Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

A question of S-video

11 views
Skip to first unread message

Brian

unread,
Nov 28, 2002, 12:50:58 AM11/28/02
to
Correct me if I'm wrong.
A direct video input from a DVD player to a TV gives a display of 250
lines.picture
A S-Video input from a DVD to a TV gives a display of over 400 lines
of picture.
To get a sharper picture you would need to change from a direct video
input to an S-Video input?

Is this correct?

Regards Brian

Action Jackson

unread,
Nov 28, 2002, 1:34:35 AM11/28/02
to
You are confusing S-Video with SVHS.


"Brian" <bcl...@es.co.nz> wrote in message
news:tkbbuug4en5okuo91...@4ax.com...

jdcarswell

unread,
Nov 28, 2002, 1:36:32 AM11/28/02
to
Yep, you've got it, although the number of lines of resolution are
theoretical and there are other factors that play a role in determining
percieved clarity and resolution. From best to worse, here's how common
analog signals stack up:

Component video (also sometimes called RGB)
S-Video
Composite (this is what you call direct video)
Coax (75ohm modulated video+audio.Avoid this one completely if possible)

"Brian" <bcl...@es.co.nz> wrote in message
news:tkbbuug4en5okuo91...@4ax.com...

Nigel Mellor

unread,
Nov 28, 2002, 8:55:18 AM11/28/02
to
Correct me if I'm wrong, but here in Europe, component video and RGB are
distinct.

Component video has separate R,G,B signals but the synch is carried on
Green.
RGB scart has seperate R,G,B lines but also uses the Composite signal to the
provide the synch.

Most European domestic AV equipment uses RGB SCART, whereas in Japan and
Australasia component video is more common.

--
Nigel Mellor
nigel....@ronime.com

"jdcarswell" <johnhyphe...@shaw.ca> wrote in message
news:QPiF9.64853$ea.11...@news2.calgary.shaw.ca...

Tim Mitchell

unread,
Nov 28, 2002, 11:32:53 AM11/28/02
to
In article <as5787$hmd$1...@newsreaderm1.core.theplanet.net>, Nigel Mellor
<nigel....@reply-to.in.signature> writes

>Correct me if I'm wrong, but here in Europe, component video and RGB are
>distinct.
>
>Component video has separate R,G,B signals but the synch is carried on
>Green.
>RGB scart has seperate R,G,B lines but also uses the Composite signal to the
>provide the synch.
>
>Most European domestic AV equipment uses RGB SCART, whereas in Japan and
>Australasia component video is more common.
>
You're kind of wrong ... component video is used in professional
equipment and uses colour difference signals not RGB. There is RGB with
sync-on-green but it's not normally called component
--
Tim Mitchell

Nigel Mellor

unread,
Nov 28, 2002, 12:22:47 PM11/28/02
to
"Tim Mitchell" <T...@sabretechnology.co.uk> wrote in message
news:2UYEsTf1...@tega.co.uk...

> You're kind of wrong ... component video is used in professional
> equipment and uses colour difference signals not RGB. There is RGB with
> sync-on-green but it's not normally called component

My apologies.

Just out of interest, what are the color component deltas referenced to,
luminance perhaps?


David McCall

unread,
Nov 28, 2002, 12:46:46 PM11/28/02
to

"Nigel Mellor" <nigel....@reply-to.in.signature> wrote in message news:as5jd7$nb8$1...@newsreaderm1.core.theplanet.net...
Yes.

When S-video first came out, it was often referred to as component,
because it is made up of the "components" of composite vide.
The only differences are that the modulated color subcarrier and
luminance are carried on separate wires, and that this allows the
luminance bandwidth to be higher (allowing for better detail).
Unfortunately the color signal is still being beat to death by the
subcarrier, so it is only a minor improvement when it comes to
color :-( Due to the popularity of component video as used in
Betacam and other common formats, this term is no longer
used when talking about S-video.

