From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Photography (Rated C-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Photography, a collaborative effort to improve the coverage of photography 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.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.

For discussions before 2009, see Talk:Pixel/Archive 1


Not to get into the merits of Foveon but that article and several others use the term "photosite" which, far as I see, is nowhere defined in Wikipedia. Perhaps this Pixel article would be the best place to define it and explain its relation to the various meanings of "pixel". Jim.henderson (talk) 03:45, 16 June 2009 (UTC)

You'd need a sourced definition to start with; it's probably just as ambiguous as "pixel" is. Dicklyon (talk) 04:11, 16 June 2009 (UTC)
For example, if it were sourced, you could use the wiktionary definition, but it disagrees, I think, with some of the uses like in Foveon X3 sensor. Dicklyon (talk) 05:12, 16 June 2009 (UTC)


The term "px" appears to be the abbreviation of the word "pixel." Can we confirm or debunk this?

This section created by and I'm just adding attribution. Looking at the user's talk page, it appears that this user has been a serial adder of unsigned comments and warned about vandalism before. DQweny (talk) 22:14, 18 October 2009 (UTC)

The phrase "the smallest item of information in an image"[edit]

I have been checking copy-vio problems in the computer graphics article and happen to notice the phrase "the smallest item of information in an image" used there and here, apparently based on the source

  • Rudolf F. Graf (1999). Modern Dictionary of Electronics. Oxford: Newnes. p. 569. ISBN 0-7506-43315.

Now I was wonder whether or not quotation marks have to be added? Eventually I think not, because this seems to be an exception in Wikipedia:Plagiarism, see here, which states:

Here are some examples where attribution is generally not required:
  • Use of common expressions and idioms, including those that are common in various sub-cultures such as academic ones. In order to qualify as a "common expression or idiom":
    • the phrase must have been used without attribution at least 2 years ago by someone other than the originator and in a reliable source, in other words one that is likely to have watchful editors and lawyers.

The particular phrase is used in at least two other books, see here

  1. Jill Marie Koelling (2004). Digital imaging: a practical approach‎. Rowman Altamira. p.1
  2. Joash Moo (2009). Art in life. Pearson Education South Asia. p.180 ‎

The first source expressed "A bit is the smallest piece of information in an image file", and the second source states: "Pixel : A picture element, which is the smallest piece of information in an image."

I am beginning to get second thoughts here. The second source could be based on Wikipedia, the first source is talking about "bits" and the original source Graf (1999) only mentioned the phrase "The smallest part of information in a binary notation system". I guess this leaves me with the question if Graf (1999) could be considered the source in the first place...!? -- Mdd (talk) 23:31, 18 November 2009 (UTC)

