PDA

View Full Version : Educating People: MP3 Compression and other Issues


June 13th 12, 08:29 PM
I'm an audio-video NUT. I've had several jobs as an AV technician or sound man and know a few things about the subject.

When it comes to MP3s most people simply don't understand exactly what kind of compression is going on. This is why I feel "compression" is a bad word to describe the data parsing that goes on in MP3 files. There is little to no DYNAMIC compression via MP3 files. There is DATA compression and psycho-acoustic masking going on, but the only dynamic compression I've heard is on a MP3 ripped from a digital "remaster" file or CD.

The only area dynamic range *might* be affected by mp3 is on lower bitrate(128kbps or lower) files. A lot of high-freq and some low-freq information is thrown away at or below that threshold, so you may reduce the dynamics of instruments occupying or reaching into those realms. On 192kbps and up, I honestly have not heard any such loss.

On another topic: Display Calibration.

First off, most "calbration" can be done by the user following the steps on a DVD or Blu-Ray. Most flat panel displays can be taken out of "Vivid" or Retail mode and the Custom mode will allow precise tweaking of each setting. Professional(back of set or service menu) calibrations costing hundreds of dollars are not necessary unless you have a set that no amount of tweaking seems to yield an accurate picture. Or, you just absolutely must make sure everything is adjusted optimally.

Secondly, unless you are color-blind or have trouble seeing things at normal light intensity, there is no such thing as "personal preference". Period! Your evening news reporter should not have rings around their eyes nose & mouth(sharpness) nor should their face color approach that of Barney the dinosaur or the Incredible Hulk(tint setting by eye - big NO NO!)

There is only one correct position for each setting(Brightness, Contrast, Sharp, etc.), and those positions will be determined by both test patterns and your local room viewing environment.

The goal of video calibration is to align your TV/Monitor/Projector with ATSC/NTSC source standards. Your TV will not add or detract from the signal and will display it as it arrived. You will see, more or less, exactly what movie producers and TV directors intended a program or movie to look like..

Video calibration exists for the same reason we have standards in architecture: doorknob heights, widths of stairways, spacing between wall electrical outlets, etc.

-ChrisCoaster

Audio Empire
June 15th 12, 06:54 AM
On Wed, 13 Jun 2012 12:29:37 -0700, wrote
(in article >):

> I'm an audio-video NUT. I've had several jobs as an AV technician or sound
> man and know a few things about the subject.
>
> When it comes to MP3s most people simply don't understand exactly what kind
> of compression is going on. This is why I feel "compression" is a bad word
> to describe the data parsing that goes on in MP3 files. There is little to
> no DYNAMIC compression via MP3 files. There is DATA compression and
> psycho-acoustic masking going on, but the only dynamic compression I've heard

> is on a MP3 ripped from a digital "remaster" file or CD.

Yes, This is confusing to many people. How do you explain to a layperson that
when FM signals are compressed, the difference between the loudest and the
softest sounds being broadcast is reduced, but when the same station is
"compressed" for Internet radio, the digital data that represents the radio
station's sound is reduced, throwing away up to 97% of the original signal?
>
> The only area dynamic range *might* be affected by mp3 is on lower
> bitrate(128kbps or lower) files. A lot of high-freq and some low-freq
> information is thrown away at or below that threshold, so you may reduce the
> dynamics of instruments occupying or reaching into those realms. On 192kbps
> and up, I honestly have not heard any such loss.

Dynamic compression of the program material in MP3 files may or may not be
happening. For instance, some radio stations that also broadcast over the
Internet, might take off the program feed either before the stations signal
processing chain (compression and limiting), or after it. If before it, then
the dynamic range is not compromised before digital compression, if after the
radio station's signal processing, then the dynamic range IS compromised.
Also most popular recordings these days are purposely heavily compressed in
order to sound loud and exciting (but they end up just sounding BAD). Many
over-the-air FM stations, that broadcast popular music tend to highly
compress the dynamic range of their broadcast audio in order to sound louder
than their competition in that broadcast market in order to grab the
attention of "dial-twirlers".