Talk:Analogue filter

Latest comment: 2 years ago by 30103db in topic "Analogue" in the context of electronics
Good articleAnalogue filter has been listed as one of the Engineering and technology good articles under the good article criteria. If you can improve it further, please do so. If it no longer meets these criteria, you can reassess it.
Article milestones
DateProcessResult
July 31, 2009Good article nomineeListed
Did You Know
A fact from this article appeared on Wikipedia's Main Page in the "Did you know?" column on July 11, 2009.
The text of the entry was: Did you know ... that in designing a new analogue filter, Sidney Darlington found tables of the exact elliptic functions required in an 1829 Latin paper by Carl Jacobi in the New York City Library?

Renamed to analogue

edit

Moved from Analog to Analogue (as in analogous filter) - Rehnn83 19th June 06

I don't think the adjective "analogous" has been used in an engineering context. 84.227.254.143 (talk) 12:22, 30 March 2014 (UTC)Reply

Wrong name/description

edit

An analog filter need not be described by a differential equation. In fact, an analog signal need not be a continuous-time signal. All that is required is that the values ("range") of the signal come from a continuum, just like quantities in the physical world. Discrete-time signals are still analog signals. They only become "digital" signals after they are quantized (i.e., "coded"). This page content (or its name) should be updated as to not mislead people to think "analog" has anything to do with continuous time. In fact, perhaps it should be removed entirely. —TedPavlic (talk/contrib/@) 00:03, 4 July 2009 (UTC)Reply

Are you seriously suggesting deleting such a substantial article merely because you don't like the title? That's almost trollish. SpinningSpark 17:52, 4 July 2009 (UTC)Reply
Nearly every section has a {{main article}}, and much of the other content can probably be merged into (or may already be a part of) Electronic filter. Additionally, the title is misleading/easily misunderstood. Furthermore, much of the content on this page would have a better chance of getting upkeep if it were on more visible (and already existing) pages. —TedPavlic (talk/contrib/@) 12:26, 6 July 2009 (UTC)Reply
That is a stunning series of invalid reasons.
  • much of the content on this page would have a better chance of getting upkeep if it were on more visible (and already existing) pages.
Do you mean it would be on more peoples' watchlist? That is not an acceptable reason for removing an article, articles exist on the merits of their content only. Besides which you actually have no idea how many watchlists this is on, that information is kept secret, deliberately so to prevent vandals obtaining lists of unwatched pages. As for your comment about already existing pages, if you had troubled to check the history of this page you would find that it has existed since 23rd January 2003, one of our more elderly articles. On the other hand I note that you have proposed merging into Electronic filter which has only existed since 10th May 2005. By the way, it is rather poor etiquette to propose a merge at another page and not place a link to the discussion at the page proposed for merging.
First of all that is obviously not true, even at a cursory glance at the article: in particular realisability, equivalence and approximation, all important concepts in filter design are not covered in any sort of detail anywhere else on Wikipedia as far as I know. Nor is Darlington's famous and much used insertion loss method, it does not get so much as a mention in other articles. Secondly, even if it were so that there were a main article for every section, there would still be a place on Wikipedia for an overview of the whole subject.
The thrust of this objection seems to be the implication that the article consists of pasted bits of other articles. Even if true, it does not invalidate the justification for an overview article, however, it is not even close to being true. This article has been written from scratch using new sources, in some cases specially obtained for the sole purpose of writing this article. In the case of the main two main articles on network synthesis and image filters, there may be a common turn of phrase, but that is only because I wrote those articles also. Much of the material was chosen for this article for the very reason that it did not appear in the previous articles.
  • much of the other content can probably be merged into (or may already be a part of) Electronic filter
The claim that material here may already be part of electronic filter is a claim that could only be made by someone who has not read both articles. It categorically is not. If you merge this article in whole, you will then have an article that is so long that it will be suggested to break it up again. Besides which, this article covers acoustic and mechanical filters which is really not appropriate material for electronic filter.
  • An analog filter need not be described by a differential equation
A correct statement which the article does not dispute. The article merely claims that linear filters are described by linear differential equations. Even were the article in error, correction of the error is the proper course, not suggesting deletion.
The only point I accept (although I certainly do not accept it as a reason for deletion) is that the article does not fully cover all analogue filters. The limitation in the scope is clearly stated at the beginning of the article: passive linear analogue electronic filters from an historical perspective (although the "electronic" bit should now be dropped since the article has strayed into other areas). I limited the scope merely because the article had been under construction in userspace for a long time and, frankly, that was the limit of what I wanted to write about. There is no restriction on others adding to the article to extend it. It could be moved to a new title of course, but it would remain the only "analogue filter" article in existence and analogue filter would become a redirect to the new article. That makes it a pretty pointless exercise in my book, there is no point doing anything like that until there is more material on Wikipedia making it worhtwile breaking into different articles.
SpinningSpark 14:53, 12 July 2009 (UTC)Reply
Responding to "This page content (or its name) should be updated as to not mislead people to think "analog" has anything to do with continuous time. In fact, perhaps it should be removed entirely." I have no idea which readers would be so misled. There's no need to change the name of the article. There's certainly no need to remove it (!?!) entirely. Binksternet (talk) 17:28, 12 July 2009 (UTC)Reply