While RGB is also technically component, it is rarely referred to
by this name, for the same reason that it isn't used for S-video.
RGB is generally referred to as, RGB. Each component of RGB
has the same bandwidth, so would require a very robust system
to record it with any decent quality.

Component, today, primarily refers to an Y, R-Y, B-Y signal.
The Y part is exactly like the Y used in S video, but the
color information is carried as 2 bandwidth limited color
difference signals, with no subcarriers to mess it up. This
makes it be an improvement over S-video.

David


Steve Simpson

unread,
Nov 28, 2002, 1:35:39 PM11/28/02
to
> To get a sharper picture you would need to change from a direct video
> input to an S-Video input?
>

Exactly the same information is on both outputs.

S-video carries the Chroma (color) information separately so that it does not
have to be processed into and back out of the (composite) signal.

Steve Simpson

unread,
Nov 28, 2002, 1:44:25 PM11/28/02
to
> Correct me if I'm wrong, but here in Europe, component video and RGB are
> distinct.
>
> Component video has separate R,G,B signals but the synch is carried on
> Green.

RGB is video formed from mixing those three colors. Sync can be on green, mixed
on a separate single channel or completely separated on two channels (horiz &
vert).

Component video is basically a Black and White picture (luminance) and separate
blue and red chroma.

In 'home theater' speak, component is sometimes referred to as 'RGB' which is
where the confusion come from.

S-video is similar except that the colors are not separated.

Composite is all signals mixed on one line.

David McCall

unread,
Nov 28, 2002, 3:06:19 PM11/28/02
to

"Steve Simpson" <simps...@NIXcox.net> wrote in message news:dutF9.18730$kz4.1...@news2.west.cox.net...

> > Correct me if I'm wrong, but here in Europe, component video and RGB are
> > distinct.
> >
> > Component video has separate R,G,B signals but the synch is carried on
> > Green.
>
> RGB is video formed from mixing those three colors. Sync can be on green, mixed
> on a separate single channel or completely separated on two channels (horiz &
> vert).
>
Not exactly. RGB is 3 separate color signals that can remain discrete
through out the whole process. Once you mix it, it isn't RGB any more.
RGB directly from the camera is by far the best source for chroma keying,
but there is no way to record it. For many years, we did all chromakey
work live. Composite and S were never any good for that, and even
component isn't all that wonderful.

Sync can be on the green signal, all 3 signals, or on a separate sync
line, or separate H and V sync lines, as you mentioned.

> Component video is basically a Black and White picture (luminance) and separate
> blue and red chroma.
>

The blue and red, are color difference signals ( R-Y and B-Y, Y is luminance)

> In 'home theater' speak, component is sometimes referred to as 'RGB' which is
> where the confusion come from.
>

I would correct any salesman that tried to tell me that. It is a totally
incorrect statement. Calling RGB component has limited merit though.

> S-video is similar except that the colors are not separated.
>

Worse than that. The color information is encoded into a phase
differential signal (similar to wiggeling the hue control, and a
color level signal, similar to wiggeling the color level up and
down real fast, and then those 2 signals get mushed together
by amplitude modulating the color level onto a subcarrier
(like AM radio), and phase modulating the hue information
onto that same subcarrier (like FM radio). What a mess,
but it works, sort of. The luminance is carried on a separate
wire, and, in that way, very similar to the component "Y" signal.

> Composite is all signals mixed on one line.
>

Composite is the same as S-video, except that the Y signal is
mixed with the subcarrier mentioned above. the Y signal has to
be bandwidth limited (reduced detail) before mixing the signals
so as to reduce the luminance detail's effect on the color (an
example would be the rainbow effects on a pinstriped shirt or tie) .

David


Steve Simpson

unread,
Nov 28, 2002, 8:39:07 PM11/28/02
to
> > RGB is video formed from mixing those three colors. Sync can be on green,
mixed
> > on a separate single channel or completely separated on two channels (horiz
&
> > vert).
> >
> Not exactly. RGB is 3 separate color signals that can remain discrete
> through out the whole process. Once you mix it, it isn't RGB any more.

