell by ear when a recording has been upped in level and compressed: "Softer" passages, like when the drums and guitars stop playing for a bridge or something don't actually sound any softer. And it is rather bland and irritating. Try No Doubt's "Return of Saturn." Average level pumped all the way up, very little dynamics.

Also, most musicians are NOT audiophiles. They could care less.

[donradick]

> Is there an easy way to measure if or how
> much a CD is compressed?

Here are a couple of links to expert analysis on CD compression. One guy even claims to have an application that will measure audio compression.

<A HREF="http://www.cdmasteringservices.com/dynamicrange.htm">http://www.cdmasteringservices.com/dynamicrange.htm</A>
<A HREF="http://georgegraham.com/compress.html">http://georgegraham.com/compress.html</A>

My favorite:
<A HREF="http://www.itwriting.com/blog/?p=91">http://www.itwriting.com/blog/?p=91</A>

HTH,
-Don

[kene]

Raise our young people to be audiophiles early.

Reprinted from your last site:

"I’ve noticed that highly compressed/loud mastering is common on material aimed at a young pop audience."

I've noticed the exact same thing, so even when a song sounds kind of fun if it sounds top 40's-ish I don't buy it. Thing to me is that compression makes a lot of sense on mass-fi equipment with narrow useable dynamic range. Kids don't get to use Daddy's big rig very often or are listening to cheap portable headphones with lots of HF hash.