Wikipedia:Reference desk/Archives/Computing/2011 January 28

Computing desk
< January 27 << Dec | January | Feb >> January 29 >
Welcome to the Wikipedia Computing 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.


January 28

edit

Slowness

edit

I use Roadrunner, which bills itself as the 'fastest ISP for the lowest price'. However I find that whenever I am just starting up the computer, the Internet is very slow in loading anything. After about an hour, it gets fast. Why is this? If it helps, I run my modem through a router to get to my computer so I can connect multiple computers, but I find this happens even when I have the modem directly connected to the computer. 24.92.70.160 (talk) 02:24, 28 January 2011 (UTC)[reply]

It could be a few things, many ISP's bill themselves as the "fastest for the lowest price" but there is no way to prove this. This could have nothing to do with your internet, rather your computer itself, or even your modem. It could also have something to do with your web browser -- do you use IE or FireFox. Or Safari, or Opera. Tofutwitch11 (TALK) 02:30, 28 January 2011 (UTC)[reply]
Have you tried quantitatively testing your Internet speed to see whether it really does get faster after one hour? If not, try this speed test: http://www.speedtest.net/index.php?nojs=1
Also, are you sure that no other applications are using the Internet during that first hour? Close all other Internet applications, don't let Windows or Java update anything, and retest. If the Internet is still slow, you can try Wireshark (http://www.wireshark.org/) to see what traffic is going through your computer. --99.237.234.245 (talk) 03:01, 28 January 2011 (UTC)[reply]

I suspect observer bias, and recommend the quantitative methods suggested above. If you can establish that an actual change of speed does occur, two probable reasons exist.

  1. First, your network connection may be cached (or even pre-fetched) at various stages of the connection between you and the network service provider - including your browser, your operating system, your cable modem, or any of the routers or servers in the network path. If you're loading the same page, as opposed to loading new and completely random pages that, do you notice a speed difference? (To test this feature particularly, you need to avoid prefetch logic in any cache scheme - so you need to navigate to a page that isn't linked in any way from any pages you've already loaded).
  2. The next possibility is that during your hour of "boot up", traffic patterns are statistically changing. At some level, your network connection is shared. So, maybe you sign on at 4 PM, when more people are logged in at work and using the shared network connection heavily; and by 5PM your shared network connection is less utilized. Compounded by the fact that your network service provider may use traffic shaping and network provisioning to vary the actual, deliverable performance, based on measured or estimated network demand, this can get pretty hairy to measure. So to quantify your speed properly, you'll really need to measure a variety of network-speed-vs.-time-of-day profiles; and you should probably average them over a month or two.

Finally, I'll just point out the difference between latency and throughput - two different metrics of "speed" that you should always keep in mind when you're quantitatively measuring network performance. By properly training internet-users in the conceptual basics of network performance, we can end the infamous "my network is slow" complaint and replace it with an actual, diagnosable, (and possibly fixable) report of true network performance. Nimur (talk) 18:29, 28 January 2011 (UTC)[reply]

Hard disk failure

edit

Five years ago, there was a power outage, and a very low-quality desktop computer I had mysteriously stopped working. If I can trust the memory of my former 9-year-old self, the computer simply didn't react when I pressed the power button. The BIOS screen didn't appear, no LEDs turned on, and the monitor never received any signal. Because there were quite a few precious childhood photos on there, I disassembled the computer and attached its hard disk (it was a PATA disk, if that matters) to a working desktop as master, hoping it would boot. No such luck.

A few weeks ago, I bought a IDE-to-USB adapter and tried to access the hard disk. It didn't even spin up or react in any way, so obviously that plan failed. I have two questions: (1) why would a power outage cause such serious problems? (2) Anything else I could do to recover my data? --99.237.234.245 (talk) 02:49, 28 January 2011 (UTC)[reply]

