Wikipedia:Reference desk/Archives/Computing/2017 April 27

= April 27 =

Monitor cracking sound
My monitor (ROG Swift '27) sometimes, but very rarely, make a relatively loud cracking sound. Like cracking your knuckles, whether it's on or when it's been off for a while (even hours). I've read that it's apparently the plastic shell that's expanding and contracting due to temperature?

I believe that's relatively normal and not harmful, at least according to Google? I believe my PlayStation did the same when I owned one. My PC does the same. Matt714 (talk) 09:24, 27 April 2017 (UTC)
 * By your description, it could be the same phenomenon as in a pop pop boat, but it is hard to know for sure if you have not identified where the noise comes from. Tigraan Click here to contact me 11:45, 27 April 2017 (UTC)


 * It's relatively normal for plastic components to make cracking/popping noises as they expand and contract. Plastic expands and contracts quite a bit as it heats/cools. (Expansion of fitted parts, if they don't slide smoothly, puts them under stress. I believe the actual noise occurs when two pieces under stress suddenly slip.) Sometimes improperly installed plastic plumbing will have the same problem when you run hot water through it.
 * Some Apple Mac desktop machines are notorious for this.
 * I can't find a good article explaining this, but here's a FAQ on Samsung's website about one of their TVs.
 * ApLundell (talk) 14:16, 27 April 2017 (UTC)


 * As for how to fix the problem, if the plastic case is attached with multiple screws, perhaps loosening or even removing a few might give it room to expand and contract. If this is not done, the stresses may ultimately cause the case to crack and split. StuRat (talk) 15:08, 27 April 2017 (UTC)


 * Cracking sound usually comes from sparks of high voltage. Since LED back lights is should be obsolete. CCFL back lights still use about 3 kv inverters. LED inverters use higher voltages as wells due the LEDs are in series circuit. ASUS specified the ROG Swift '27 with an in panel back light inverter. If nobody took the monitor wet, it might be the power supply caught dirt of some capacitors begin failing. -- Hans Haase (有问题吗) 23:47, 28 April 2017 (UTC)

The Samsung thing was reassuring. I've asked elsewhere and many reported the same thing, but no failure even after years. The constantly vacillating temperature in Québec right now probably does not help. My PC does it as well. Matt714 (talk) 00:37, 29 April 2017 (UTC)

Non-support of font tags
As I understand it,  tags were deprecated in HTML 4.0 (in the late 1990s, according to HTML), which is why their continued use in customized Wikipedia signatures is controversial. I'm on one side of the fence or the other, depending on the answer to one question: What is likely to happen with all the font tags in our archives if (when) browsers finally drop support? Will browsers continue to recognize them but ignore them, or will they be exposed on the rendered page? Are there any precedents to judge by? &#8213; Mandruss  &#9742;  16:08, 27 April 2017 (UTC)


 * Currently, every web browser that I use ignores unsupported tags. So, if a tag is not supported, it doesn't show up to the user. It is simply ignored completely. I see no reason to assume that it will change if a browser stops parsing a font tag. It will just see the tag and ignore it. No harm. Remember that failing gracefully is important in HTML. 209.149.113.5 (talk) 16:20, 27 April 2017 (UTC)

