Jump to content

Wikipedia:Reference desk/Archives/Science/2016 September 15

From Wikipedia, the free encyclopedia
Science desk
< September 14 << Aug | September | Oct >> September 16 >
Welcome to the Wikipedia Science 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.


September 15[edit]

the other memory effect[edit]

hello, I remember reading on the Internet where someone accidentally discovered a different kind of memory effect in rechargeable batteries, namely that variations in current during the charging phase manifested as tiny variations in voltage during the discharging phase. The guy (or girl) was thus able to record, and successfully retrieve, something like 1000 bits. I can't for the life of mine google the page up again, but am pretty sure I'm not hallucinating. Does someone by chance know what I'm talking about? Asmrulz (talk) 04:26, 15 September 2016 (UTC)[reply]

Are you referring to [1] [2] [3]? If so, it doesn't seem to refer to 1000 bits of information but an deviation in voltage that is a few parts per thousand (although I didn't read the sources that carefully). Nil Einne (talk) 08:31, 15 September 2016 (UTC)[reply]
no, that's not it... Asmrulz (talk) 23:31, 15 September 2016 (UTC)[reply]
The effect is not implausible. For many galvanic cells, metal is deposited on the cathode during charging, and the rate of charging may affect some property of the metal being laid down, for example concentration of inclusion of impurities or size of crystals formed. During discharge, variations of this property at the surface of the electrode as it is becoming redissolved might in turn affect the output voltage in a predictable way. Such an effect should be pronounced for a cell with a mixed electrolyte: two metals with similar but nonidentical electronegativity (e.g. nickel and copper), where the ratio of metals in the alloy formed at the cathode should be significantly affected by the charging rate. —Quondum 23:53, 15 September 2016 (UTC)[reply]
When searching for information on this sort of thing, the phrase "side-channel" helps a lot. Try [ side-channel battery ] on Google, for example. --Guy Macon (talk) 23:57, 15 September 2016 (UTC)[reply]

Optical connection[edit]

Did optical connection had a ADSL line? How did work a special government line in optical connection? I think that optical connection always did had a solid signal of data, did I'm right?--Alex Sazonov (talk) 09:39, 15 September 2016 (UTC)[reply]

Why optical connections did not had included dedicated lines of data transfer? I suppose that it did had a voltage problem, because optical connections did never had supported a high voltage transfer.--Alex Sazonov (talk) 11:17, 15 September 2016 (UTC)[reply]
Did I know right that frequency of electron oscillation done most than frequency of photon oscillation, that’s why a light done weak?--Alex Sazonov (talk) 11:57, 15 September 2016 (UTC)[reply]
I suspect as always, few here are really sure what you're saying. If you'd write better in whatever language you have a decent command of (which no one seems to know since AFAIK you've failed whenever you've tried in English and Russian) maybe you could get more help. However it's very common that *DSL lines have fibre optics somewhere upstream. Very often copper is only in the last mile and maybe some internal networks. The rest is fibre optics, which is obviously optical although not necessarily using visible light. Generally the length of the copper last mile is getting shorter and shorter. And some of us including me right at this very moment have no copper last mile (again discounting internal networks) and are using FTTH. Fibre connections can use less power, although this is likely only relevant in data centres [4] [5]. (Although possibly copper is lower power consumption for some very short interconnects [6] [7]? And I'm also not sure how power consumptions compares at lower data rates.) The bigger advantage with fibre is that you can generally get higher data rates a lot easier, especially over longer distances. Fibre does tend to require a fair amount more professional expertise for installs. Fibre connections can be in various forms e.g. Passive optical network, point to point, terms like DSL do not apply to the fibre part of the connection. Nil Einne (talk) 14:13, 15 September 2016 (UTC)[reply]
Another property of FTTH (Fiber to the home), though not really relevant for 99.9% of users, is a very low ping. See here (sorry,could not find any better source). TigraanClick here to contact me 14:32, 15 September 2016 (UTC)[reply]
To illustrate this, I get about 2-4 ms to the first external hop (GbE to the ONT connected to a GPON network). It probably could be lower but I'm on a residential connection and despite my ISPs advertising, they aren't really that well optimised for super low latency. By comparison IIRC I got about 10-13 ms on VDSL with interleaving off. This is a resonable difference but most of the time it makes very litte difference since even when I'm doing something latency sensitive the difference is a relatively small proportion if I'm hitting Australia, let alone the US let alone further afield. (Of course it adds up and if I'm only hitting Australia even VDSL adds a resonable percentage. But the ultimate point is VDSL without interleaving is reaching a level that even for somewhat latency sensitive applications it's often not significant.) If your ISP won't let you turn off interleaving or your connection can't support it, you're at a level that it's starting to be significant. Nil Einne (talk) 14:51, 15 September 2016 (UTC)[reply]
You might want to avoid saying "the last mile is getting shorter" to somebody with poor English skills. Instead, I suggest "the length of the copper wire at the user's end is getting shorter". StuRat (talk) 14:34, 15 September 2016 (UTC)[reply]
I always knew imperial units were confusing. TigraanClick here to contact me 14:54, 15 September 2016 (UTC)[reply]
That's more of a proof that trying to use two systems at once is a bad idea. As for the current case, "the last km is getting shorter" would be just as confusing. StuRat (talk) 15:04, 15 September 2016 (UTC) [reply]
I know. I just saw the opportunity for a cheap-shot joke, and took it. TigraanClick here to contact me 09:05, 16 September 2016 (UTC)[reply]

