Wikipedia:Reference desk/Archives/Computing/2015 January 15

= January 15 =

minimum USB latency versus bluetooth
how does minimum USB latency compare with minimum bluetooth latency? Is it noticeable due to the presence of the wireless protocol? (e.g. on mice). Thanks. 212.96.61.236 (talk) 04:26, 15 January 2015 (UTC)
 * My gut tells me that USB will have lower latency than Bluetooth. Personally I prefer wired keyboards and mice for many reasons.Fractal618 (talk) 18:11, 15 January 2015 (UTC)


 * Bluetooth latency in somewhat noticeable when playing fast-twitch computer games, although I can't give specific figures. I "upgraded" to a good wired keyboard and mouse setup (through USB), and all problems were resolved.  If you are just using your computer for browsing/email/etc., you most likely will never notice it. Justin15w (talk) 18:10, 16 January 2015 (UTC)

Explanation for Laptop Charger Voltages
Hello, I have noticed that my current laptop uses a 19.8V charger. What is the reason for such value? Apart from shorter charging times and having a thinner cable I can't think of other reasons for using a relatively (for modern electronics standards) high voltage. Are these the main reasons behind that or am I missing something.

Thanks in advance. Regards 197.148.33.57 (talk) 11:12, 15 January 2015 (UTC)


 * Assuming you are in the US the 19.8V charger uses a transformer to step down 120VAC to 19.8VDC, this causes a six fold increase in available amperage, if the charger were a 10VDC charger it would cause a twelve fold increase in available amperage, which would actually mean even faster possible charging times. As for the reason why your laptop charger outputs 19.8V, I can only assume it is a proprietary reason, in place to discourage "universal" chargers and force consumers to buy their chargers. On another note, 19.8V seems very normal for a laptop charger. Are you asking why it is 19.8 and not 20, or 19.5?Fractal618 (talk) 18:08, 15 January 2015 (UTC)


 * Careful, you can't assume that charging speed depends only on current and not voltage. On the last point, 197.148 asked why it is "a relatively (for modern electronics standards) high voltage", so I think the question is why it isn't something like 5 or 10 volts, not why exactly 19.8. (Nothing is going to care about the difference between 19.8 and 20 anyway.) --65.94.50.4 (talk) 19:24, 15 January 2015 (UTC)
 * Yeah I agree on the OP compared 19.8V to significantly lower voltages. But I don't think it's particularly surprising, the OP seems to be only thinking of certain electronics, there are definitely plenty of others with such a "relatively" high voltage. In the particular case of laptops, AFAIK many more traditional types ones still have a battery pack made up of multiple cylindrical lithium ion cells, perhaps 18650. Even if they are more specialised cells, for a relatively large laptop I suspect these will still often be several cells rather than one large cell, see e.g. . If these are LiCoO2, which AFAIK are still the most common for such purposes, they probably need a voltage of ~4.20 for charging (the precise voltage will depend on the specific cell type and any limitations imposed by the manufacturer to increase lifespan). These will normally be charged in series albeit with individual cell monitoring and balancing. If you use 12V, a fairly common voltage for a number of different devices, you can't even charge 3 cells without needing a stepup/boost powersupply. 19.8V allows 4 cells without problem. You can even charge 5 cells of other chemistry, or simply 5 LiCoO2 if you limit the CV charging to ~3.96V (which probably won't limit capacity that much, may be at most 20% compared to 4.20V). Even if you need to charge 5 or 6 cells at a CV of 4.2V, stepping up from 19.8V to your needed voltage is likely to be easier than stepping up from 12V. Note that the voltage shouldn't affect charging times as the OP seemed to imply. Unless you have to limit current because you can't supply sufficient current at a very low voltage (e.g. you will need quite a high current if you're using 5V to give a decent charge rate, remembering also you may need to both charge the battery pack and supply the laptop at the same time). Note in particular, for a relatively high powered laptop even if you did use 12V, there's a very good chance the power (or current) requirements would be well beyond what many 12V adapters could safely provide if you wanted a similar charge rate. Nil Einne (talk) 05:24, 16 January 2015 (UTC)

Hello all! Thanks for your input. Yes, I was asking why 19.8V and not 5V, for instance. Einne's explanation makes lots of sense. It is probably the reason behind the value itself and also behind the use of a voltage rating defined to the tenth of a Volt. Regards 41.63.165.48 (talk) 11:02, 16 January 2015 (UTC)

Removing name assigned in pdf numbering field
(Using Adobe Acrobat Professional 2008). I converted a series of tiffs to a single pdf. Now, for whatever reason, the field at the top where it lists the page number has the name of the document the tiff was originally assigned in it at every page. For example, at page four that field says "CAP 1 VISA STATEMENTS_Page_04.tiff" (the first part being the name of the tiff files and the file extension even though the doc is now a pdf. I've also added entirely unrelated documents to this pdf, and the same "numbering" is assigned. I've never seen this before That field normally just says just the numeral, that is, if I go to page four the field will say "4" -- that's it. Anyone know how to remove this? 108.46.137.89 (talk) 14:36, 15 January 2015 (UTC)
 * I can't help with this specific issue but if I were in your situation I would first try to convert with another method, rather than find a way to solve this problem. In my experience that is often faster for problems with format conversion. E.g. here is a tiff->pdf converter that might not have the same behavior. SemanticMantis (talk) 15:50, 15 January 2015 (UTC)
 * Thanks but I needed to fix this document. It had been laboriously compiled from various separate documents (and bates-stamped) and would have required hours to reconstruct. However, I have fixed it myself. Here's what I did. First, I'm sure there is a way to actually remove this page assignment directly; still no idea how to do that. My fix worked and was simple though it's not as direct. I took a small existing pdf document that had normally page numbering. I then simply went to "Insert Pages" in that pdf and inserted the problematically numbered much larger document in its entirety and saved this Frankenstein. I then deleted the pages that weren't originally part of the larger documenst and saved this over the prior doc. 108.46.137.89 (talk) 16:46, 16 January 2015 (UTC)

Decimal - long conversion and long - DateTime conversion in C#
In late November I posted a question to the Mathematics Reference Desk about linear regression analysis for a project we have at work. I'm now currently implementing the project. The sequence maximum values in the database can be so large that even the 64-bit  type in C# won't suffice, I have to use the 128-bit   type to read them from the database. Now I then do the actual linear regression calculation in  type. When I end up with the result, I have to convert it to a. I do this by first converting the  into a   and then using the   constructor. But is there any way to check if the  will not fit into a , or the   ticks value is too large to construct a   object, at which point the calculation would result in "estimated time is too far in the future to give a specific answer"? I could use ...   in case an exception occurs, but in my opinion it's an ugly style of coding. J I P &#124; Talk 19:56, 15 January 2015 (UTC)


 * I don't think there's anything ugly about using error handling in that way. StuRat (talk) 20:04, 15 January 2015 (UTC)


 * Compare your  to   and   -- Finlay McWalterᚠTalk 20:47, 15 January 2015 (UTC)


 * If you mean the DateTime structure in .net 4.5, try looking at http://msdn.microsoft.com/en-us/library/vstudio/system.datetime.maxvalue%28v=vs.110%29.aspx which gives the maximum allowed value, which is 100 ns before 00:00:00, January 1, 10000. I have read that the DateTimeOffset structure is a more capable replacement for DateTime, but I have not tried it myself. Jc3s5h (talk) 21:04, 15 January 2015 (UTC)
 * Use the TryConvert operator, or wrap it in a try-catch block? LongHairedFop (talk) 10:45, 16 January 2015 (UTC)