[net.audio] 2 questions, 2 replies

goodmon@encore.UUCP (Brian Goodmon) (02/27/85)

I've been wondering about a few things:

What speed does the tape run on the machine that is recording
the information for a digital master tape?  The reason that I ask is
that I was wondering if you could put that same info onto a cassette
tape, and have a D2A/A2D converter at home.  This would allow you to
reap the benifits of digital reproduction without a CD player, and would
have the added advantage that you could record digitally yourself.  However,
some quick figuring seems to suggest that this isn't feasible:  16 bit
samples at 44.1 K per second means roughly 800 K bits per second;  guessing
that a cassette tape runs at around 3 inches per second means that the
information density on the tape would have to be 266 K bits per inch.
That seems impossibly high, since I seem to remember that computer mag tapes
have a density on the order of 10 K bits per inch or less.  That makes me
wonder how fast a digital master tape runs, hence this question.  Have I
missed something in my calculating?  Also, I guess that another reason that
this isn't a great idea is that the cassette deck would probably need extra
functions, like logic to search for beginnings of "tracks", etc.  Oh well.

Are there advantages to a linear tracking tonearm other than the obvious
removal (you hope) of tracking error?  An article that I copied from Stereo
Review states that there is no need for anti-skating compensation.  Is this
true?  Can you explain why?  Their explanation was not clear.  Do they really
mean "since many linear tonearms are driven by servo motors, the servo motors
compensate for skating force so that no *additional* compensation is
necessary"?  What about linear tonearms that don't use servo motors (the
Souther, for instance)?  Also, what is the audible manifestation of tracking
error?

Now, some responses to questions by other readers:

To the person who asked for headphone suggestions:  listen to Stax.  Their
electrostatic "earspeakers" are wonderful.  They are also expensive.  Oh
well.  Even if you can't afford $400, listen just so that you can compare
other headphones to the best (the best that I've heard anyway).  A cheaper
brand that has very good sound is Sennheiser (sp?).  They don't have nearly
the resolution of the Stax, but come in a respectable second.  Unfortunately,
the only model that I found comfortable to wear (ie. fits around the ear
instead of pressing against it) is their top-of-the-line.  That translates
to "most expensive", but it is still cheaper than Stax's bottom-of-the-line.
Stax also makes electret headphones that beat the Sennheiser sound for only
slightly more money (around $150 I think).  This model (Lambda Junior it's
called) might be a good compromise cost/sound-wise.  Electret headphones
work in much the same way as electrostatics...the difference is that the
thing that moves to create the sound is a heavier, permanently magnetized
piece of metal rather than a thin, light, electrically polarized piece of
saran wrap.

To the person who asked about Harmon-Kardon T-60 turntables:  the T-60 is
a nice machine, with some nice features.  Before you decide though, give
a listen to an AR with a Linn Basik tonearm (this happens to be what I
ended up buying).  The T-60 is more convenient to use, what with the
auto shut-off/tonearm lift, and the front panel 33/45 rpm selector (on
the AR you must lift the platter off and change the belt to another pulley),
but the AR/Linn sounded better to me.  The bass is, to my ear, tighter and
more realistic.  The difference is due to the Linn being a better tonearm than
the one that comes with the H-K (ie, stiffer arm, and better, tighter
bearings).  The tonearm that AR makes is not held in high regard...most places
don't even sell it, usually mounting the Basik on the AR.  Another good
separate tonearm is the Grace 707; depending on the compliance of your
cartridge, it might work as well or better than the Linn.  The decision between
the H-K and AR/Linn is a tradeoff between slightly better sound and much greater
convenience.  Both the H-K and AR/Linn go for about $450.  Another table that
you might consider is the Kyocera.  It sounds at least as good as the H-K, is
even more convenient (more handy semi-automatic features), and, if I remember
right, costs less (around $375 I think).  For a medium cost manual or semi-auto
turntable, you won't go wrong with any of these three.