Well, it's an interesting question. The second source you cite is a 2009 source, so you're probably right they got it from wikipedia. And the first is not about pixel per se. GBS will not show me the page with the pixel entry in Graf, so I can't check what he said. Decent definitions for pixel are notoriously difficult to find (at least, I hope I've made this problem notorious via my own paper and talk). Maybe we should look for a couple of alternative definitions to use instead. Dicklyon (talk) 23:41, 18 November 2009 (UTC)
Thanks. I have the same GBS problem. I guess the easiest thing to do here is to find what Graf did say about pixels. I am not sure what paper you refer to. If they are indeed notoriously difficult to find, your last suggestion to search for some alternative one's might not be that easy? -- Mdd (talk) 23:52, 18 November 2009 (UTC)
If you look at the history, e.g. this edit of mine, it appears that there's no reason to suspect that that phrase came from that source. I think you just moved it there, so you should probably move it back. The guy who cited Graf here said it was just for pixel = picture element. And then here an anon changed both the definition and what the source was attached to; I think it would be best take the lead back to something like what it said before this one. Dicklyon (talk) 23:54, 18 November 2009 (UTC)
Ok, this is an interesting turn. I noticed you already made the changes... and removed the phrase. At the moment I think this is a good think to do: Since that phrase isn't based on Graf, my mistake, it could be possible that that phrase is a copy-vio from Koelling (2004). -- Mdd (talk) 00:15, 19 November 2009 (UTC)
P.S. I also like this restored definition better "a pixel... is a single point in a raster image". Just plain and simple. -- Mdd (talk) 00:19, 19 November 2009 (UTC)
I don't see how the phrase "the smallest piece of information in an image" for pixel could have been taken from "A bit is the smallest piece of information in an image file". Sometimes phrases just happen. "smallest piece of information" is in over 600 books. Dicklyon (talk) 00:28, 19 November 2009 (UTC)
Mmm... The reason I started this talk item in the first place was, that the specific expression "the smallest piece of information in an image" seemed like a phrase that could be copyrighted. I read somewhere that an expression can start with just three words. But I don't know all ins and outs... one way or on other. I don't think I started this discussion for the wrong reason, and I appreciate your feed back and solution. I already implemented it in the computer graphics article as well. -- Mdd (talk) 00:44, 19 November 2009 (UTC)
P.S. The phrase was added here without a source. Regarding the anon's other edits, see here this person does seem to be an expert in the field, and knows what he or she is talking about. -- Mdd (talk) 02:46, 19 November 2009 (UTC)
Right, that's one of the diffs I linked. I don't see why you think he's an expert. Dicklyon (talk) 06:42, 19 November 2009 (UTC)
Yes, you are right, sorry. The user did make some advanced additions to the article, and his last remark here. However I often have difficulties assissing situations like this. Do you think he is not? -- Mdd (talk) 12:13, 19 November 2009 (UTC)
It's hard to guess. He seemed to be into the pixel article, and not much else, but not clear why. Dicklyon (talk) 00:40, 20 November 2009 (UTC)
Hi! This is "Anon" from 2008, still posting anonymously to protect my employer, and still an expert in the field. Let me reassure you right now that the text I wrote earlier is in no way a copyright violation, it's simply a statement of fact using the language of signal processing: A pixel is indeed the smallest piece of information in an image.
If I may be so bold as to contrast that with a misleading statement: "In digital imaging, a pixel, or pel, (picture element) is a *single point* in a raster image..." (emphasis mine)
(Us in the industry would call that a single sample of a point-sampled image.)
Mr Lyon, if you're still confused on the difference between a "sample" and a "point", please re-read the (somewhat quaint) first link listed:
Everybody, repeat it with me now, it feels good: "A pixel is *not* a little square." — Preceding unsigned comment added by (talk) 00:39, 28 January 2012 (UTC)
Anon, thanks for responding; just noticed. I'm in the industry, too; see my referenced publication [1]. It does go into the origins and widespread divergent meanings of pixel. And I'm pretty familiar with Alvy's paper, as it was I who linked it. I do realize that "sample" and "point" don't necessarily mean exactly the same thing; but changing point to sample probably also won't go far to making the article more correct or clear. Dicklyon (talk) 19:34, 5 February 2012 (UTC)

First Sentence Confusion[edit]

The topic is pixel, and yet the first sentence starts making odd divisions in its useage by stating that it is used in digital devices (only?) and that "raster" differs in comparison to "display devices". It then speaks of smallest screen element, but then one finds that display devices includes tactile devices for the blind. Very confusing.

First, pixels may be used in analog displays, and second, raster is as opposed to vector. Is that not correct? Thanks. - KitchM (talk) 21:24, 7 September 2011 (UTC)

I also had the feeling that pixels could also be used in the context of analog displays, but apparently the term emerged in use to refer to digital displays. I agree that the first sentence and first few paragraphs need to be clarified. ozhu (talk·contribs) 18:46, 17 July 2013 (UTC)

Display Screen[edit]

I'm not sure how the meaning of "display" in LCD is but I thought it was the noun. Therefor LCD Screen would be wrong, wouldn't it? I'm not a native English speaker but where I come from people tend to say that even stranger, they say "LCD Display". I always thought "LCD" was enough. -- (talk) 08:49, 27 October 2011 (UTC)

LCD can sometimes be enough, but LCD display or LCD screen is not wrong, just more explicit. Dicklyon (talk) 14:58, 27 October 2011 (UTC)

Contradictory - factor of three - are [RGB] one pixel or three pixels ?[edit]

Pixel is probably used in both senses, even throughout the article.

