5

Over the years and with the large variety of hardware I've dealt with I've used quite a few of different video inputs and outputs. While I usually have no issue just plugging cables into devices, I really started to wonder how video cables differed from each other.

Could anyone specify the difference between s-video, rgb, rca, component, composite, vga, hdmi, displayport, dvi, and any others I'm forgetting? Why do these cables keep evolving; what does an hdmi cable offer that an s-video cable does not, for example? What are the advantages of using one cable/input over another? Refresh rate? Input lag?

closed as off-topic by JAL, Aaron, fadden, Stephen Kitt, Chenmunka Aug 15 '16 at 8:52

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question does not appear to be about retrocomputing, within the scope defined in the help center." – JAL, Aaron, fadden, Stephen Kitt, Chenmunka
If this question can be reworded to fit the rules in the help center, please edit the question.

  • 5
    This question might be more fitted to Electrical Engineering. Some of the video inputs you note are not at all retro. – wizzwizz4 Aug 10 '16 at 8:58
  • @wizzwizz4 If I don't get an answer here or it is not well received I will ask it over on EE – wcarhart Aug 10 '16 at 9:00
  • 3
    Simply put, Bandwidth. You can't pump 4K video through a composite input and expect tolerable results. – R Drast Aug 10 '16 at 11:44
  • This has been flagged as off-topic, but I'm pretty sure it's on topic somewhere... If there are any other suggestions as to where it should go, they would be very helpful. – wizzwizz4 Aug 10 '16 at 13:47
  • 3
    What is your question specifically [even if it's to ask elsewhere]? Folks are not always receptive to a non-specific request like, "just tell me lots of stuff about {x}..." Without knowing what type of problem you are trying to solve specificially, folks have no way of knowing if their efforts are actually helpful... or if someone is simply hoping to trawl through whatever people type later. – Robert Cartaino Aug 10 '16 at 14:33
10

To stay on-topic for the RetroComputing SE, I'll only answer for the video connectors commonly seen on retro systems:

The first thing to keep in mind is that you cannot always rely on the connector shape alone to know what it is carrying. The best example of this is the venerable RCA connector, which is used for all kinds of different signals. Sometimes (but not always) you can tell what an RCA connector (or set of connectors) is for based on color coding.

Systems that were designed with the intention to be connected to your television often had a "TV out" or "RF out". This was usually done via an RCA connector (the connector used for composite and component video, and also commonly used for audio) which would attach to your TV via a switch box. The switch box usually had either a 300-ohm antenna output (2 screw terminals) or a 75-ohm antenna output (F-connector, commmonly used for cable TV). You could tell that a system had "TV out" or "RF out" because it often had a channel selector switch that let you change what TV channel was used (2/3 or 3/4 were common options in the USA). An RF TV signal was basically a composite video signal transmitted on a radio-frequency carrier wave, in a very similar fashion to how audio is transmitted by radio stations. the RF signal was able to carry both video and audio (which allows TV programs to have sound, albeit mono because at that time only 1 audio channel was used).

The composite video signal carried by RF was a standard signal, following the NTSC standard in the USA, the PAL standard in much of Europe, etc. The NTSC standard defined the refresh rate and it also defined the number of visible lines (483, split into 2 frames due to interlacing). Video chips at that time often generated composite video signals directly, then used an "RF modulator" to convert the signal to RF, which the TV would then convert back to composite again. the RF conversion step would introduce some distortion, so you could get better picture quality by skipping it and sending the composite signal directly, but this required you to have a "monitor" with a composite input because TV's at the time only had RF inputs. about the time VCRs were getting popular, it became common for TV manufacturers to include composite inputs in their TVs, so you no longer needed a special monitor. Composite video commonly used the RCA connector, but to avoid confusion a convention was adopted to color the video connector yellow and the audio connector white (eventually white and red when stereo audio became common in TVs).