That depends on whether you consider the display device to be part of the 'whole
process'.

The screen image ('video') is composed of the three 'colors' mixed, yes?

I was trying to make the point that there is no 'Y' in RGB . . . . :-)

Of course, as you correctly pointed out, the channels are kept separated in
transmission.

Brian

unread,
Nov 28, 2002, 9:07:09 PM11/28/02
to
Thanks Steve and others for your replies.

Does that mean I might notice a small difference if I changed to a
S-Video connection when connecting a DVD player to a Television.
Both have S-Video connections.

On what type of connection is their a limited video bandwidth?
I always thought that the reason for S-Video was to overcome a limited
video bandwidth.

Regards Brian

JD

unread,
Nov 28, 2002, 10:15:01 PM11/28/02
to

"Brian" <bcl...@es.co.nz> wrote in message news:hniduu8dhme76v8sm...@4ax.com...

> Thanks Steve and others for your replies.
>
> Does that mean I might notice a small difference if I changed to a
> S-Video connection when connecting a DVD player to a Television.
> Both have S-Video connections.
>
> On what type of connection is their a limited video bandwidth?
> I always thought that the reason for S-Video was to overcome a limited
> video bandwidth.
>
If you use S-Video, you are limiting the color bandwidth to about
1.5MHz at best, and true S-Video is even more limited. If you
use component, you can get over 2.5MHz of color bandwidth.

(True S-Video can be directly combined to create true NTSC,
S-Video need not adhere to all of NTSCs limitations, but some
chroma decoders will make errors on wideband S-Video.)

The luma bandwidth of S-Video can theoretically be as great as
component.

The differences in color aren't always obvious, and it is dependent
upon source material quality, and the quality of your display.

If the original source (to make the DVD) is DV or NTSC (like from
D2 masters), then the advantage of component would nearly be
insignificant. If the original source was carefully mastered with
all 4:2:2 digital equipment, you'd likely be able to discern a
difference. It all depends upon your viewing situation, and the
quality of your display.

S-Video can look subjectively very good. It isn't perfect, and
unless you have a wonderful set-up (and view very well mastered
DVDs), then the benefit of component can be measurable, and
can sometimes be obvious, but you have to decide upon how much
you want to spend. Good quality cables aren't expensive.

The only reason for using composite is for convienience or a lack
of component/S-Video inputs. The loss of picture quality can
be significant (lost diagonal resolution along with other comb
artifacts.)

John

RJ

unread,
Nov 28, 2002, 11:25:41 PM11/28/02
to
Brian...

You should notice a big difference in something like "dot crawl" on high value
color edges, like red.

Of course, if your TV has decent built in Y/C separation, the difference you see
switching to S-Video may no be that great. But I think you will see a difference.

RJ
...

David McCall

unread,
Nov 29, 2002, 12:12:08 AM11/29/02
to

"Steve Simpson" <simps...@NIXcox.net> wrote in message news:%yzF9.19396$kz4.1...@news2.west.cox.net...

> > > RGB is video formed from mixing those three colors. Sync can be on green,
> mixed
> > > on a separate single channel or completely separated on two channels (horiz
> &
> > > vert).
> > >
> > Not exactly. RGB is 3 separate color signals that can remain discrete
> > through out the whole process. Once you mix it, it isn't RGB any more.
>
> That depends on whether you consider the display device to be part of the 'whole
> process'.
>
> The screen image ('video') is composed of the three 'colors' mixed, yes?
>
Get out your magnifying glass. You will see completely separate
Red, Green, and Blue phosphers. They don't actually mix until they
hit your eyes, if you are talking about a computer monitor or an
RGB monitor in a studio (not that comon these days). TVs and
other video monitors do recieve encoded signals and then they
separate ity into an RGB signal that is delivered to the actual
display surface.

RGB isn't used as much in video anymore. everything went
component, and now digital, but most studio cameras used to
have RGB outputs suitable for chroma keying.

> I was trying to make the point that there is no 'Y' in RGB . . . . :-)
>

True. Y can be derived by summing the 3 signals together though.