Cheap LCD photo frames tend to cheat on the specs We recommend a minimum resolution of 480 x 234 for 7" digital frames, and 800 x 600 for high resolution 10.4" digital frames That is really "480 x 78" or "160 x 234" not "480 x 234" - note that 800 isn't divisible by 3 ! So we have both meanings in one sentence! (Actually I'm having doubts about this - may have to check physically with some I own !)

Sigma claim other cameras inflate their pixel counts - I doubt it. That's really why I came here seeking clarification !

From the article: (I've used [x3] and [÷3] to show which definition is used ...)

  • [X3] smallest, addressable element
RG+B are individually addressed
  • CRT pixels correspond to their timing mechanisms and sweep rates.
I used to think the [RGB] phosphor dots were pixels, but they can be intersected by 2 raster lines
The image to the right is confusing, despite the caption - can we get a photo showing raster lines within phosphor dots ?
Geometry of color elements of various CRT and LCD displays; phosphor dots in a color CRT display (top row) bear no relation to pixels or subpixels.
I have a 21" Hitachi with more pixels than dots, but would have to index-match the roughened anti-aliasing surface for photo-micrography. Anyone got a Trinitron and a photo-microscope ?
  • In some contexts, the term pixel is used to refer to a single scalar element of a multi-component representation while in others the term may refer to the entire set of such component intensities for a spatial position.
Yes! That sentence showed it is ambiguous...
... but people have stuck enough junk in brackets to hide the meaning ...
  • In some contexts (such as descriptions of camera sensors), the term pixel is used to refer to a single scalar element of a multi-component representation (more precisely called a photosite in the camera sensor context, although the neologism sensel is sometimes used to describe the elements of a digital camera's sensor),[1] while in others the term may refer to the entire set of such component intensities for a spatial position. In color systems that use chroma subsampling, the multi-component concept of a pixel can become difficult to apply, since the intensity measures for the different color components correspond to different spatial areas in a such a representation.
... Clarify, please !
  • A photograph of sub-pixel display elements
[÷3] one pixel = 3 sub-pixels
  • The number of pixels in an image is sometimes called the resolution
[÷3] One wouldn't add red + green + blue resolutions, unless discussing information theory ?
  • "640 by 480 display"
[÷3] except cheap photo frames
  • Subpixel rendering
[÷3] one pixel = 3 sub-pixels
  • Pixels on computer monitors are normally "square"
[÷3] one pixel = 3 sub-pixels
  • Each pixel is made up of triads
[÷3] one pixel = 3 sub-pixels
  • 2 bpp image can have 4 colors
[X3] 4 levels each for R, G and B = 12 colours
(Not to be confused with sRGBWeb_colors#HTML_color_names) !
[÷3] one pixel = 3 sub-pixels
I'm tempted to add "Sometimes known as 'Pixels' " !
This section has been expanded until it loses all sense !
  • It can be an ambiguous term

Who definitively, authoritatively owns the definition ?

  • VESA defined SVGA etc, but charge for their published standards - anyone have a copy handy ?

There may be other industry bodies in different fields ?

I feel that at least the article in Wikipedia needs to be self-consistent - saying the global usage is ambiguous is not enough. Given the use of 'sub-pixel', and the absence of 'super-pixel', we should change the [x3] usages in the bulleted list above. That would be fewer changes, and consistent with HTML and Wiki-code.

Thoughts ?

---19S.137.93.171 (talk) 04:07, 8 May 2012 (UTC)

1 Pixel = 3 Pels ?
3 Pixels = 1 Pel ?
1 Pixel = 1 Pel !
Shame - I had hoped that we could use one word for [RBG] and the other for [R], [B] or [G]
Instead we use both words for both things !
-- (talk) 06:31, 25 May 2012 (UTC)


technical difference between software/hardware[edit]

is there one? is there a difference in nomenclature between software pixels & the pixels on screen? there should be... Lostubes (talk) 19:45, 5 March 2013 (UTC)

Maximum info possible to contain in a pixel[edit]