No citation templates in this article?

edit

Is there anyone besides User:Spinningspark who condemns the use of citation templates in this article? --bender235 (talk) 12:25, 26 July 2009 (UTC)Reply

I am going to support User:Spinningspark in this specific case (see GA review): usage of templates is not obligatory up to GA level, it indeed scares novice WP editors, but the major argument is historical - old sources of this article can't be checked by bots, thus no need for templates. Materialscientist (talk) 22:37, 26 July 2009 (UTC)Reply

Those templates aren't just there for bots, but also for human readers, who sure find it helpful to have a consistent citation style within Wikipedia. Also, I don't see how it "scares" new editors, since no one demands them to add citations only with those templates. That can be done later by other Wikipedia users. --bender235 (talk) 22:42, 26 July 2009 (UTC)Reply
Don't get me wrong, I myself use templates everywhere, but please understand the others' arguments: (i) consistent formatting has little to do with templates; (ii) novice editors often do not understand how to input templates and mistype their fields, messing the citations, whereas they are fine without templates; (iii) mixing up templates and no-templates does result in inconsistent formatting. Nothing is clear-cut here, off course you can emulate any template by text. Materialscientist (talk) 22:48, 26 July 2009 (UTC)Reply
But emulating a citation template by text destroys the whole idea of having flexible references that can be modified in style in the future (if necessary). Also, those citation templates will implement a citation microformat once it is available, which enables bots and search engines to index those references much easier (just like the hCard microformat we implemented to all those biography infoboxes a while ago). --bender235 (talk) 07:21, 27 July 2009 (UTC)Reply
Those are benefits of templates, and they are true, but there are drawbacks as well (just to add one, a typo is usually invisible in templates, but is clear in no-template formats). The point is "current policies" do not require use of templates. If you glance here you might realize how unshaped are the citation standards yet. Materialscientist (talk) 07:56, 27 July 2009 (UTC)Reply

Four word title? Why not make it five? Or go back to two.

edit

This page was recently renamed to make it four words long. It went from "analogue filter" to "passive analogue filter development." Going back to my old comment and its responses:

  • If it's a good idea to make the title this precise, then why not add "continuous-time" to that? Analog does not imply continuous time. There are plenty of discrete-time analog PWM devices, for example. Common applications of 555 times generate analog-"coded" PWM signals. Likewise, sample and hold devices can be used to do analog discrete-time linear filtering (e.g., sum weighted versions of samples of analog signals). So if you want the TITLE of this page to really match its content, you should either add a discrete-time section or add "continuous-time" to the title.
  • It was argued that most industry people say "analog filter" when they are actually picturing a "passive continuous-time analog filter." Hence, even though it's incorrect to be so general, it's so commonplace that it's better to be recognizable than be correct. By this logic, the page should be un-renamed.

TedPavlic (talk/contrib/@) 13:07, 27 July 2009 (UTC)Reply

Forgive me, I know nothing about the previous discussion. Just outsider's comments: (i) The title should be understood by non-specialist. Most people can guess what digital/analogue means, but "continuous-time" is jargon (time is always continuous) (ii) Words "passive" and "development" were necessary to specify the article coverage, but the title should be as concise as possible. Materialscientist (talk) 22:46, 27 July 2009 (UTC)Reply
I'm saying that most people can't guess correctly at what digital and analogue mean. In particular, most lay people assume that digital implies discrete time (and discrete value) and that analogue implies continuous time (and continuous value). Likewise, this page implies that all "passive analogue filter development" goes on in continuous-time domains when that is certainly not the case. Moreover, this page reflects the misconception that the two major schools of filter development are analogue and digital. In fact, they are continuous time and discrete time (and now, with the emergence of hybrid systems showing up in control and signal processing, hybrid time). Additional details about whether a signal is analogue or digital (i.e., whether the value space is constrained) have more to do with the noise characteristics of the filter than the actual filter design. —TedPavlic (talk/contrib/@) 13:10, 28 July 2009 (UTC)Reply
This, with proper referencing, might be worth adding into some article as a note, but it is too heavy for the titles. Materialscientist (talk) 22:24, 28 July 2009 (UTC)Reply
Regardless, I think it's strange to have a title that cannot be bolded within the first sentence on the page. —TedPavlic (talk/contrib/@) 12:47, 29 July 2009 (UTC)Reply
I noticed that too. Is it necessary to have the title bolded in the lead? Materialscientist (talk) 12:50, 29 July 2009 (UTC)Reply
Yes and no. From WP:BOLDTITLE:

As a general rule, the first (and only the first) appearance of the article's subject should be as early as possible in the first sentence and should be in boldface.

