On 16 Dec 2011, at 12:14 PM, Chuck Guzis wrote:
On 16 Dec 2011 at 11:19, Christian Kennedy wrote:
Then there's the political angle; if RCA
hadn't sued to stop
deployment of the CBS system we might have all grown up with spinning
color wheels :P
Why? Do you think that development of the tri-color CRT would have
been prevented by adoption of the field-sequential color system?
I don't think that development of the *CRT* would have been prevented but the
deployment of a *television system* that used them to advantage probably would have been.
The CBS system was incompatible with the existing NTSC system, reduced line count from 525
to 405 and while it increased the field rate to 144 it had an effective rate of only 24
frames per second. In some respects RCA's lawsuit and the National Production
Authority order were clearly designed to make the CBS system non-viable thanks to the
growing installed base of B&W NTSC receivers, but that doesn't mean that the much
more complex NTSC system wasn't in the end technically superior to the CBS field
sequential system. Seriously, the Columbia 12CC2 receiver was an impressive kludge, but a
kludge nevertheless, and one that pointed up a problem with the color wheel system with
respect to larger screen sizes (although the subsequent drum receivers somewhat addressed
that problem).
Note that I'm not arguing that field-sequential is inherently inferior, only that the
NTSC solution was a highly serviceable one that preserved the installed base. Many modern
DLP projectors use color wheels running at some multiple of the frame rate.
Then there was the whole colordaptor thing: NTSC color signals converted to
field-sequential in order to drive a color wheel placed in front of an existing NTSC
B&W television. These things were both homebrewed and available from some television
manufacturers, but the consumer saw them as a kludge.
Instead, we had to put up with terrible color for
decades.
In part that's due to the difference between the original NTSC (1953) colorimetry and
the response curves of the phosphors that manufacturers ended up using. The phosphors
originally specified had better color (probably better than modern CRTs) but were
inefficient and slow (to the point that images smeared); receiver manufacturers responded
by changing the phosphors in their tubes and then doing parlor tricks with the signals to
try to map from the 1953 colorimetry to the response curves of their chosen phosphors.
Given that the transmitted signals are gamma corrected it's hard (or perhaps
impossible) to accurately perform this mapping in analog space so the results were always
a manufacturer-specific approximation of the broadcast colors. Somewhere in 68 or 69 RCA
and Conrac proposed a new set of control phosphors for use in broadcast video monitors
that became the SMPTE C spec, in 1987 (this dragged on for a while) it was adopted into
SMPTE 170M. Of course both PAL and SECAM used the original NTSC colorimetry as well, at
least until 1970, but rather than allowing color correction in the receiver the EBU
directly encoded signals for EBU colorimetry values. That's what NTSC should have
done from the outset, but going back and taking the covers off the spec to add that was a
non-starter.
The other source of crappy color in early NTSC was the fact that the color burst phase
would tend to drift when you changed the channel, hence the dreaded "tint"
control (both NTSC and PAL sets had hue controls, but in the case of PAL sets it was
usually buried in the set and only touched at the factory or during repair). The problem
largely resolved itself when vacuum devices were replaced by solid state devices.
I
remember the CBS color system--some CBS-branded sets of the time even
had octal (or dodecar) sockets at the back of the chassis for the
colorwheel adapter.
True CBS-compatible sets also had a "color/monochrome" switch that changed the
behavior of the vertical oscillator and horizontal sync circuits as well as the gook to
sync the color wheel to the vertical sync pulse and to brake the wheel and then position
it to "clear" segments when viewing B&W programming.
Compared to the green skin tones of the RCA system, the CBS color
rendition was superb. Europe had the right idea--instead of rushing
something half-baked to market.
CBS had great color because it didn't use phosphors. PAL and SECAM have great color
because they wedged correction factors into the signal. NTSC had crappy colors because
they assumed that they knew what the phosphors would look like but they were wrong; the
analog would be the CBS system with each manufacturer using differing hues on the color
wheel. How fabulous would that color rendition have been?
Granted, I saw only demonstrations of the CBS color system, but I
remember that the local electronics surplus dealer had big wooden
crates full of now-useless color wheels. I'm sorry that I didn't
pick up a few.
They weren't useless, since it was possible to obtain (or build) aftermarket adapters
to allow NTSC color to be rendered on a NTSC B&W set using a color wheel -- the
consumer just didn't want them.
--
Dr. Christian Kennedy
chris at
mainecoon.com AF6AP
http://www.mainecoon.com PGP KeyID 108DAB97
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
"Mr. McKittrick, after careful consideration..."