HamCall (October 1991)/HamCall (Whitehall Publishing)(1991).bin/bcast/bcstcble/ccode.txt

Text File | 1990-10-14 | 6KB | 108 lines

EVALUATION OF A COPY PREVENTION METHOD FOR DIGITAL AUDIO TAPE SYSTEMS

National Engineering Laboratory
National Bureau of Standards
February 1988

Executive Summary:

Digital audio tape (DAT) systems have all the attractive audio qualities now
associated with compact disc technology plus the ability to make very high
quality copies of copies.  Concern over loss of intellectual property rights
has produced proposals for legislation requiring that DAT systems sold in
the United States be fitted with a copy prevention decoder such that
suitably coded material could not be copied by the DAT recorder.  The CBS
Records system, the basis for the proposals, removes a narrow band of
frequencies from the audio signal - termed putting a "notch" in the
frequency spectrum of the signal.

In considering this proposal, several committees of the Congress asked the
National Bureau of Standards (NBS) three questions about the copy prevention
system:

     1.  Does the copy prevention system achieve its purpose to prevent
digital audio taping machines from recording?

     2.  Does the system diminish the quality of the prerecorded material
  into which the notch is inserted?

     3.  Can the system be bypassed, and, if so, how easily?

The National Bureau of Standards received from CBS REcords descriptive
material, specifications, circuit diagrams, and operating encoding and
recording/decoding devices for the tests.  The results and conclusions of
the NBS tests apply to these devices.  Suggestions for recorded material to
be used in listening tests were received from several sources.

NBS conduced a series of laboratory studies to understand the copy
prevention system and to exercise it in several ways to obtain answers to
question 1 above.  The NBS also constructed and evaluated several circuits
designed to bypass the system (question 3).  With the aid of an expert in
acoustical evaluations on contract from a university, listening studies were
conducted using subjects drawn from the local section of the Audio
Engineering Society, NBS staff and a few local audiophiles and musicians.
These listening studies were carried out to answer question 2 above.

The NBS findings are as follows:

     1.  Answer to question 1 - "does the system achieve its purpose?":

     NBS Conclusion:  The system does not achieve its stated purpose.

          The systems does prevent the copying of notched material much of
          the time.  However, for about half of the recorded tracks studies,
          the system exhibited FALSE NEGATIVES; i.e., notched material was
          nonetheless recorded.  In addition, the system also exhibited
          FALSE POSITIVE behavior, i.e., the system failed to record
          unnotched material.  NBS studies 502 tracks on 54 compact discs
          and found false positives for 16 tracks on 10 discs.

     2.  Answer to question 2 - "Does the system diminish the quality of
the prerecorded material?":

     NBS Conclusion:  The system's encoder alters the original electrical
     signa.  For some listeners for some selections, this results in a
     discernible difference between prerecorded notch and unnotched
     material.

          NBS interpreted this question both in terms of objective
          electrical measurements and in terms of whether or not listeners
          can detect differences between notch and unnotched material in
          carefully controlled listening tests. k Electrical measurements
          showed that the encoded signal is degraded relative to the signal
          as it was prior to the copy prevention coding, degraded in terms
          of adding a deep notch in signal amplitudes near 3840 Hz, and
          degraded in terms of changes in signal phase relationships near
          3840 Hz.  In a double-blind series of listening tests, 87 subjects
          listened to a prerecorded tape of short selections from 24
          different sources - 20 compact discs and four selections produced
          on a keyboard synthesizer.  The results show that, although the
          effects of the encoder are fairly subtle for some musical
          selections, there are some selections for which the subjects
          detected differences between notched and unnotched material.

          In a second series, 15 experienced listeners worked with 10
          selections presented on parallel tape tracks such that the subject
          could switch back and forth from notched to unnotched material.
          For 2 of the 10 selections the encoded version was correctly
          identified 12 out of 15 times; these results are statistically
          significant.

     3.   Answer to question 3 - "Can the system be bypassed, and, if so,
          how easily?:

     NBS Conclusion:  The copy prevention system can be bypassed easily.

          This question was interpreted by NBS as asking whether or not a
          DAT recorder can be made to record notched prerecorded material.
          It is not possible to restore all of the actual information
          removed by the encoding process.  NBS engineers designed and
          constructed several electronic circuits for implementing five
          different methods to circumvent or defeat the copy prevention
          system by the use of external signal conditioning.  All five
          methods succeeded in bypassing the copy prevention system.  The
          circuits are simple and easy to construct.  The cost would be in
          the order of $100.