Wikipedia:Reference desk/Archives/Computing/2015 December 31

= December 31 =

Why is Li-Fi faster than Wi-Fi?
The article on Li-Fi claims that it is up-to 100 times faster than Wi-Fi. What Why would data transmission be much faster using visible light communication, infra-red and near ultraviolet?--Denidi (talk) 20:24, 31 December 2015 (UTC)
 * I take it you mean why would data transmission be faster. Presumably it's because visible light has a much higher frequency.  That means you can modulate it faster without introducing "ringing" artifacts that confuse the receiver.  I'd start with bandwidth and Shannon entropy. --Trovatore (talk) 20:32, 31 December 2015 (UTC)
 * Yes, I meant 'why'. Thanks for the links, they help me understand why it's faster, although I still doubt the 100x bit, unless you are looking at the slowest WiFi (b) and running the Li-Fi under optimal lab conditions. The 802.11n can be pretty fast, and real-life implementations can imply further limitations.--Denidi (talk) 20:37, 31 December 2015 (UTC)
 * You are correct to doubt the hundred times faster, because it's true only when compared with IEEE 802.11 (legacy mode). The problem with Li-Fi comes when you stand between the transmitter and the receiver, since the signal is almost completely blocked.    D b f i r s   20:55, 31 December 2015 (UTC)
 * Really? The Li-Fi article claims you can get up to 224 Gbits/s.  That's more than 300x faster than the fastest value I see claimed for 802.11n, namely 600 Mbits/s.  Supposedly with 802.11ad you can get up to 7 Gbits/s, so it'd be only 30x or so faster than that. --Trovatore (talk) 21:36, 31 December 2015 (UTC)
 * I think I got my sums wrong. I've striken part of my comment above.    D b f i r s   09:28, 1 January 2016 (UTC)


 * It's really not all that complicated. The electromagnetic spectrum covers many orders of magnitude.  A single near-infrared channel has an inherent frequency of about 340 THz, while a single microwave channel has an inherent frequency of either 2.4 or 5 GHz.  That's a difference of 68,000 fold.  Actually sending information means modulating that channel in useful ways.  For WIFI, advanced modulation schemes come increasingly closer to the theoretical single channel limits, and recent technologies have moved to MIMO schemes that use multiple channels to overcome the limits of a single channel.  By contrast, LIFI systems are still using only a tiny fraction of the theoretically available single-channel bandwidth, and usually operate with only a single channel (though similar fiber-optic technologies use multiple channels).  For comparison, consider that we have fiber optic systems that deliver as much as 1 Petabit / s (1000000 Gbit / s) over a single fiber.  LIFI is more-or-less about finding ways to adapt the science behind fiber optics to work over the open air without the need for the glass fiber cables.  Dragons flight (talk) 10:39, 1 January 2016 (UTC)
 * ... but open-air Li-Fi will always have problems of dust, weather, birds and people getting in the way, etc. I can't see it ever being useful except in laboratory conditions, but I'm happy to be proved wrong ...?    D b f i r s   13:30, 1 January 2016 (UTC)
 * Nothing will be a magic bullet solution, but we've been using wireless communications that require line-of-sight for decades. One strategy is to design systems like Li-Fi to interoperate well with "regular" wireless and fall back to the latter when you don't have line-of-sight. --71.119.131.184 (talk) 14:48, 1 January 2016 (UTC)
 * I'd say my laptop or smartphone never have line-of-sight to my router, and I would not like to worry about it. A further problem is that even if I could get more than several hundred Mbit/s, I would also need another type HDD or SDD, since normally they cannot handle these rates.Llaanngg (talk) 15:01, 1 January 2016 (UTC)
 * NVM Express SSDs can normally get speeds 1-2 Gb/s and they're probably saturate the a 4 lane PCI Express 3.0 link soon (actually I suspect a number already are). Probably within a couple of years of PCI Express 4.0 becoming common for NVM Express many will be saturating that. Of course this also depends what you're competing, with, what purpose etc. For example, your SSD or HD speed isn't relevant if what you're doing is sending uncompressed UHD video from a device to a display where neither device is planning to store it on a SSD or HD. Likewise, while you may be able to get 600-700 mbit in the same room on 5ghz ac for a single device, what happens if you have a hundred people in a conference centre? Even fairly open plan office buildings can have problems particularly when the designers decide it's a good idea for everyone to be using wifi. For that matter, while li-fi would seem likely to have worse LOS issues than 5ghz let alone the 60ghz proposals, there's the open question of range comparisons. How will li-fi compare in a stadium or other open but large area with a lot of people? Even the LOS is a complicated thing. While 5ghz has quite a few channels available, as it improves and people demand more people may begin to hit spectrum saturation problems due too many competing networks again, particularly in high density apartments and similar. Nil Einne (talk) 07:13, 2 January 2016 (UTC)
 * I'd say my laptop or smartphone never have line-of-sight to my router, Well, then we better tell them to stop wasting their time then :/ All you would have to do is put a LiFi Access Point on your ceiling. Where I work we have open plan offices with hundreds of employees, and I AM within line of site to my nearest AP pretty much anywhere I go. Currently the wireless here isn't quite as reliable or fast as the wired connection, so I still "dock" my laptop into the LAN when I'm at my desk, but a great LiFI network could probably replace almost all of the LAN on the floor. Vespine (talk) 23:11, 3 January 2016 (UTC)

Centre point of the monitor screen
Hello,

How do I find the centre point of the monitor screen? A cross sign would be helpful... Regards. -- Mr. Zoot Cig Bunner (talk) 23:36, 31 December 2015 (UTC)
 * Under Windows, you can just use Screen.GetBounds. To find the actual centre of the actual monitor (especially if it's a CRT monitor), you'll need to use a ruler. Tevildo (talk) 10:53, 1 January 2016 (UTC)
 * I don't understand it Tev. I guess its something to do with coding... Never mind, thanks anyway. Regards. -- Mr. Zoot Cig Bunner (talk) 21:13, 1 January 2016 (UTC)
 * What do you actually want to do? When you find the centre of your monitor, what do you want to do with it? Tevildo (talk) 22:09, 1 January 2016 (UTC)
 * Depending on what you're trying to do, you could set an image like this as your desktop background. -- BenRG (talk) 00:58, 2 January 2016 (UTC)
 * Tevildo, BenRG: I want to make sure that the software window(s) I open/re-open sometimes/many times, stays in the middle of the monitor. The imbalance becomes irritating sometimes when suddenly it muddles up, especially after you fix it using your ruler...  What Ben stated is desirable, but something alike or different could probably work, if that's not available... -- Mr. Zoot Cig Bunner (talk) 18:42, 2 January 2016 (UTC)
 * Btw, thanks Ben. I'm doing what you stated. -- Mr. Zoot Cig Bunner (talk) 18:46, 2 January 2016 (UTC)
 * Thanks for the explanation. It's not generally possible to control where a window opens on the screen.  There are applications (such as CMDOW) that you can use to move a window after it's opened - however, for what you want to do, that might be too complicated.  One simple suggestion is to edit your desktop wallpaper to put a pair of little mitre marks &#x231c;&#x231f; where you want the corners of your window to be, and move it if it's out of alignment. Tevildo (talk) 19:39, 2 January 2016 (UTC)
 * -- Mr. Zoot Cig Bunner (talk) 21:23, 3 January 2016 (UTC)