Nigel Mellor

unread,
Nov 29, 2002, 4:34:01 AM11/29/02
to
"David McCall" <david...@attbi.com> wrote in message
news:%GuF9.167469$QZ.26783@sccrnsc02...

> The blue and red, are color difference signals ( R-Y and B-Y, Y is
luminance)

Presumably the green component is "reconstructed" or derived by some
function Yb,Yr and Y?


GMAN

unread,
Nov 29, 2002, 3:33:48 PM11/29/02
to

Please dont confuse "Component" with "Composite" . Composite is very poor when
compared to S-Video connections.

COMPONENT = best
S-Video = better
Composite = fair

Steve Simpson

unread,
Nov 29, 2002, 9:24:57 PM11/29/02
to
> > The screen image ('video') is composed of the three 'colors' mixed, yes?
> >
> Get out your magnifying glass. You will see completely separate
> Red, Green, and Blue phosphers.

That describes every color monitor on the planet regardless of the method of
transmission, so I don't see the specific relevance.

Besides, I think we are getting far OT of the OP's question, which was simply
what is the difference between composite and s-video.

He was under the incorrect assumption that there was a big difference on
resolution, which could mean he was simply confusing composite and s-video with
vhs and s-vhs . . . as someone else pointed out.


Stradling, Ed

unread,
Nov 30, 2002, 4:15:38 AM11/30/02
to
S-video is only the second best way of watching the DVDS, the best is
through a scart wire if both your machines have scart sockets which are
RGB-capable


"Brian" <bcl...@es.co.nz> wrote in message
news:hniduu8dhme76v8sm...@4ax.com...

Richard Crowley

unread,
Nov 30, 2002, 9:29:20 AM11/30/02
to
"Stradling, Ed" wrote ...

> S-video is only the second best way of watching the DVDS, the best is
> through a scart wire if both your machines have scart sockets which are
> RGB-capable

You understand, of course, that SCART is a European phenomenon. Most people
in America likely never heard of it and don't have a clue what it is.
Equivalent over here would be one of the "component" formats like Y/Pr/Pb,
RGB, etc.


Dave Haynie

unread,
Nov 30, 2002, 5:43:35 PM11/30/02
to
On Thu, 28 Nov 2002 06:36:32 GMT, "jdcarswell"
<johnhyphe...@shaw.ca> wrote:

>Yep, you've got it, although the number of lines of resolution are
>theoretical and there are other factors that play a role in determining
>percieved clarity and resolution. From best to worse, here's how common
>analog signals stack up:

>Component video (also sometimes called RGB)

Actually, for US digital TV, this isn't RGB at all, but YPrPb. That's
a luminance plus two color encoding channels. Some older component
systems, as well as some SCART video systems in Europe, use RGB.
They're carrying the same amount of information, but the YPrPb
encoding is more favorable to video images (a better color gamut).

>S-Video

This is also called Y-C video; or going way back, "split luma-chroma"
video. The Commodore 64 home computer used this type of connection,
though long before the consumer electronics industry dubbed it
"S-Video" and gave it a one-piece mini-DIN connector.

>Composite (this is what you call direct video)

This shouldn't be called "direct" video, it should be called
composite, or perhaps CVBS (Composite Video Broadcast Standard).

>Coax (75ohm modulated video+audio.Avoid this one completely if possible)

Yup. When you have that, you're taking your CVBS signal, some form of
audio, and modulating it over RF, to make it appear as a broadcast
signal. The only reason to use this is for playback on television with
no video inputs. Ok, it's also useful for long runs: a composite,
split, or component signal is probably good for 10m or thereabouts,
not much more (they each have specs). Once you modulate it, you can
run the RF signal over 100m or more of good cable, at most requiring a
booster amp (available at any Radio Shack).


Dave Haynie | Chief Toady, Frog Pond Media Consulting
dha...@jersey.net| "The weather is here, I wish you were beautiful" - J.Buffett

0 new messages