Quickly reading the article it seems that a pixel is what I have always thought, roughly put: the smallest piece of information in a picture, most commonly rendered as a square but which can also be interpreted so as to blur its edges and overlap it with its neighbouring pixels, as the diagram seems to show. Now, while I have always laughed at the photo-enhancing programs that are the staple of the CSI etc series (e.g the barely visible and miniscule reflection in the sunglasses of somebody in a photo, being enhanced to show in high resolution the face of someone standing behind the camera) due to their ubiquitousness I am now beginning to doubt myself. Surely there can be no more information in a photograph (for instace) than is contained at the level of its pixels. Photo-enhancing is presumably simply the blurring/softening referred to, but can someone confirm how much information can be contained in one pixel? An episode of the sublimely absurd Numbers last night had a pixel containing half a dozen names. Secondly I came across a web-design tutorial regarding single pixel gifs that can be stretched to any size. Is this simply a mix-up in terminologies or can a pixel span many pixels?!? It's not possible from the article to determine the answers to these quite basic questions. LookingGlass (talk) 14:22, 4 April 2013 (UTC)

Yes, the CSI stuff is nonsense. The single-pixel GIF is just a transparent rectangle that you can stretch to cover any number of screen pixels. There's no information in it. Dicklyon (talk) 06:55, 5 April 2013 (UTC)

Thanks. Sometimes the sheer volume of BS makes me doubt my sanity! LookingGlass (talk) 19:07, 5 April 2013 (UTC)

Another, related usage[edit]

Pixel can also refer to a hardware device that consists of an RGB LED, usually in conjunction with a small controller. These devices are often connected together in a string, similar to strings used as Christmas decorations. These pixel strings are often used to create electronic signage, as well as for holiday decorations. (talk) 16:35, 13 April 2013 (UTC)

All points addressable[edit]

I'm trying to better integrate All Points Addressable into the encyclopedia. I don't think that should be a proper noun, all points addressable is what that should be titled, but since it's really never going to be more than a glossary term, it should probably be merged somewhere. Which gets me to the definition in the lead: : "a pixel is... the smallest addressable element in a display device". Shouldn't that say, "a pixel is... the smallest addressable element in an all points addressable display device"? Because certainly in text mode display devices such as the IBM Monochrome Display Adapter a character is the smallest addressable element, not a pixel. Maybe the definition of all points addressable can be moved to a new section of the pixel article, as all pixels addressable or simply pixel-addressable are equivalent terms. – Wbm1058 (talk) 02:29, 23 November 2013 (UTC)

Makyen's lead[edit]

I reverted again Makyen's change of the lead sentence to read:

In digital imaging, a pixel (px), or picture element (pel), ...

The problem is that the cited sources do not support "px", and they support both pixel and pel as abbreviations of "picture element". If we're going to change it, we need to find what sources to use to support the new scheme, and be sure to at least write it consistently with those sources that we cite. In all my research on this topic (see my cited paper), I never encountered discussion of "px" that I can recall. Dicklyon (talk) 16:02, 2 January 2014 (UTC)

There's also the "Mpx" for megapixels. In book search, this seems to be only about 6% as common of Mp for megapixel; essentially an outlier, made up by authors who don't know better, and not really supported by anything authoritative. We should take it out to reduce the confusion. Dicklyon (talk) 16:06, 2 January 2014 (UTC)