If it won't spin up on it's own, you're going to need the help of a Data Recovery Service. They can retrieve data from disks that aren't functioning.
A google search turns up lots of them. It wouldn't hurt to contact them and get some quotes. APL (talk) 03:51, 28 January 2011 (UTC)[reply]
Before you spend a lot of money, I assume that you received an independent power supply with your adapter. Is this working properly and firmly connected to your hard drive motor? It seems unlikely that the power outage destroyed the motor of your hard drive, though I suppose anything is possible if there was a large power surge at the same time. Dbfirs 09:39, 28 January 2011 (UTC)[reply]
Yes make sure you plugged in both the IDE and power connectors into the hard drive 82.44.55.25 (talk) 10:41, 28 January 2011 (UTC)[reply]
Sometimes, drives get stuck if they haven't been used for a while (several years). If you're sure you don't want to go for a professional data recovery service, you could try the following procedure - which may make things worse, may decrease the chance that a professional data recovery service will still be able to pull data off the drive, should you later change your mind, and may increase the price such a service will charge you.
Hook the drive up to your USB adapter or a long IDE cable, and an appropriate power supply. As soon as you connect power to the drive, turn it - more or less gently- around the disk axis.
I've seen disks spring to life after that procedure, though it usually meant they didn't live very long - so start backing up the data as soon as possible. -- 78.43.71.225 (talk) 16:41, 28 January 2011 (UTC)[reply]
I too have briefly revived a disk with this procedure, but I can't say I'd recommend it for an archive of irreplaceable photographs or for a drive that stopped suddenly and mysteriously.
The sort of "Sticky drive" this fixes tends to be more of a gradual failure where first the drive starts making horrible noises, then it occasionally doesn't start, then it doesn't start more often than not, etc. APL (talk) 17:00, 28 January 2011 (UTC)[reply]
What is the model of the drive? Early models suffered from stiction and could be temporarily fixed by applying percussive maintenance. ---— Gadget850 (Ed) talk 16:58, 28 January 2011 (UTC)[reply]
I've heard of data recovery companies, but before mailing my disk in for a quote, can anybody give me an approximate price range? I've read that recovery off of a physically damaged disk costs >$1000. I'm 13 years old, and even if I had that kind of money, I wouldn't spend it on something that only has sentimental value.
The disk in question is a Quantum Fireball (Plus LM), 3.5 series. I've tried 78.43.71.225's suggestion, as well as gentle "percussive maintenance", but neither of them had any effect. I'm also sure that the power adapter is plugged in. The adapter is also a SATA-to-USB adapter, and it works perfectly on my SATA disk, so that's not an issue. Unfortunately, I don't have a working PATA disk to test it on. --99.237.234.245 (talk) 17:49, 28 January 2011 (UTC)[reply]
If it is SATA, then it won't suffer from stiction. ---— Gadget850 (Ed) talk 11:49, 29 January 2011 (UTC)[reply]
One office friend did spend US$1000 for one of those places to recover the data from a hard disk. (The service did get all the data in his case.) You could always wait 15 years until you're loaded with cash and try then. Comet Tuttle (talk) 17:58, 28 January 2011 (UTC)[reply]
Yes, disk recovery can be expensive but prices do vary depending on what service you want. Google for "disk recovery" or similar keywords, and get a few quotes - you don't have to follow up any of the quotes. Astronaut (talk) 11:37, 29 January 2011 (UTC)[reply]
Question: Does the drive show up, throwing errors, or does it not show up at all? If the latter: With older hard drives, it is sometimes possible to salvage the electronics from a drive of the same make and model, so in case the logic board got damaged, yet the mechanical parts are still ok, you might be able to recover the data yourself by buying the same model on ebay and swapping the board.
If it does show up, then you probably managed to fry the drive motor, and swapping the board won't help.
I've read reports of people building their own makeshift cleanroom in the bathroom for "open-disk surgery", but that's a rather risky approach (and would still require you to find a donor disk first). -- 78.43.71.225 (talk) 21:27, 29 January 2011 (UTC)[reply]
It doesn't show up at all. The computer doesn't even recognize it as a USB device. I'll try buying a new control board, although I don't know how easy that is considering the age of this disk. --99.237.234.245 (talk) 05:13, 30 January 2011 (UTC)[reply]
Before you go too far, you might try it with some other, known-good, disk drive if you have one. There may be a problem with your USB to ATA converter. APL (talk) 02:38, 31 January 2011 (UTC)[reply]

wlan printer

edit

For connecting a wlan printer, I connected both the printer and the computer to a wlan_xy. However, is there a way of connecting the printer directly to the computer (which is connected to a wlan)? Quest09 (talk) 10:51, 28 January 2011 (UTC)[reply]

