SJ-MD100 (serv.man2) - Panasonic Audio Service Manual Supplement (repair manual). Page 7

– 7 –

Y

“ATRAC” (adaptive transform acoustic coding) is the name given to the signal compression technology that is one
of the most important technologies used in the MD system.  The principles of this technology are extremely complex,
so here we will only explain the basic concept.

Y

Simply speaking, utilizing the characteristics of the human auditory sense, the sounds that cannot be heard by the
human ear are eliminated so that only the sounds that can be heard remain, thus reducing the amount of data that
must be recorded.  This principle is similar to the PASC signal compression that is used for digital compact discs.

Y

The amount of data that can be recorded per second (the bit rate) is 16 bits x 44,100 x 2 channels = 1,411,200 bits
per second, or approximately 1.4 megabits per second (1 megabit = 1,000 kilobits = 1 million bits).

Y

However, because the diameter of an MD is only about one-half that of a CD, it only has a capacity of about 160
megabytes.  In order to record the same 74 minutes of data as a CD, the bit rate of 1.4 megabits per second must
be compressed to one-fifth.  “ATRAC” is the signal compression technology that was developed in order to do that.
(Signal compression is rather like the technology for orange juice concentrate, in which the best part of the fresh
orange is concentrated for easy transportation and then reconstituted with water before drinking.)

Y

The basic concept of ATRAC utilizes the characteristics of the human auditory sense.  The frequency spectrum
that cannot be heard by the human ear (Fig. B) and the frequency spectrum that cannot be heard because it is
masked by high-level sounds (Fig. C) are cut, and the bit rate is compressed by appropriately arranging the bits,
thus making it possible to fit all of the necessary data within the capacity of the disc.

Y

Technically speaking, the analog signals are converted to 44.1-kHz 16-bit digital signals, and these signals are then

processed by the ATRAC encoder.  Using a maximum time of 11.6 ms as a single unit, the encoder converts the
digital waveforms within each single unit into about 500 different frequency spectra and then analyzes the strength
level of each frequency.  Then, as shown in the figures below, utilizing the two principles of the “threshold of
audibility” and the “masking effect”, only the signals actually heard by the human ear are selected and compressed
to one-fifth the original volume.

Y

This complex process is performed by an LSI chip called the ATRAC.

D

 ATRAC signal compression technology

U

What does the word ATRAC
mean?

U

Why is signal compression
necessary?

U

Basic concept of ATRAC

Fig.A

Fig.C

Fig.B