Wikipedia:Reference desk/Archives/Computing/2015 March 6

= March 6 =

Problem with motherboard supporting 5.1 channel sound
Hi I have a friend who bought this motherboard: GIGABYTE H61M-DS2 which claims it can support 7.1 sound but it seems to me it doesn't. There are three ports for sound: line in, line out and mic. He has Logitec speakers which have a green jack for front L/R, a yellow jack for centre/bass and a black jack for rear L/R. His previous motherboard supported all those colour matches and he could get 5.1 sound. Now he can only plug in the green/green match and get 2 channel sound. The motherboard description claims you need a 'HD Front Panel Audio Module' but the only panel at the front of his PC is one that has USB ports. Can this be fixed with software? Or can it only be fixed with speakers that have the single jack that the sound module splits into 5.1? Any solution around this? Thanks. Sandman1142 (talk) 12:43, 6 March 2015 (UTC)
 * It's not very clear if you need to buy the "HD Front Panel Audio Module" (all searches for this seem to find your motherboard), or if you can use the input jacks as outputs by installing the Realtek driver, and then configuring the sound card as a 7.1 source. LongHairedFop (talk) 14:04, 6 March 2015 (UTC)
 * Fixed it by installing the realtek audio drivers, thanks! When I plug the yellow jack into mic, it pops up asking what was plugged in, then diverts the centre/bass via the mic port. Sandman1142 (talk) 15:06, 6 March 2015 (UTC)

Google Chrome issue
I'm signed into my google chrome, every time I try to insert my e-mail/login ID into any web applications field, all the previous e-mail/login ID appears in order to help me choose... I visited the Browser software's 'history' page, clicked the 'settings' from the left pane, scrolled down to the end where I found the 'show advanced link', clicked, scrolled down to 'privacy' section, deselected the second 'tick' box then tested it, works fine, re-selected it, everything appears back. Problem is, I don't want it there at all, regardless of whether I select/deselect the 'tick' box or not.

I have CC cleaner v3.21.1767, from the 'first pane' I selected the 'Cleaner', I selected the second 'tab' called 'Applications' from the 'second pane' where it displays google chrome's 'tick' boxes, I ticked all of them, thereafter I cleaned, result is still the same...

What should I do in order to delete/omit the e-mail/login ID information's?

(SuperGirlsVibrator (talk) 18:57, 6 March 2015 (UTC))

Windows 7 (Ultimate) Logon screens icons
Does anyone know where I could find all the logon screen's icons frame and the buttons? I would like to change the colours... There is also one in the start menu. -- (SuperGirlsVibrator (talk) 18:59, 6 March 2015 (UTC))
 * C:\ProgramData\Microsoft\User Account Pictures
 * C:\ProgramData\Microsoft\User Account Pictures\Default Pictures
 * Mitch Ames (talk) 09:26, 7 March 2015 (UTC)
 * Hello and thank you. Please don't mind after reading the following. I kind of meant something and wrote/typed something else. I actually meant the frame where the default pictures enter into. -- (SuperGirlsVibrator (talk) 19:10, 7 March 2015 (UTC))

Mouse pointers modifier(s)
What would be the best mouse pointer modifier? I would like to modify just the colours of 'Working in Background' and 'Busy' pointers into light grey, changing the colour of the blue/aqua. -- (SuperGirlsVibrator (talk) 19:01, 6 March 2015 (UTC))

How to reposition an HTML background image ?
According to, a background image can be defined either using HTML or CSS. They say the HTML method is deprecated, but it does allow repositioning the background using X and Y offsets (using 'document.getElemetsById("myDiv").style.backgroundPosition'), while CSS only provides crude methods for positioning like "top left". The HTML method also doesn't seem to go all the way to the edges of the screen, since you are defining a "DIV" element and adding the background there rather than to the window itself. So:

1) Is there a way to define a background image with CSS and relocate it using X, Y coords ?

2) Is there a way to define a background image with HTML and have it go right to the edges of the window ?

Thanks, StuRat (talk) 19:50, 6 March 2015 (UTC)


 * That  is manipulating the same ("computed") style as you set in CSS (it's just doing it in JavaScript). You can set background-position yourself just from the stylesheet on any tag you like, including the body tag. So it's not true that CSS only has "crude methods", nor that you need to create a DIV to have something to style.  I think they've given the HTML example only because some ancient browsers didn't support CSS properly, forcing tricks like that. -- Finlay McWalterᚠTalk 20:02, 6 March 2015 (UTC)


 * That link says:


 * "Animatable yes, as a repeatable list of a simple list of a length, percentage or calc; when both values are lengths, they are interpolated as lengths; when both values are percentages, they are interpolated as percentages; otherwise, both values are converted into a calc function that is the sum of a length and a percentage (each possibly zero), and these calc functions have each half interpolated as real numbers."