Thanks for starting the thread. You beat me to it by a sleep period 8-).
I agree that I, also, have seen "p" used for abbreviating pixel. Sources, certainly for anything for which there is contention, are definitely needed. I also don't have a specific need for mentioning "px" in the lead. However, it is my natural inclination to put abbreviations at the first use of a term.
The primary place that I have seen px used for pixel, in the last several years, has been in webpage development, including the Wikipedia software. The basis for this is the CSS specification. When searching for references, there were a plethora of additional third party sources which discussed this, including ones like Wikipedia:Markup#Images. Given the effectively ubiquitous use of "px" in the web design context, it is non-trivial to separate out sources which used "px" in other contexts (using web based sources for finding cites). W3C is also the only standards body, which I am currently thinking of, that has addressed the issue of an abbreviation of pixel. However, this may just be my more recent experience overshadowing memories from earlier years. If it is a book reference that you desire, I am sure that there are a large number of printed books which would provide such. Reference for W3C definition:
<ref>{{cite web |url= |title=Syntax and basic data types |first= |last= | |year=2011 [last update] |quote=4.3.2 Lengths |accessdate=January 2, 2014}}</ref>
There is a countering argument. The "px" abbreviation has been around through a sufficient development of technology such that it is becoming unclear if it is actually being used only for "pixel", as defined. In fact, the use of "px" as an abbreviation for "pixel", at least in this context, is sufficiently ingrained such that there is third party coverage (a blog, not appropriate for citing) attempting to explain where "px" is not actually intended to mean a device "pixel", but an abstraction due to pixels becoming too small to accommodate the intent behind the hundreds of millions of places "px" is used to mean pixel.
In addition to the desire to add the abbreviation "px", I have an issue with how the first part of the lead is written. I did not re-write it, particularly after the first revert, because I felt that such rewrite would be considered to be more contentious. However, since we are discussing this, we should open up the issue of how that porton of the article is written. Honestly, I would want to spend some more time thinking about how I would consider it best stated. My initial issue is with the first sentence of (current article text):
"In digital imaging, a pixel, or pel,[1] (picture element[2]) is a physical point in a raster image, or the smallest addressable element in an all points addressable display device; so it is the smallest controllable element of a picture represented on the screen. The address of a pixel corresponds to its physical coordinates. LCD pixels are manufactured in a two-dimensional grid, and are often represented using dots or squares, but CRT pixels correspond to their timing mechanisms and sweep rates."
When first reading it, this sentence feels clunky. In addition, it does not mention pixel, the title of this article, as early as possible in the lead (recommended practice, MOS:BOLDTITLE). I would suggest something that begins something like:
"A pixel (px),[3] or picture element (pel),[4][5] is a physical point in a raster image, or the smallest individually addressable element in an all points addressable display device. In digital imaging, it is the smallest, complete element of an image. The address of a pixel corresponds to its physical coordinates. LCD panels are manufactured with fixed pixels in a two-dimensional grid. CRT pixels correspond to the timing and activity of a beam of electrons striking phosphor on the glass. Pixels on print devices correspond to deposits of ink, or other material, on the print medium."
Note: I removed "on a screen" because that excludes other output devices (e.g. printing). Added "complete" because there is an issue about sub-pixels. Other changes, really would like to put it aside and come back to it fresh. [Also note: initially I had intended to confine this to just the first sentence, but expanded it to the first paragraph.]
I am not wedded to the above wording. I would usually want to write it down, leave and then come back to it later in order to look at it fresh. I do desire to see wording that is inclusive of other representations of a pixel than just specific to a screen. At a minimum, print devices should not be excluded and their mention needs to not exclude 3D printing. The rest of the first paragraph, as currently in the article, implies, by emphasis on screen types, that non-screen representations are excluded.
As to Mp, or Mpx, p and px: Both MP and Mpx are mentioned in the text of the article. The problem with trying to easily find the relative popularity is that "p" is just too short for an accurate search; there are too many false positives. MP provides a bit better discrimination, and is a common abbreviation for megapixel. However, I would assert that the vast majority of those uses are where the context supports no other interpretation other than pixel. I would argue against the removal of any of them. Makyen (talk) 00:45, 3 January 2014 (UTC)
If we're going to include px, we need something better than it being used as the code in computer languages like html/css and wiki markup. And as you note, the css px is not even about the same concept as pixel. Here are some useful searches for comparing Mp and Mpx: [2] and [3]. Try your own variants in books, scholar, web, or whatever. Looks like about 2% in books for Mpx. I don't see a major camera company using Mpx for megapixels; Sony and Canon use MP; Nikon and Sigma and Olympus appear to not abbreviate megapixel. Dicklyon (talk) 05:28, 3 January 2014 (UTC)