Does the printer have a USB port? Many networked printers do, and if it does, you can connect it by a USB cable to your computer and use it just like a non-networked printer. Or are you trying to do something more complicated? --Colapeninsula (talk) 14:15, 28 January 2011 (UTC)[reply]
yes, I know I can connect to it with a cable. The problem is that I use the wlan of my neighbor, and if he disconnects it, I wouldn't have the printer connected to the computer via wlan. It seems that they suppose that you have to have an access point for a wlan connection. Quest09 (talk) 17:19, 28 January 2011 (UTC)[reply]
Yes, that is the way the printer and the computer work: you need a wireless access point to establish a wireless network. The USB cable is your best solution if your neighbor switches off his router. Comet Tuttle (talk) 19:01, 28 January 2011 (UTC)[reply]
Wireless access point is not mandatory, there is ad hoc mode. All wireless cards, i have encountered support it (though, some specific devices might not). In ad hoc mode, there is no need for access point, because devices connects to each other directly. Maximum achievable data rate might be lower than in infrastructure network (because of less efficient protocol), there also in no WPA available, only WEP. Setting it up also might be more complicated. Printer probably will be able to connect to existing network. So, you will need to make ad hoc network on your computer and connect printer to it. There probably also will be necessary a DHCP server (if that thing will not work with zeroconf). -Yyy (talk) 08:02, 29 January 2011 (UTC)[reply]

Calendar of worldwide events/days/festivals/occasions etc.

edit

Hello all, I think this query just about falls under the remit of the Computing Ref. Desk. I've spent the last few days looking for an online calendar that charts a range of events including:

  • Inter/National days of celebration (Independence days etc.)
  • Inter/National and regional events/festivals etc
  • Awareness weeks ("Green Week", "Mental Health week" etc.)

And any other notable events on an International, National and regional level.

I've found some reasonably useful resources such as the Earth Calendar, 2011 and http://en.wikipedia.org/wiki/Category:Annual_events, but nothing that really works as a suitable "catch-all".

Does anyone on here know of a resource that might fit the bill? Thanks, 161.76.127.89 (talk) 11:43, 28 January 2011 (UTC)[reply]

It's not free, but it appears that this is what you are looking for: MultiCultural Calendar --Zerozal (talk) 14:13, 28 January 2011 (UTC)[reply]
Hi Zerozal, thanks for the link, it oculd be quite useful. 161.76.127.89 (talk) 15:21, 28 January 2011 (UTC)[reply]
Go to Google and login (assuming you have made your free Google account). Select "Calendars" under the more option. In Calendars, you will see your calendar on the left and a settings option. Click on settings. On the settings page is the option to "View interesting calendars". Click on that. You will see a list of holiday calendars for pretty much every country and religion. -- kainaw 19:27, 28 January 2011 (UTC)[reply]

Storage capacity of 480i SD video on Blu-Ray

edit

Considering the maximum 10Mb/sec of a SD DVD, how much 480i video could a single layer 25GB Blu-Ray disc contain? Google calculator and Wolfram Alpha gave me about 5 ½ hours (25GB ÷ 10Mb/sec), but that seems a little low. --70.167.58.6 (talk) 15:41, 28 January 2011 (UTC)[reply]

My math agrees. You probably already know this, but our article section DVD-Video#Data rate says professionally encoded DVDs run about half that maximum data rate, so you'd double your runtime in that case. Comet Tuttle (talk) 17:57, 28 January 2011 (UTC)[reply]
Since the question seemed to concern 480i video rather then specifically MPEG2 encodec 480i video, you could also consider using H.264 and an even lower average data rate achieving roughly the same visual quality. Nil Einne (talk) 19:27, 28 January 2011 (UTC)[reply]

Unprintable symbol in firefox (iceweasel)

edit

When I am looking at math formulas using iceweasel ( Debian version of firefox) I get the following problem.

U preserves the inner product 〈 , 〉 on the Hilbert space, i.e. for all vectors x and y in the Hilbert space,

I assume that the symbols after "U preserves the inner product" are also unreadable for you. I have the browser set to UTF-8 and Ariel type face. I have had the same problem using other type faces. What is the problem and how do I fix it.

