BOC-L Digest / Still in the Shadows

[Eric Siegerman]

(I'm not an expert on this stuff either; I just pretend to be one
on the Net :-)

On Tue, Jun 01, 2004 at 09:51:54AM -0400, Paul Mather wrote:
> On Tue, Jun 01, 2004 at 11:27:31AM +0200, Henderson Keith wrote:
> => *does this give a sharper picture, or eliminate 'resonance' patterns,
> => if the frequency matches?  I would *guess* this must be the case...
>
> Not that I'm an expert, but I believe it does eliminate resonance
> patterns (rolling hum bars) and gets rid of flicker problems with TV
> cameras in the studio.

As well, by taking the vertical-sync timing from the line
frequency rather than generating it, they could save the extra
components that a clock circuit would have required.  Don't
forget how old these standards are:  the NTSC frequency and
number-of-scan-line parameters were set in March 1941 (gotta like
the Net!); I assume roughly the same time frame for PAL.  In
those days, circuitry was a *lot* more expensive than it is now
-- think what the first computers looked like in the mid-40s, and
extrapolate back a few years.  So I don't know, but it's possible
that that stupid 50- or 60-Hz clock circuit represented
significant cost savings in those days.

That said, I've seen what 50/60 interference looks like.  It
ain't pretty.  A company I once worked for took our (Canadian, so
NTSC-based) computer-music system (McLeyvier, in case anyone's
interested in lunatic-fringe synth arcana --
http://www.synthmuseum.com/mcleyvier/mcleyvier01.html) to the
Musik Messe (huge, monster, truly astounding music trade show) in
Frankfurt.  The McLeyviers ran fine through a step-down
transformer, but the screens were basically unreadable.  As I
recall, it looked rather like trying to see a reflection in a
still pond that someone's just heaved a rock into, but the waves
were linear, not circular, and rolled slowly up the screen.  (If
you brought a similarly crappy PAL monitor over here, would the
waves roll the other way? :-)

Lars, the hardware guy who was along for the ride, pulled the
power supplies out of the monitors' bases, spliced in extension
cables between them and the monitors themselves, and put the
supplies down on the floor.  The idea was to put some distance
between the A/C circuitry of the power supplies (whose output was
D/C) and the monitors' video circuitry.  It worked; though there
was still a small amount of ripple, that extra meter of
separation was enough to make the displays more than readable
enough to demonstrate.

(Ob-HW-Reference: I took personal time after that trip (I wasn't
about to waste free airfare to Europe!), and used it to go to
England in search of Hawkwind.  I was stupendously lucky -- they
were touring just when I was there.  So that's how I got to my
first two HW gigs.  Winter '84 tour; Slough & Manchester.)

> => One question I have though...why don't they have *all* the features of
> => these chips activated for every model?  My guess is that it would be
> => harder for them to sell more expensive models, if the cheaper ones had
> => all the same features.  Which means the companies are basically making
> => their products "inferior" intentionally, so that you will pay more for
> => something that isn't really any more advanced, but just has been 'turned'
> => on.  OK, I'm a professional cynic when it comes to mega-corporations,
> => but what other answer is there?

> The reason the cheaper models have all the functionality built in (but
> selectively enabled) is because it's cheaper for, say, a Far-East DVD
> player manufacturer to design and build a single unit they can
> mass-produce out the wazoo and sell in all the world markets (Europe,
> North America, Asia, etc.) than to have several specialised units that
> can only be sold in a particular geographic area.  It's a matter of
> economies of scale.

Yup; standard procedure.  Making two million of one chip costs a
lot less than a million each of two similar, but not quite
identical, chips -- specifically, the very significant design,
testing, debugging, and manufacturing-setup costs of the second
chip.  That doesn't make it any less infuriating to us end-users,
though :-(

> The other main reason would be to cut down on pilot error.  If it were
> easy to, say, switch to PAL-encoded output, then there'd be lots more
> people messing up their units and losing picture on their NTSC TVs by
> playing around with the settings.  I believe that's why a lot of the
> features are accessible only via these hidden "engineer" codes.
> They're ostensibly "for experts only."

Also, perhaps this again saves them from hardware differences.
(Here I'm completely guessing.  It sounds plausible enough, but I
really don't know much about the ins and outs of
consumer-electronics manufacturing, so this paragraph is pure
supposition.)  To lock down the region code, you'd have to add
custom circuitry like a ROM or PAL (uh, that's a different "PAL",
nothing to do with the video standard) that end users can't
easily mess with.  But for differences like PAL/NTSC that aren't
backed by legal requirements, the "engineer code" might let them
manufacture absolutely identical units, then customize them for
different markets after they've come off the assembly line.

--

|  | /\
|-_|/  >   Eric Siegerman, Toronto, Ont.        erics at telepres.com
|  |  /
It must be said that they would have sounded better if the singer
wouldn't throw his fellow band members to the ground and toss the
drum kit around during songs.
        - Patrick Lenneau