ron@brl-tgr.ARPA (Ron Natalie <ron>) (03/02/85)

No, you can't use cassettes.  You can however, use videotape.

herbie@watdcsu.UUCP (Herb Chong [DCS]) (03/02/85)

In article <173@encore.UUCP> goodmon@encore.UUCP (Brian Goodmon) writes:
>What speed does the tape run on the machine that is recording
>the information for a digital master tape?  The reason that I ask is
>that I was wondering if you could put that same info onto a cassette
>tape, and have a D2A/A2D converter at home.  This would allow you to
>reap the benifits of digital reproduction without a CD player, and would
>have the added advantage that you could record digitally yourself.  However,
>some quick figuring seems to suggest that this isn't feasible:  16 bit
>samples at 44.1 K per second means roughly 800 K bits per second;  guessing
>that a cassette tape runs at around 3 inches per second means that the
>information density on the tape would have to be 266 K bits per inch.
>That seems impossibly high, since I seem to remember that computer mag tapes
>have a density on the order of 10 K bits per inch or less.  That makes me
>wonder how fast a digital master tape runs, hence this question.  Have I
>missed something in my calculating?  Also, I guess that another reason that
>this isn't a great idea is that the cassette deck would probably need extra
>functions, like logic to search for beginnings of "tracks", etc.  Oh well.
sharp has made a prototype digital recorder that uses a standard TDK MA-R
cassette (they will use no other) at the same recording speed as a regular
analog cassette.  they use something like 20! channels across the width of
the tape and do strange and wonderful things to bias and equalization
to get something back again.  the system is 14bit logarithmic encoding
at the 44.1kHz sampling rate.  if i remember correctly, sharp stated
that this was a technical excercise to show that it could be done and they
had no plans of any kind to build a version suitable for commercial or
home use.  seems to me that dropouts would be a major problem.

>Are there advantages to a linear tracking tonearm other than the obvious
>removal (you hope) of tracking error?  An article that I copied from Stereo
>Review states that there is no need for anti-skating compensation.  Is this
>true?  Can you explain why?  Their explanation was not clear.  Do they really
>mean "since many linear tonearms are driven by servo motors, the servo motors
>compensate for skating force so that no *additional* compensation is
>necessary"?  What about linear tonearms that don't use servo motors (the
>Souther, for instance)?  Also, what is the audible manifestation of tracking
>error?
antiskating is required in conventional tonearms because of the tonearm
geometry and the resultant forces on the stylus.  the resultant forces
on a tangential turntable happen to be tangential and directly in line
with the tonearm.  thus, no antiskating force is required.  draw yourself
a force diagram of a pivoting tonearm and you can see that there is a net
inward force on the groove which must be compensated for unless you can
live with uneven record wear.

>To the person who asked for headphone suggestions:  listen to Stax.  Their
>electrostatic "earspeakers" are wonderful.  They are also expensive.  Oh
>well.  Even if you can't afford $400, listen just so that you can compare
>other headphones to the best (the best that I've heard anyway).  A cheaper
>brand that has very good sound is Sennheiser (sp?).  They don't have nearly
>the resolution of the Stax, but come in a respectable second.  Unfortunately,
>the only model that I found comfortable to wear (ie. fits around the ear
>instead of pressing against it) is their top-of-the-line.  That translates
>to "most expensive", but it is still cheaper than Stax's bottom-of-the-line.
>Stax also makes electret headphones that beat the Sennheiser sound for only
>slightly more money (around $150 I think).  This model (Lambda Junior it's
>called) might be a good compromise cost/sound-wise.  Electret headphones
>work in much the same way as electrostatics...the difference is that the
>thing that moves to create the sound is a heavier, permanently magnetized
>piece of metal rather than a thin, light, electrically polarized piece of
>saran wrap.
another one to try are the electret headphones by audio technica.  they are
the next best to stax that i have personally heard, and i have listened
to many.  electret headphones do NOT work as described as above.  electro-
static headphones have a metallized mylar film suspended between two plates
that are charged by the external power supply to some 2 to 3 thousand
volts.  the signal is fed to the film and depending upon the polarity of
the signal, is attracted to one plate or the other because of the strong
electric field.  an electret headphone uses a mylar film which has imbedded
charges so that there is a constant electic field around it.  the plates
are fed the signal which has been boosted to the range of about 2 thousand
volts max by the adaptor.  the charged film moves inside this changing electric
field.  that is why both electrostatic and electret headphones required
adaptors or special amplifiers, but the signal is fed to two different places.