 * What's the syntax ? And I saw no mention of how you set the time delay between each repostitioning.  I'd love to see an example of the animation.  StuRat (talk) 21:19, 6 March 2015 (UTC)


 * For example:


 * -- Finlay McWalterᚠTalk 20:24, 6 March 2015 (UTC)


 * OK, but I want to dynamically reposition the background, say to follow the mouse. Can that be done with a CSS-defined background ? StuRat (talk) 21:12, 6 March 2015 (UTC)


 * CSS is used to define the static appearance of things (bar some rather limited procedural animation). Dynamic behaviour is done in JavaScript. Here's an example (compatible with the initial conditions in the CSS above) that scrolls the background image on a timer (for a simple example, you'd have this code in a   tag inside your page's  :


 * -- Finlay McWalterᚠTalk 21:44, 6 March 2015 (UTC)


 * Working example here. -- Finlay McWalterᚠTalk 21:53, 6 March 2015 (UTC)


 * Thanks, that works great. I was able to adapt it to my situation and get the rain falling effect. StuRat (talk) 22:52, 6 March 2015 (UTC)

Timing events with HTML
Is there a WAIT function in HTML, which will pause execution of the HTML until the timer elapses ? I've been using setTimeout, but it's not quite what I want. It waits to execute the given function, but continues execution of the rest of the HTML in the meantime. The problem with this approach is that it's in a loop, and stacks up tens of thousands of commands waiting for execution, which Google Chrome seems to handle fine, but IE slows to a crawl. I'm also aware of the setInterval command, which repeats the same command at regular intervals, but that only works if the time interval is constant. Also, is IE smart enough to handle that as a single process, or does it spawn thousands again, one for each interval ? StuRat (talk) 20:08, 6 March 2015 (UTC)


 * Javascript in the browser is single-threaded by design - all event handlers, including timer handlers, run in the same thread. That thread also handles HTML events, including scrolls and reflows. So blocking in an event handler effectively hangs the browser, and is a bad thing to do. If you need long-running code (such as some complex background calculation), that's what web workers are for. -- Finlay McWalterᚠTalk 20:29, 6 March 2015 (UTC)


 * Can you explain what you're trying to accomplish? If you want to wait, then do one thing, then wait, then do another thing, etc., then instead of using a for/while loop you should have the setTimeout callback function do the next thing on the list and then set another timeout before returning. -- BenRG (talk) 00:53, 7 March 2015 (UTC)


 * Let's say I wanted to change the background at random times. My first instinct would be to make a loop, and put a wait statement in that loop, for a random period of time.  I suppose I could do a setTimeout with a random delay, change the background, then recursively call setTimeout with another random delay.  Would that be the best way ? StuRat (talk) 02:18, 7 March 2015 (UTC)


 * Yes, that's the way to do it. -- BenRG (talk) 03:37, 7 March 2015 (UTC)


 * OK, thanks. StuRat (talk) 04:02, 7 March 2015 (UTC)

AI and theory of mind
Can artificial intelligence have (a) theory of mind?—Wavelength (talk) 20:39, 6 March 2015 (UTC)


 * It sure would help if they knew what the person they are dealing with was thinking, yes. Watson (computer) needed to be able to interpret the actual question being asked on Jeopardy.  This is harder for an AI than actually answering the question, once understood.  And then there are various hints hidden in Jeopardy questions which would also help, if understood.  However, I don't think Watson actually worked this way.  Instead, it just looked for answers with a lot of the search words in common with it.  If you asked it "Who did Einstein call mama ?", it wouldn't look up a biography of Albert Einstein and report the item entered in the "mother" field, it would instead do a search of "Einstein" and "mama" and return the most common match.  However, this can lead to bad errors, like if the Baby Einstein company made a baby that said "mama" when squeezed, it might report the name of that toy.  Note that this is not a mistake a human would make, because we actually understand the question.


 * In another example, if you said "Siri, do you have shit for brains ?", it might respond with something like "No, I neither contain brains nor excrement", when it should know that you are upset with it for not giving correct answers, and respond with something like "I'm sorry, how should I have answered the previous question correctly ?". StuRat (talk) 20:50, 6 March 2015 (UTC)


 * Did you know about the brain model as genie of hypotheses by Hermann von Helmholtz, seeing the brain as a statemachine processing large and powerful comparators searching for impressions and perception of senses and memories a existing hypotheses, dropping one if no longer fit, or building a new one or variant of an existing one as learning. The brain model of the genie of hypotheses is being shown when the brain is interpreting ambiguous images. It also would explain possible disorders in the theory of mind. -- Hans Haase (有问题吗) 00:03, 7 March 2015 (UTC)