A composite video signal was called "composite" because it was made up of 2 signals, a 'brightness' signal (luminance) and a 'color' signal (chrominance). On a black-and-white TV or a monochrome monitor, only the brightness signal was relevant (this allowed the same composite signal to work on both color and black-and-white sets). Just like you could eliminate distortion due to RF modulation by skipping the RF step and sending composite video directly, you could avoid distortion due to the 'composite' conversion by sending luminance and chrominance signals on 2 separate wires. This is exactly what S-Video did (and also what Commodore did on the C64 and the 1702 monitor). Commodore used RCA connectors on the 1702 monitor, while the S-Video standard defined a new connector which probably has a technical name but we just call it the S-Video connector.

The chrominance signal in S-Video is itself made of 2 sub-signals that are combined into 1, and you can get a slight improvement in color quality by sending them on separate wires instead of combining them. That's what component video does, but it goes back to using RCA connectors for each signal instead of inventing something new (like S-Video did).

Each of these standards was related to the previous one and a common theme in all of them was to make 'luminance' or 'brightness' a separate signal to support monochrome / black-and-white monitors. but color TVs don't actually use luminance and the 2 chrominance sub-signals directly, they use RGB (red/green/blue) to display colors. So if you're generating RGB from your video chip you can avoid the entire process of converting to and from component by just sending RGB directly (there's another 2 signals called horizontal sync and vertical sync which are used to control screen refresh, so RGB will often have 5 connectors, or 4 if H-Sync and V-Sync are combined on 1 wire, or 3 if they are both combined with the green signal). VGA is a common example of RGB (analog), but EGA and CGA qualify as RGB signals as well (albeit digital RGB). It is also common to see RGB using BNC connectors (these are similar to the F connector used by cable TV, but only require a quarter-turn to secure instead of threading a nut onto a bolt like the F-connector does). Modern TVs that have component inputs will sometimes allow you to reconfigure them as RGB inputs via a setup screen, which means that you will sometimes see RGB using RCA connectors also.

  • Just to be more correct: F-connector was not used in Europe for TV - older IEC 61169-2 connector was (and is) used for that purpose, and you could often find it in European computers with the (PAL) RF output. F-connector was/is used for satellite TV. – Radovan Garabík Aug 4 '17 at 14:53
  • The S-Video connector is a Mini-DIN connector. – tofro Aug 25 '18 at 15:41
6

Here are the practical differences in the context of retrocomputing:

At the lowest end is RF. The picture is pretty noisy and requires a TV with antenna screw terminals or coax, or a modulator to convert the signal to composite + audio. Audio and video is sent as a single signal.

The next step up is composite which provides a much clearer picture by passing only video through the 1 signal channel. Audio goes through separate cable(s). Unfortunately, the picture quality improvement makes some color fringing more obvious.

Next is S-Video which eliminates that color fringing by passing luma and chroma through 2 separate signal channels. (The more signal channels, the better quality.)

Next is component (3 signals), SCART (1-4 signals + audio), or VGA (usually 5 signals1). These connectors (especially component and VGA) all support progressive scan to eliminate flicker (depending on the source signal), and component and VGA support higher resolutions.

The remaining connectors you listed are digital2, which eliminates any remaining noise, jitter, and other analog artifacts. HDMI and DisplayPort both carry audio, while DVI does not.

Notes:

1The Playstation 2's optional VGA adapter uses only 3 signals in the VGA cable.

2Depending on the connector, DVI cables can also carry analog signals.

  • 1
    SCART isn't really a signal standard like the rest you described. Rather its a universal connector standard designed to be able to carry any of (the older) signals on one cable to simplify hooking up devices and to allow for easier switching between inputs. That's why you'll see things referred to as, for example, RGB-SCART, indicating that the signal passing through the SCART cable is an RGB one. – mnem Aug 10 '16 at 20:06

Not the answer you're looking for? Browse other questions tagged or ask your own question.