The Russian state telephone company of Moscow did wanting to install optical connection in my flat as free services without money, but I’m did not wanting it, because this optical connection did not had supported the ADSL line which used by my ADSL modem, did I’m doing right?--Alex Sazonov (talk) 16:05, 15 September 2016 (UTC)[reply]

It's unlikely anyone here can tell you if you're doing the right thing since we have no idea what you're trying to achieve and ultimately you have to decide what works best for you. Is there some reason why you need your ADSL line & ADSL modem? If your telco is offering to upgrade you for free it would seem likely this includes a free router. Alternatively some ADSL router modems also support a WAN port connection. In any case, a decent router with a WAN port is fairly cheap and you're much less restricted when you don't need one with a DSL port (you can of course connect a DSL router to another router but depending on options may have problems due to double NAT etc). So I don't see any reason to want to keep your ADSL modem. The internet service provided over the fibre connection would likely be faster for a comparable price (although I admit I have no idea about prices in Moscow). So it doesn't seem there's any real reason to keep the ADSL line.

If you want a voice or analog data line, the fibre service may be a bit more limiting since you'll likely need a UPS to either the ONT or the ONT & voice service device/s (probably the router) if you want continued service in a power cut depending on how the voice line is provisioned (which will depend on the telco and/or ISP or voice service provider). And even if one is provided by the telco, it probably won't last as long as the ones in their cabinets (or whatever) compared to the situation if still using a copper last mile. Additionally, while there are standards from all I've read fax over ATAs is a bit hit and miss. You really should be moving to all digital fax on your end if you move to a fibre connection. But you didn't mention either of these, but only the ADSL line which as I've said you haven't provided a reason why you want to keep that I can understand.

Nil Einne (talk) 16:37, 15 September 2016 (UTC)[reply]

