Talk:Duplex (telecommunications)

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing / Networking (Rated Start-class, High-importance)
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.
Taskforce icon
This article is supported by Networking task force (marked as High-importance).

Completely omitted terms, and unclarity[edit]

The term "diplex" has been completely ommited from this and related articles.
1) In a "simplex" system, only one transmitter can send on the channel at a time.
2) In a "diplex" system, two transmitters can send on the channel at a time.
3) In a "multiplex" system, multiple (three or more) transmitters can send on the channel at a time.

The abbreviation "TDD" cannot be used for "time division duplexing", because in the United States, "TDD" already means "Telecommunications Devices for the Deaf". I dislike it when people try to give duplicate meanings to acronyms. For example, "USA" always means the "United States of America", and there is no confusion, and likewise, "U.K." always means the "United Kingdom". (talk) 03:37, 1 October 2008 (UTC)

Don't argue about the terms[edit]

It doesn't make any sense to discuss about which term came first, or whether "two times half-duplex equals full-duplex". These terms are historical ones that came up as the need arose, and not through some sort of a rational system. They just didn't - and we have the historical terms and continue to use them. (talk) 03:44, 1 October 2008 (UTC)

Simplex vs. Half-Duplex[edit]

I know that the article follows tha ANSI standard. Is there any documentation of how that came to be the standard? There are a couple of confusing points:

  1. Using the road analogy, everyone agrees that (full) duplex corresponds to a two-lane road, with traffic flowing in either direction, each in its own lane. Half of that would seem to be one lane travelling only in one direction.
  2. Similarly, two of what ANSI calls half-duplex gets you full-duplex plus a lot of overhead for managing the direction of traffic in either direction in either lane. When does two times a half equal more than a whole? In duplex communication, I guess. Maybe we could call the result double-half-duplex and confuse everyone?
  3. Before duplex communication was invented, people used one line to talk to each other (say on a telegraph). It seems odd that such a system would be called half-duplex, when presumably that was just the way it is prior to duplex.
  4. Nobody that did not have a duplex telegraph in mind is likely to have invented a one-way circuit, so does that mean simplex requires duplex, but half-duplex came along independently and first?

If it is possible to document why the decision was made to make the current naming the standard, it might help inquiring minds make sense of the situation. Dpv 15:15, 8 June 2006 (UTC)

Hm, all my education in electronics has always taught me the ANSI spec.
To me, "Dup-" as in "Duplex" means two. Probably why "(full) Duplex" means two-way communication. "Simp-" as in "Simplex" means simple, as in lowest working system, i.e. a one-way broadcast like TV, AM radio stations, etc. Certainly adding traffic management systems and controlling the flow over a single channel is not "simple".
"Half-Duplex" still contains the "dup-" part, so to me it means its still two-way, albeit over one channel with switching.
I don’t agree that two half-duplex channels give you full-duplex. As the entire connection (both channels) can be switched to the same direction, then to me this is simply a wider bandwidth half-duplex channel. Assuming each channel had a 50 Mbit bandwidth, you then have a 100 Mbit half-duplex system, that can transfer 100 Mbit at a time, either up or down or a combination of the two with the proviso that one can't exceed 50 Mbit while the other is happening.
Whether it is done by allocating the available channels to different directions, and switching these as needed or allocating the same channel to different directions via a TDMA protocol, then its still a half-duplex system to me. Bonding multiple half-duplex systems just gives you a finer granularity than an 'up/down' system
Calling a one-way system like TV/radio "half-duplex" to me doesn't sound right. "Uniplex" maybe?? :/
A clear term for one-way already exists and is in use: "simplex". There is no need to invent another.
-- 07:28, 17 June 2006 (UTC)

I know that most people are trained according to the current standards, and once trained they tend to think what they were taught is logical. My question is how the standard came to be the way it is now. It seems odd that two half-duplex don't make a full-duplex. It seems odd that the way telegraphs worked before anyone had duplex is now called half-duplex. But I know standards don't make sense to everyone all the time. (Does anyone really use their favorite term for one-way only communication, whether half-duplex, or simplex, for broadcast radio? I would normally only think of it for something like a telegraph wire with a Morse-code key on only one end and a buzzer only on the other end. "Broadcast" is a good enough term for the radio example.) I really want to know if we can explain the current terminology with documentation of how it came to be the standard. Dpv 11:48, 17 June 2006 (UTC)

"These terms also apply to early PC sound cards, however almost all are now full-duplex." - is it really still possible to buy a half-duplex PC sound card?

