NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/rec/video/16236

Internet Message Format | 1993-01-22 | 5.4 KB

Path: sparky!uunet!cs.utexas.edu!sdd.hp.com!crash!cmkrnl!jeh
From: [e-mail]
Newsgroups: rec.video
Subject: Re: RGB Input on Mitsubishi TV
Message-ID: <[e-mail]>
Date: 22 Jan 93 00:43:12 PST
References: <[e-mail]>
Distribution: world
Organization: Kernel Mode Systems, San Diego, CA
Lines: 94

In article <[e-mail]>,
 [e-mail] writes:
> I have an 8-pin RGB input on my 26" Mitsubishi TV (CS-2669R).
> Can this be used to input a Hi-8 source via an S-video cable,

No.

> or is an S-video output compatible only with an S-video input?

Essentially, yes.

> Are such cables available? Also, are there any other potential
> uses for connecting a camcorder or VCR directly to the RGB input?

Only if you have RGB outputs on the camcorder or VCR.

Here's why:

The two signals in the "S-Video" connector consist of the "Y", or luminance
signal, which is the black and white part of the NTSC signal plus the sync
pulses, and the "C", or chroma signal, which is the 3.58 MHz carrier with the
color information encoded onto it (via a complex mechanism which we needn't get
into here).

The "NTSC composite" video signal -- the one that's present at the yellow RCA
"phono" jack -- looks like a black and white video signal (just like the "Y"
signal in the S-video connector) with the 3.58 MHz chroma signal added to it.
When you feed video to a monitor via this jack, the first thing it does is to
pick that 3.58 MHz carrier out of the rest of the mess.  The monitor then is
dealing with separate "Y" and "C" signals, just as if you'd gone in through the
S-Video jack.

Now, the monitor decodes the chroma signal to get Red and Blue levels, and
feeds these to a matrix circuit that adds these two together and subtracts
the sum from the "Y" or luminance signal, the result being the Green level,
since Y = R + G + B.

(Actually Y = .3 R + .59 G + .11 B, if I remember right.  These proportions are
based on the eye's sensitivity to light.  I mention this not because it
contributes to the understanding of composite video vs. S-video vs. RGB
signals, but because if I don't, some wiseass who would never dream of
answering a question directly, but who just loves to pick technical nits in
other people's answers, is bound to jump up and correct me.  So, to all
wiseasses out there, if I have these proportions wrong, please correct me, and
I'll save the information where I can find it next time.  For the rest of us,
what matters is that once you have Y and R and B, you can derive G.)

(Someone is bound to wonder:  If NTSC video can encode three signals -- overall
brightness (Y), Red level, and Blue level, why bother with G = Y - R - B?  Why
not just send G in place of Y, and be done with it?  The answer is:  For
compatibility with b&w tv sets.  These pay no attention to the 3.58 MHz chroma
carrier; the Y part of the signal *is* a b&w video signal, so the b&w set is
happy.  If G were sent instead of Y, the picture on the b&w set would show only
the amount of green that's in the scene, rather than the total brightness
level, which includes R and B.)

Where was I?  Oh yeah... The results are three signals representing R, G, and B
brightness levels. These are used to control the emissions of the three
electron guns in the CRT.

The signals you would feed to the RGB inputs *almost* look like the RGB signals
present in the set after decoding Y and C.  I say "almost" because to accept R,
G, and B from external inputs the monitor also needs horizontal and vertical
sync information.  The de facto standard in the consumer tv monitor industry is
that RGB inputs expect "sync on green", meaning that the sync signals are added
to the green signal. The result looks just like a regular monochrome video
signal, except that the video information reflects only the amount of green in
the original scene, not the overall brightness (luminance) level.

In other words you could take the G signal that you would normally feed to the
G input of an RGB monitor, and feed it instead to the video input of a black
and white monitor, and you would see a picture -- a black and white picture
representing the amount of green in the original color picture.  But try that
with R or B and you'll just get hash on the screen. That's because there are no
sync pulses in the R or B signals to tell the monitor where a scan line or a
vertical field is supposed to start.

As you probably know, the standard in PC monitors is different -- the
horizontal and vertical sync signals are on separate wires.  It's easiest that
way, since the computer's video card produces the sync signals separate from
the R, G, and B signals, and since extra conductors in a six foot monitor cable
are cheap, there is no point in combining the sync with the green on the video
card only to have the monitor pick them apart again.  (Some computer monitors,
DEC's at least, still use sync on green.)

You mentioned that your set has an 8-pin "RGB" input.  The extra pins might
allow for separate sync signals.  Check the manual.  But even with separate
sync inputs, the usefulness of RGB inputs on tv sets is pretty limited.  unless
you are running an IBM PC with a CGA card, or an early Commodore Amiga, the
computer needs a monitor with much faster sweep rates than a tv monitor will
do.

    --- Jamie Hanrahan, Kernel Mode Systems, San Diego CA
Internet:  [e-mail], or [e-mail]  Uucp: uunet!cmkrnl!jeh