The Solar System consists of the Sun and those celestial objects bound to it by gravity.

The article's subject is usually the same as the page title, but not always. In lists (including outlines, indexes, and glossaries), the subject is generally preceded by the article type (such as "List of"). The article type should not be presented as the subject of the article, only the part after it should. For example, in Outline of Africa, the first sentence of the lead should describe Africa, and present it in bold – not Outline of Africa (the article is not about outlines of Africa, it is presenting an article on Africa in outline form).
If the title of a page is descriptive it does not need to appear verbatim in the main text, and even if it does it should not be in boldface. So, for example, Electrical characteristics of dynamic loudspeakers begins with:

A dynamic loudspeaker driver's chief electrical characteristic is its electrical impedance versus frequency.

Additionally, from WP:MOSBOLD:

If the article topic does not have a commonly accepted name, but is merely descriptive (e.g., history of the United States), the title does not need to appear in the first sentence, and is not bolded if it does.

So, if you interpret "subject" literally, then maybe "analog filter" should be bolded. However, with the loudspeaker example, maybe nothing should be bolded. —TedPavlic (talk/contrib/@) 13:41, 29 July 2009 (UTC)Reply
Ted, you give 555 timer applications as an example of a discrete-time application, but this application is excluded by "passive". Do you have any examples of passive discrete-time applications? I do not think such a thing is strictly possible, so there is no point making the title more specific if you are still encompassing the same set of circuits. Anyway, I was personally quite happy with the original 2-word title. SpinningSpark 18:38, 29 July 2009 (UTC)Reply
I will concede that so long as the article's topic is explicitly passive AND focuses on physical implementations, it's safe to assume that signals discussed will all have continuous-time domains. My issue was when the title was "analog filter." An "analog filter" arguably describes an extremely broad class of filters encompassing anything that outputs a signal whose values come from a continuously ranged values. That being said, as Wikipedia is for the lay person, and the lay person (in fact, the lay industry expert) will probably look for "analog filter" when "continuous-time filter" is actually meant. So, just as it's now acceptable to pronounce "often" with a hard "t" in the middle of it because lots of people do it, it's probably acceptable to have an "analog filter" title for this subject matter just because that's what lots of people will look for. However, if the title stays as it is (PAFD), it should probably not be bolded (certainly not when it's paragraphs away from the lead). —TedPavlic (talk/contrib/@) 12:13, 30 July 2009 (UTC)Reply

don't say passive

edit

i think when people say 'analog filter' now they very likely mean an active one with lots of op-amps and thrifty on the 'passive' parts (capacitors and inductors) — Preceding unsigned comment added by 72.193.24.148 (talk) 00:52, 22 February 2012 (UTC)Reply

Article is excellent

edit

This article is a cut above the usual Wikipedia article. It concentrates on historical development, which distinguishes it from most of the electrical engineering articles. The writing is well-paced, erudite. 84.227.254.143 (talk) 12:22, 30 March 2014 (UTC)Reply

"Analogue" in the context of electronics

edit

This article and Analogue electronics is pretty much the only place it's spelled this way. Even within the article, the infobox, the books/articles in the references and bibliography and the category box spells it "analog". Are there actually electronics books/publications which spell it this way or is this a case of pushing a british variant in a context where it doesn't belong (ie, "computer programme")? 30103db (talk) 19:21, 24 March 2022 (UTC)Reply

Such a foolish comment could only be made by soneone who has not done a search of the literature beforehand. SpinningSpark 14:00, 25 March 2022 (UTC)Reply
So snarky... Searching through IEEE publications, JSTOR, arXiv, google scholar, google ngram, or even simply looking at the number of google search results shows not only that "analog" is clearly the preferred spelling (even within the sources on this article), but that sources which do spell it "analogue" tend to be older (over 30 years ago) and predominantly british. I'm not going to bother changing anything since this appears to be a charged topic, but if this really is a case of WP:ENGVAR, the MOS:COMMONALITY and MOS:ARTCON (ie, the title of the infobox at the top of the page is currently "Linear analog electronic filters") guidelines should apply. 30103db (talk) 14:29, 30 March 2022 (UTC)Reply