Echo cancellation[edit]

Shouldn't echo cancellation (as used by V32, etc) be mentioned here, or am I misunderstanding the page? (I can add it, I just want to be sure this page isn't intended to be specific to, say, radio technologies or something.) Squiggleslash 23:13, 2 January 2007 (UTC)

Ethernet example[edit]

Full-duplex ethernet schemes are more varied than the example given in this article. Below are some methods used to achieve full-duplex operation:

10BASE-T: two simplex twisted pairs, one in each direction.
100BASE-TX: two simplex twisted pairs, one in each direction.
100BASE-T4 (rare): four simplex twisted pairs, one in each direction and two subject to negotiation.
100BASE-T2 (rare): two duplex twisted pairs with echo cancellation.
1000BASE-T: four duplex twisted pairs with echo cancellation.
1000BASE-TX (rare): two simplex twisted pairs, one in each direction.

That's just for ethernet over twisted pairs. Some of the optical standards use wavelength-division multiplexing for full-duplex over a single fiber. There's weird older stuff like 10BROAD-36, too. —Ryan (talk) 20:02, 18 October 2009 (UTC)

Semi-Duplex Missing?[edit]

Why is the popular Semi-Duplex missing in this article? — Preceding unsigned comment added by (talk) 02:41, 8 August 2013 (UTC)

Silly half-duplex analogy[edit]

From the article: "A good analogy for a half-duplex system would be a one-lane road with traffic controllers at each end, such as a two-lane bridge under re-construction. Traffic can flow in both directions, but only one direction at a time, regulated by the traffic controllers."

The boldface part works fine as an analogy, no problems there. The bold-italic part doesn't track with the nature of half-duplex systems, because there is no external 'regulator' akin to traffic controllers when using half-duplex systems (at least in the few obvious examples that come to mind) - when using walkie talkies, the users themselves must manage the exchange of transmissions, analogous to road traffic (for example by ending their transmission with 'over') and likewise on the one-land road cars would just wave to each other to indicate intentions (I can't speak for the everywhere, but in England, Wales, Scotland and France (much of rural, hilly parts of Europe) I can say there are definitively one-lane, two-way roads, without traffic co-ordinators or traffic lights, and drivers must negotiate these roads themselves - they are usually short, quiet roads, and in the event two cars meet in opposite directions, the car closest to the end of the road simply acknowledges the other car and backs up).

Upon reaching the italic part of the analogy it just becomes absurd; I am sure readers with even the most basic level of education can envisage a one-lane road with two-way traffic without having to bring into the analogy a bridge and some re-construction work - even if you live in a region without such roads and have never travelled anywhere with one. To reduce it to the absurd, the bridge inclusion adds as much to the analogy as would the following "...such as a two-lane 17th century cable-stayed bridge under re-construction due to corrosion caused by decreasing pH levels in local precipitation as a result of climate change from government-backed deforestation for the purpose of clearing land for increased soybean production in order to meet the growing global economic demand." - it's entirely superfluous.

I would like to edit it, so that it reads "A good analogy for a half-duplex system would be a one-lane road that allows two way traffic, traffic can only flow in one direction at a time." or if absolutely necessary, "A good analogy for a half-duplex system would be a one-lane road with traffic-lights allowing two way traffic, traffic can only flow in one direction at a time - when the light is green at one end it is red at the other."

The reason I exposited this here instead of just making this trivial edit is that I am sure it would either be flagged as vandalism or somebody would complain without understanding my reasoning, because I happen to choose to edit anonymously from an often changing IP (having no desire to have every edit I ever make collated and searchable). I'm also rather bored so I thought I might as well. So I will check back in a few days and if nobody has complained I will make the edit then. If anybody reverts it after that, I will put the change back, and direct them here to explain why they reverted it and to wait until consensus is reached before further edits are made, as a full explanation and opportunity to object was given prior to the edit and in good time. (talk) 00:46, 24 September 2015 (UTC)