Thanks a lot. I’m did decide to install optical connection as it did a free services without money. I have a small theoretical question, that is, if in mechanical models a frequency of body oscillation always did determined a speed and acceleration of this body, so that did a frequency of elementary particulars oscillation always determine a speed and acceleration of those elementary particulars?--Alex Sazonov (talk) 17:29, 15 September 2016 (UTC)[reply]
  • ADSL won't do fibre to the cabinet. You need VDSL, which is time for a new modem. They're not expensive, they probably look identical to your existing modem, but they are different.
You might be tempted to use an existing ADSL router with a new VDSL modem, using PPPoE. In my experience this doesn't work well. It works (there is a connection), but there's no performance gain as there ought to be with fibre and it can become unreliable. Andy Dingley (talk) 17:55, 15 September 2016 (UTC)[reply]
That's a bit confusing. It depends on the what the DSLAM installed in the cabinet supports and what your telco and ISP supports. A significant chunk of NZ is cabinetised and most people are using ADSL2+ to the cabinets. You can get VDSL2, but as it costs more many people don't bother. The cabinetisation isn't useless as it means you generally get 16-24 mbps (although the standard is I think more than 10 mbps) compared to often under 10 mbps when connected to the exchange although your upstream is still piss-poor as we don't even use annex M. Additionally most ISPs here require a splitter installed in the house for VDSL2 and a fair few people don't have these or have older ADSL2+ ones. Generally ISPs here will absorb the cost if you sign up for a 12 month contract, as they will also provide a VDSL2 router modem for free, but not everyone will be happy with that. I believe some ISPs are providing VDSL2 router modems with WAN ports to all customers who take up the option for the router modem, so it's possible your "ADSL2+ modem" will also work with VDSL2 or FTTH although these tend to be fairly cheap and crappy devices. If your telco has alread rolled out cabinets in the form of FTTN and is now moving to closer cabinets in the form of FTTC, it may be that they won't bother to support ADSL2+ with FTTC although as our article hints at, it's also possible they won't use VDSL2. Nil Einne (talk) 05:32, 16 September 2016 (UTC)[reply]
It did only one solid pocked pocket of data for everyone line of signal transaction in optical connection, is it right?--Alex Sazonov (talk) 08:19, 16 September 2016 (UTC)[reply]
It seems me that be better using one pocket of data for everyone signal.--Alex Sazonov (talk) 08:45, 16 September 2016 (UTC)[reply]
Again, it's very difficult to know what you're asking, but as I mentioned above, fibre connections especially for home users are often Passive optical network which are point to multipoint so shared among several users because it's cheaper, but they don't have to be. There's no reason why they can't be point to point and some providers do offer these generally for business connections. If your telco offers dark fibre with sufficient control you can probably find someone who'd offer such a connection but you'd have to pay accordingly. PON are often GPON nowadays with a rate of 2.488 Gbps down and 1.244 Gbps up so even with a contention ratio of 1:32 you shouldn't get that bad speeds. If the telco pushes it and goes for 128 you're starting to get low speeds although since you're comparing it to ADSL, even that not so much. More importantly, contention happens basically throughout the entire network and there's no guarantee the GPON is going to be where your connection is saturated. Nil Einne (talk) 10:06, 16 September 2016 (UTC)[reply]
Thank you. Did the distance access working in passive nets, it did as a clever packed pocket of data in solid pocked packet of data transfer in optical connection?--Alex Sazonov (talk) 10:30, 16 September 2016 (UTC)[reply]
I be thinking that distance access always did working as a dedicated line in line, but optical connection did never had a dedicated line in line, that's what about I'm talking.--Alex Sazonov (talk) 11:25, 16 September 2016 (UTC)[reply]
Anyone answer me, did a optical connections always using only as passive nets?--Alex Sazonov (talk) 13:14, 16 September 2016 (UTC)[reply]
Could be a light spector consist of another light spectors, so could a light spector be in another light spector in one optical line - one optical conductor of optical cabel, if color could consist of another colors, so that could be a dedicated optical line in optical line, that is, could be two or more light spectors in one optical line - one optical conductor?--Alex Sazonov (talk) 14:43, 16 September 2016 (UTC)[reply]
Light could specular reflecting and specular refracting light, that is an optical problem.--Alex Sazonov (talk) 15:07, 16 September 2016 (UTC)[reply]
I think that signals of opposite electromagnetic potential could transferring by one conductor – one line on different frequency, so that did optical conductors do it? — Preceding unsigned comment added by Alex Sazonov (talkcontribs) 15:52, 16 September 2016 (UTC) --Alex Sazonov (talk) 15:53, 16 September 2016 (UTC)[reply]
So signals of same electromagnetic potential for example + 5 Volt and + 10 Volt of course could transferring by one conductor – one line on different frequency, is a same optical effect being?--Alex Sazonov (talk) 06:57, 17 September 2016 (UTC)[reply]
Yes. Wavelength-division multiplexing is a technology which multiplexes a number of optical carrier signals onto a single optical fiber by using different wavelengths (i.e., colors) of laser light. Such networks use Optical add-drop multiplexers to route channels of light into or out of a single mode fiber. You may find more information at the linked articles and at the articles about Passive optical network and Optical burst switching which is a compromise between the yet unfeasible full optical packet switching and the mostly static optical circuit switching. AllBestFaith (talk) 14:50, 17 September 2016 (UTC)[reply]
Much thanks. May being optical paradoxes do it possible if optics done a base of quantum?--Alex Sazonov (talk) 18:13, 17 September 2016 (UTC)[reply]
So if elementary particulars could oscillating in all directions in magnetic or electromagnetic waves it’s may being do additional focuses and additional spectors in optical environments.--Alex Sazonov (talk) 19:47, 17 September 2016 (UTC)[reply]