Herb Chong...

I'm user-friendly -- I don't byte, I nybble....

UUCP:  {decvax|utzoo|ihnp4|allegra|clyde}!watmath!water!watdcsu!herbie
CSNET: herbie%watdcsu@waterloo.csnet
ARPA:  herbie%watdcsu%waterloo.csnet@csnet-relay.arpa
NETNORTH, BITNET, EARN: herbie@watdcs, herbie@watdcsu

sjc@angband.UUCP (Steve Correll) (03/03/85)

>  Are there advantages to a linear tracking tonearm other than the
>  obvious removal (you hope) of tracking error?  An article that I copied
>  from Stereo Review states that there is no need for anti-skating
>  compensation.  Is this true?  Can you explain why?...Also, what is the
>  audible manifestation of tracking error?

Look down on your turntable from above, and imagine a line down the
center of your cartridge from front to back. If you have a pivoted
tonearm, you'll notice an angle between that line (I'll call it the
"centerline") and the tonearm: this angle is responsible for skating
force. Friction on the disk surface exerts a force on the stylus, and
because of that angle, the force pulls sideways on the tonearm (or, in
Newtonian terms, the force has two vector components, one of them
parallel to the tonearm and the other perpendicular to it and directed
inward).

Why the angle? Well, the machine which initially cut the groove kept
its centerline tangent to the groove, so one goal of any tonearm is to
keep the cartridge's centerline tangent, too.  The obvious approach
(which nobody uses) is to mount the cartridge with its centerline
parallel to the arm and make the arm just long enough to be tangent
to the outermost groove. A little Euclidian reasoning will show that as
such a tonearm approaches the innermost groove, it must somehow grow
longer to remain tangent.

The popular solution is to make the tonearm longer in the first
place, but mount the cartrige so its centerline is angled inward from
the tonearm, retaining tangency at the outer groove. Given the right
length and angle, such a tonearm will be tangent at the innermost
groove, too, though it will be a little bit wrong in between.
(Manufacturers' hype to the contrary, it doesn't matter whether the
angle is achieved by bending the arm or mounting the cartridge at an
angle at the end of a straight arm.)

Tangential (linear-tracking) arms, whether electrical or mechanical, exhibit
no skating because all the force exerted on them by the record surface is
parallel to the arm.

I have never heard any audible manifestation of tracking error. It rarely
exceeds a degree or so, whereas the vertical angle of the stylus can vary
by 5 or 10 degrees between different brands of cartridge. The minimization
of tracking error is popular, however, whereas the standardization of
vertical angle is not. People don't seem to worry about the 5 to 8dB resonances
which some highly praised moving-coil cartridges exhibit around 15kHz, either.

I'm (overly) fond of citing an article in "Audio" magazine about a year
ago which concluded that tangential arms actually have about as much
error as pivoted arms, because their servos dither a bit. However,
tangential arms can be much shorter, and therefore less massive, and
therefore (with highly compliant cartridges) more immunity to record
warp.  (Even if warps don't bounce the stylus out of the groove, they
make life harder for the stylus by making the downward tracking force
fluctuate.)

-- 
                                                           --Steve Correll
sjc@s1-b.ARPA, ...!decvax!decwrl!mordor!sjc, or ...!ucbvax!dual!mordor!sjc