Wikipedia:Reference desk/Archives/Computing/2014 August 1

Computing desk
< July 31 << Jul | August | Sep >> August 2 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


August 1

edit

IC chips

edit

why does the pictures of IC chips carry a photo of metal coins ? — Preceding unsigned comment added by 101.221.131.80 (talk) 07:42, 1 August 2014 (UTC)[reply]

I don't see any photos of coins in our IC chips article, but this is a common device for illustrating the size of a small item such as a chip. The viewer is expected to know the size of the coin, and can therefore deduce the approximate size of the chip.--Shantavira|feed me 08:32, 1 August 2014 (UTC)[reply]
Example: Contactless smart card - note the caption below the picture "Size comparison of chip (compared to a Canadian penny)". Mitch Ames (talk) 14:33, 1 August 2014 (UTC)[reply]
This use of a coin is known as a scale marker or a scale object, but surprisingly we don't have an article on that, and none of the meanings at the scale disambiguation page are appropriate! The closest one is to map scales, but that is of course specific to maps. Anyway, as others have pointed out it is to give you some sense of context in understanding how large the object is. A coin is often a good choice, but serious scientists use things like these [1]. SemanticMantis (talk) 16:31, 1 August 2014 (UTC)[reply]
Carancas meteorite shows a scale cube in action. Well, action in the passive sense. It may be worth putting a Scale object section in Linear scale, as most such objects seem to be used to indicate linear scales.
There is a guideline somewhere against using coins for scale, as they vary a lot in size. There is a template {{NoCoins}} for use on such images. --  Gadget850 talk 11:20, 2 August 2014 (UTC)[reply]

java stuff

edit

can we develop java application without having user-defined class?? means there should not be class <class_name> >>you can use pre-defined classes... response with explanation........182.18.179.2 (talk) 11:42, 1 August 2014 (UTC)[reply]

No, not really. For an application, the runtime wants the name of a class (or of a JAR with a manifest which names a class) to run. That class has to have a public static void main(String[]) method, which means only classes intended to be applications can be executed - the Java runtimes does, in essence, reflection to find that method and verify its signature - an example is here (the code after // load the target class, and get a reference to its main method). In this regard Java is unlike languages like Perl or Python, where there is essentially an implicit main at the top of the script (so all code that isn't inside a function is run when the script is). -- Finlay McWalterTalk 12:29, 1 August 2014 (UTC)[reply]

"Mpbile View" vs. "Desktop View"

edit

How can I make the "Desktop View" the default for all my devices? I see that the mobile view might be good for phones, but I usually use my iPad and prefer the original desktop view. Right now I have to switch every time I access Wikipedia.

Just to clarify: you are talking about the "desktop view" and "mobile view" of the Wikipedia websites? --Andreas Rejbrand (talk) 14:17, 1 August 2014 (UTC)[reply]
Yes. Peter Flass (talk) 21:34, 1 August 2014 (UTC)[reply]
What are your cookie settings? My iPad stays in the desktop mode just fine. Dismas|(talk) 05:58, 2 August 2014 (UTC)[reply]
"block cokies from third parties and advertisers" Peter Flass (talk) 16:10, 6 August 2014 (UTC)[reply]

Why do monitors have background light?

edit

Couldn't they just have a mirror and a light in front lightining the colored pixels? That is also the way we see stuff in general.OsmanRF34 (talk) 22:15, 1 August 2014 (UTC)[reply]