  1. ^ Foley, J. D.; Van Dam, A. (1982). Fundamentals of Interactive Computer Graphics. Reading, MA: Addison-Wesley. ISBN 0201144689. 
  2. ^ Rudolf F. Graf (1999). Modern Dictionary of Electronics. Oxford: Newnes. p. 569. ISBN 0-7506-4331-5. 
  3. ^ "Syntax and basic data types". 2011 [last update]. Retrieved January 2, 2014. 4.3.2 Lengths  Check date values in: |date= (help)
  4. ^ Foley, J. D.; Van Dam, A. (1982). Fundamentals of Interactive Computer Graphics. Reading, MA: Addison-Wesley. ISBN 0201144689. 
  5. ^ Rudolf F. Graf (1999). Modern Dictionary of Electronics. Oxford: Newnes. p. 569. ISBN 0-7506-4331-5. 

2.0 Megapixel Image Necessary?[edit]

Do you really think we need this image?


In my opinion, it does not add anything to the article, and I think it actually detracts from the aesthetics of the article. Tell me what you think. I'll delete it in a week if nobody objects. Henrib736 (talk) 03:25, 6 April 2014 (UTC)

No objections here, though if possible, maybe it could be replaced with a higher-quality image showing the same thing (megapixel markings on a digital camera). Indrek (talk) 10:21, 6 April 2014 (UTC)
Nowadays cameras usually use at least 16MP, if there are better image with bigger resolution, I think we should replace the current image.Gsarwa (talk) 18:16, 4 April 2015 (UTC)
In my understanding, the point the image is making is that the megapixel war at one point was so intense that the spec was marked in large recessed letters on the hardware, to two significant figures. Samsara 06:46, 5 April 2015 (UTC)

kilopixel should be mentioned[edit]

The search "kilopixel" redirects to this article, but the term isn't found on the page. I can't figure out whether a kilopixel should mean 1000 or 1024. I think the former, but I don't know of a reference. Squish7 (talk) 02:49, 17 March 2015 (UTC)

Do you have any examples of "kilopixel" actually being used in the real world? Because if not, and you can't find any reliable sources, then it's probably not notable enough for inclusion on Wikipedia. Indrek (talk) 08:53, 17 March 2015 (UTC)
It is usual to mention one kilo is 1,000 and one mega is 1,000,000 for people without IT background, but the truth, one kilo is 1,024 and one mega is 1,024x1,024 (difficult to remember).Gsarwa (talk) 18:29, 4 April 2015 (UTC)

Photo Labs versus inkjet printer[edit]

I don't know it is useful and can be understood for/by the readers or not, but in fact inkjet printer has left behind a lot of Photo Labs, although nowadays there are some Photo Labs using also inkjet printer with more expensive price.

So, a camera with 24MP or 24,000,000 pixels (the precise pixels is more a bit than it) can produces in 8R paper size (8"x10") = a square root of (24,000,000/80) = a square root of 300,000 = 547dpi, theoritically. But due to paper photo labs commonly only produce maximum 300dpi, so the result will always 300dpi or lower, while even the modest A4 1440x1440 pixels (1440dpi) inkjet printer can produces more image resolution than paper photo labs, with note excellent inkjet photo paper and excellent inkjet photo dye/ink are expensive.Gsarwa (talk) 18:52, 4 April 2015 (UTC)

I see a few issues with that paragraph.
First, is it notable? Wikipedia is not an indiscriminate collection of information. For something to be included in an article, it needs to have some encylopaedic value, and to improve the article in some way.
Second, is it sourced? Wikipedia does not accept original research, any information must be attributable to reliable sources. Specifically, you'd need sources that back up the claims you're making about photo labs and inkjet printers.
Finally, that text seems to be only tangentially related to pixels, and (assuming the above two criteria are met) would be better suited for inclusion in another article like Inkjet printing, Digital printing or Printer (computing). Indrek (talk) 19:17, 4 April 2015 (UTC)
I agree. It seems to be saying that print resolution is determined by pixels per inch and/or dots per inch; there are pretty loose relationships. In my experience, a good 300 pixel per inch photo printer beats a high-dpi inkjet, but I'm willing to concede that it could just as well go the other way. We'd need sources to clarify what the comparison is, who judged it, and such. Dicklyon (talk) 05:12, 5 April 2015 (UTC)