Example talk 14:24, 22 April 2017 (UTC) will display as: Exampletalk 14:24, 22 April 2017 (UTC) not as (font changed to fint): Example talk 14:24, 22 April 2017 (UTC) and we would expect that to hold true until the end of time, or the end of Wikipedia as we know it, whichever comes first? &#8213; Mandruss  &#9742;  16:37, 27 April 2017 (UTC)
 * So a current signature
 * While that's true, there's also the issue of mediawiki parsing. For example bar shows up explicitly without me using  tags because mediawiki doesn't recognize it as a valid HTML tag. That being said, I doubt the developers at WMF will ever decide to just drop a recognized tag like that.  ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  17:00, 27 April 2017 (UTC)
 * So you're ok with all of your existing sigs in the archives rendering as ᛗᛁᛟᛚᚾᛁᚱPants Tell me all about it.  . If you're happy I'm happy, it will just be 17 wasted and superfluous characters of wikitext for each of your signatures. &#8213; Mandruss   &#9742;  17:45, 27 April 2017 (UTC)
 * Well, my sig doesn't use a font to produce the characters, it uses unicode characters (elder futhark runes, to be specific). But it doesn't bother me too much for it not to render properly from time to time, because most of the time it does. ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  17:58, 27 April 2017 (UTC)
 * You will lose the green in all existing sigs when the support is dropped. &#8213; Mandruss  &#9742;  18:22, 27 April 2017 (UTC)
 * Wow, I honestly didn't realize I had used the font tag. Well, I've changed it to a span tag now. But no, it wouldn't bother me that much. I only picked green (and added the shadows) because it makes it easy for me to spot my signature on talk pages, to know where I left off. The only real vanity in my signature is the phrase "tell me all about it" and the runic characters. ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  18:49, 27 April 2017 (UTC)
 * Just out of interest, what should the seven boxes really look like if I had a runic font?   D b f i r s   19:46, 27 April 2017 (UTC)
 * Something like this: File:Mandruss-screenshot.png. &#8213; Mandruss &#9742;  20:55, 27 April 2017 (UTC)
 * Check out Runic (Unicode block) for some free fonts that support it. Arial is one of the fonts that do, and while it's not free, it should be included with most modern web browsers. If you have one of those installed and your browser doesn't display it check out this site for instructions on how to get it working. ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  21:13, 27 April 2017 (UTC)
 * Oh, and while it's acceptable to call them "boxes" under normal circumstances, when they show up in my signature, they're "squares". ;) ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  21:15, 27 April 2017 (UTC)
 * Needless to say. SpongeBob BoxPants would make no sense whatsoever. &#8213; Mandruss &#9742;  21:19, 27 April 2017 (UTC)
 * Thank you for explaining the "squares". Now that I know what they mean, I won't bother installing elder futhark runes, but just think of you as squarepants (no offence intended).    D b f i r s   07:58, 29 April 2017 (UTC)
 * Curious - why were font tags deprecated??? Whoop whoop pull up Bitching Betty &#124; Averted crashes 19:13, 27 April 2017 (UTC)
 * Because CSS handles it better and easier. Notice how the signature in my last comment is slightly larger than this one (and the previous ones?) That's because my use of the font tag broke the css class I'd put it into. I didn't realize it because I got the result I wanted, but once I switched to a span tag, the CSS class started working again. (I removed the class from my signature between then and now). ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  19:27, 27 April 2017 (UTC)


 * For the same reason, bold, italic, and underline were initially deprecated in HTML4 and then reintroduced in HTML5. I assumed font would be reintroduced as well, but it wasn't. 209.149.113.5 (talk) 11:55, 28 April 2017 (UTC)
 * Yeah, those are tags that a lot of people use inline. I figure they brought them back because  is easier to use than  where you have to set up a bold, italic and underline class for each typography used on the page. I know I appreciated it for that reason. I've also noticed that the new implementation works better with complex CSS rules. The older ones (similar to the  tag) could screw with class properties of tags like    and   , though it seemed to handle    just fine.  ᛗᛁᛟᛚᚾᛁᚱPants   Tell me all about it.  13:24, 28 April 2017 (UTC)

Javascript bitwise operations
How do bitwise operations work in javascript since everything is a float? (note that there is no error when you do something like 5.2 & 3.6). 70.190.164.57 (talk) 20:19, 27 April 2017 (UTC)
 * The bitwise operators work on signed 32 bit integers. If the operand is a double float (and this is the basic type for a JavaScript Number) then it will be cast to a signed int32 first. As doubles are pretty long (52 bit fraction) then this is a practical process, although there's obviously a minor overhead (but hey, that's going to be the least of your worries in any plausible JavaScript context).
 * There's also a lot of scope for optimisation here. The JavaScript engine doesn't necessarily convert back to the double float, unless it needs to. So complex bitwise expressions could be processed almost entirely in the int form (but you'd never know). Andy Dingley (talk) 20:55, 27 April 2017 (UTC)

Display Resolution
What is the difference between "1280*720" and "720*1280" on a phone? 43.245.120.134 (talk) 22:53, 27 April 2017 (UTC)


 * AFAIK resolution is always width x height. So the first would be wider than tall and the second would be taller than wide. Landscape vs portrait orientation. Both measurements are in pixels of course. &#8213; Mandruss  &#9742;  23:00, 27 April 2017 (UTC)
 * I'm unsure if that truly the case with the phones; you'll find websites advertising either ways ("1280*720" and "720*1280"). 103.67.158.220 (talk) 14:07, 29 April 2017 (UTC)
 * "720*1280" is uncommon tho compared to the common practice it would be technical correct. Ofcourse you could aswell switch the numbers for any TV- or PC-Screen but i bet Marketing would instantly veto that, because they want the big numbers mentioned first because people are likely to check out items in a list by the first "mark" they look for and ofcourse then more is usually better. --Kharon (talk) 04:09, 30 April 2017 (UTC)