I believe you would get glare and reflections off the glass that way. StuRat (talk) 22:18, 1 August 2014 (UTC)[reply]
With present technology, it is easier and cheaper to mass-produce large volumes of reliable monitors that use a backlight and an active matrix thin film transistor array of color-filtering pixels. In principle, the individual colored pixels could be built from light-emitting components - like any of the many varieties of light emitting diodes. But, it is difficult to make a display using that technology that satisfies all the requirements: it must be bright enough to use; must withstand an ordinary lifetime of consumer-use without suffering damage or defective pixels; and must provide competitive resolution or dot pitch.
Placing the light-emitting element in the active color matrix means a greater chance of "burning out" the pixels - this kind of wear-and-tear is user-visible and is very undesirable. Why do these devices fail? We have an article on that: List of LED failure modes. Among the prominent failures: thermal stress causes the atoms inside the semiconductor material to migrate, slowly reducing the device performance. Catastrophic failure modes, like thermal cracking of glue, epoxy, packaging material, or wirebond, would cause immediate failures. All sorts of other failures occur, like material degradation due to long term exposure to sunlight, humidity, heat, and voltage. These lifetime reliability issues become worse when you pack more thermal energy into a tiny space - like an LED the size of an individual pixel on a modern display.
Nimur (talk) 03:32, 2 August 2014 (UTC)[reply]
A possible solution to the heat problem is to use fiber optics to deliver the light for each pixel to the screen, while positioning the actual LEDs wherever they can be effectively cooled. StuRat (talk) 13:14, 3 August 2014 (UTC)[reply]
Like all technical problems, I'm sure there are zillions of possible solutions; but in this instance, none of those solutions are cheap enough, reliable enough, and functionally useful enough to justify their adoption. Don't you think that dedicated and highly-skilled experts, working in a competitive marketplace fueled by innovation, would pursue any possible solution that actually solved more problems than it introduces? Many times, a solution that requires thinking outside the box is really just uninformed about why the walls of the box were built that way in the first place.
Coincidentally, this week will be the Emerging Display Technologies conference in downtown San Jose. Among the other business presenters, Intel's CTO for Perceptual Computing will be giving the keynote speech. The highlights are expected to be new technologies in flexible, curved, and interactive displays. Market leaders like Dolby and Corning and Intel, as well as many smaller companies and start-ups, will be represented as well. The problems are many-faceted: software, hardware, electronics, materials science; not to mention the ecosystem - new display technologies must be compatible with existing multimedia content and electronics! Something to follow, if you're interested in the state of the art for displays! Nimur (talk) 23:42, 3 August 2014 (UTC)[reply]
RE: "Don't you think that dedicated and highly-skilled experts, working in a competitive marketplace fueled by innovation, would pursue any possible solution that actually solved more problems than it introduces?". That amounts to "Don't ever try to think of any improvements to any products, as somebody else will have already done so". And "experts" are often unable to think outside the box, and a fresh perspective is needed. For example, I doubt if many film camera engineers were thinking of how to replace film with digital cameras. StuRat (talk) 05:41, 4 August 2014 (UTC)[reply]
You know that the digital camera sensor was invented at Eastman Kodak, which was the world's preeminent manufacturer of camera film, right? They also invented many of the technologies we now call photolithography, which make modern microelectronics possible. The fact that their corporate head office missed the memo is well-studied; but their engineering experts were twenty years ahead of the consumer market. Nimur (talk) 05:55, 4 August 2014 (UTC)[reply]
Right, but since they didn't pursue it, somebody else must have done so. And using your logic, they should have just said "Well, Kodak are the experts here, and if they didn't pursue it they must know it's a dead end, so we shouldn't pursue it either". StuRat (talk) 04:04, 6 August 2014 (UTC) [reply]
In principle, the individual colored pixels could be built from light-emitting components. See OLED. Vespine (talk) 01:53, 4 August 2014 (UTC)[reply]
Are you able to find any OLED - or in fact, any - displays that are commercially available today that do not use a backlight? For example, look at LG's OLED television product brochure. LG is widely regarded as a market leader in OLED technology - but they still use a backlight (perhaps also made from thin film or OLED material) behind the colored OLED layer to provide uniform illumination and sufficient brightness to be useful. You'll probably find the same for all other current competitive products in the marketplace. A few companies offer special advance-prototypes, but for research and development, not for general consumer sale. Perhaps in a few years, thin film self-luminescent LED or OLED technology will overcome some of its present limitations - but at least for now, it looks like the backlight is sticking around. Nimur (talk) 05:14, 4 August 2014 (UTC)[reply]
According to [2] and seemingly confirmed by various other sources like [3], [4], [5], [6], [7], [8], the Samsung KE55S9C (or probably all Samsung OLED TVs) do in fact use individual coloured OLED subpixels without any think that could be called a "backlight".
It doesn't look like the Samsung TV is sold any more but it's definitely a commercial product you could once buy [9] [10] [11], perhaps partially because of production problems [12] (Samsung appears to deny that they stopped production, they didn't seem to deny they may be having problems) [13] [14]. Probably the fact that they were producing a 1080P display when everyone is going bananas over 4k and LG were seeming ready for 4K OLED displays didn't help either.
So while white subpixel with filter based OLED TVs may be more dominant they don't seem to be the only thing that have been in commercial production. (You may argue the quantities are too small to be concerned commercial, but I would suggest that applies to both LG and Samsung at the moment even though LG may initially at least dominate the mass-market considering Samsung's problem.)
Note that from the sources, I find your statement about LG's tech misleading anyway. As understand the LG OLED tech, it's not simply "they still use a backlight (perhaps also made from thin film or OLED material) behind the colored OLED layer". Rather there's no "also" about it. What makes the display an OLED is that it used white OLED subpixels which are then RGBW colour filtered. I can't actually find much detail on the filter design, but I don't see why they'd use an OLED as the filter, more likely it would be a fairly "simple" passive colour filter. (Yes I know simple may be a bit misleading when we're talking about filtering each individual subpixel in a ultra thin 8,294,400 pixel (so 33,177,600 or something subpixels) 68 PPI 65" curved display, but you get the idea.)
Note that while some sources use the term backlight, I've avoided it. Whether you want to call the WOLED subpixels a backlight is going to vary between individual and I'm not sure there's any correct answer. It seems political as much as anything. Obviously producing white light and then filtering it is not the most efficient process but it could still end up being more efficient than some alternatives. And do you say a traditional traffic light (or whatever) has a backlight simply because it uses a white light which is colour filtered? The fact remains even with WOLED, you're still addressing each individual subpixel and turning them on and off individually as basically a coloured subpixel. This isn't like an LCD where even with those IMO misleadingly called LED displays, the backlight is basically one or a few lights. Which you are hiding behind an opaque LCD when needed (and yes, also colour filtering to give the appropriate colour).
P.S. Speaking of Kodak, from what I can tell a fair amount of the LG OLED early tech comes from Global OLED Technology LLC i.e. was acquired from Kodak.
P.P.S. Of course we're only talking about TVs. I'm lazy to check, but I strongly suspect a number of the OLED displays on the various mobile devices lack anything which could be fairly called a backlight. Yes I know Apple doesn't use them, but they aren't the only manufacturer of mobile devices out there.
P.P.P.S. Are we not getting a bit off topic since the OP's question seemed to primarily relate to why displays were backlit instead of front lit rather than why displays had a backlight instead of individually coloured and addressed lit pixels? This has been answered in parts, but it seems to me even the first answer by Nimur didn't really address it that well. For the benefit of the OP, I'll provide [15] which mentions how displays (whether LCD or OLED) often purposely filter out any light coming from the front so it doesn't look too washed out in direct sunlight.
Nil Einne (talk) 07:34, 4 August 2014 (UTC)[reply]