Gary Roach —Preceding unsigned comment added by 96.229.235.83 (talk) 19:58, 28 January 2011 (UTC)[reply]

(They show up fine for me in Firefox on a Mac, assuming they are supposed to be somewhat stretched less-than and greater-than signs.) --Mr.98 (talk) 20:04, 28 January 2011 (UTC)[reply]
It is possible that the questioner thinks that there should be something inside the angle-brackets like <x,y>. Since those brackets were first added, nothing has been inside them. -- kainaw 20:08, 28 January 2011 (UTC)[reply]
Except a comma. --Mr.98 (talk) 21:28, 28 January 2011 (UTC)[reply]

Any particular website? ¦ Reisio (talk) 22:07, 28 January 2011 (UTC)[reply]

I should have put in a description. What shows on my browser are two square boxes with numbers (I think) 3008 , 3004. These seem to correspond to #160 in the html code. I am talking about wikpedia files here. —Preceding unsigned comment added by 96.229.235.83 (talk) 00:01, 29 January 2011 (UTC)[reply]

3008 is the left angle bracket 〈, 3009 the right 〉. 160 is equivalent to a particular space character. If you can refer us to the particular files, more can be confirmed. ¦ Reisio (talk) 00:17, 29 January 2011 (UTC)[reply]

Yes a search for Unitary Operator will get you to the page in question. Specifically, http://en.wikipedia.org/wiki/Unitary_operator

Gary R. (I guess I should log in but seldom have anything to contribute. Also retired and flat broke. Great economy) —Preceding unsigned comment added by 96.229.235.83 (talk) 00:44, 29 January 2011 (UTC)[reply]

I don't think you're missing any characters, the editors of that article have just decided to refer to certain syntax without any values inside. ¦ Reisio (talk) 00:51, 29 January 2011 (UTC)[reply]

What you're seeing is almost certainly a browser bug -- it doesn't know how to render those symbol codes. How old is your version of iceweasel? Looie496 (talk) 18:17, 29 January 2011 (UTC)[reply]

Phones as modems

edit

Hi, I've always "known" and have been told by everyone that using a Phone as a modem for my computer extensively, shortens the lifespan of the phone, but I cant find anything to back this up, does anyone have any documentation to support this? Joneleth (talk) 20:13, 28 January 2011 (UTC)[reply]

In what way are you using a phone as a modem? When I first began using modems (shortly after the fire was invented), a modem required you to place the handset on the modem. Shortly afterward, the modem became a self-contained unit. You plugged in a phone line. So, you didn't use a phone anymore - just a modem. Regardless, usage of the phone in the very old modems is no different than speaking on a phone. If anything, it is better for the phone as risk of being dropped decreases when the phone is just sitting on top of a modem. -- kainaw 20:18, 28 January 2011 (UTC)[reply]
Perhaps you mean tethering a mobile wireless digital telephone to a personal computer? I think that shortens the battery life because the radio transmitter is on for a long period of time. It should not damage the phone in any way. Nimur (talk) 21:40, 28 January 2011 (UTC)[reply]
He could have been talking about whistling modem tones into a phone. (What, no article? Shocking.) 67.162.90.113 (talk) 22:57, 28 January 2011 (UTC)[reply]
(The article you want is Phreaking.) --Mr.98 (talk) 00:20, 30 January 2011 (UTC)[reply]
Oh, I don't know; I used to be able to whistle 300 baud originate and answer tones, and it would be a stretch to call that phreaking. Phreaking has the connotation of manipulating the phone company's system, rather than just fooling the modem on the other end of the line into not giving up and dropping the connection for a few seconds, for want of a carrier signal. Comet Tuttle (talk) 18:35, 31 January 2011 (UTC)[reply]

Im taking about Tethering with Cell phones yes, and i know why they are suppose to do it but i need some kind of documentation for it! Joneleth (talk) 22:01, 28 January 2011 (UTC)[reply]

Integrated circuits do eventually fail after many years of use, but they usually get replaced long before that point. Batteries have to be replaced after a relatively small number of charge-discharge cycles, of course. Overuse may shorten the lifespan of your limited-data service plan. Other than that, I don't know of any negative effects. -- BenRG (talk) 04:10, 30 January 2011 (UTC)[reply]