Jump to content

Wikipedia:Reference desk/Archives/Computing/2009 September 19

From Wikipedia, the free encyclopedia
Computing desk
< September 18 << Aug | September | Oct >> September 20 >
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.


September 19

[edit]

Hacking a registry that isn't running

[edit]
Resolved

It's a long story, but the short version is that I have a hard-drive with a full, legal installation of Windows XP. The computer it once was in has died. The computer case which once had the registration key is long since gone to the dump. Yeah, dumb. There are plenty of free programs out there that will hack your registry and provide you with all kinds of serial numbers of stuff you've installed, but they all search the active registry, not the one sitting inert on a harddrive. Do any programs exist that will do this? And before you ask, yes, I've tried simply installing the drive and booting from it, but the computer configuration is just too different and I get stuck in a boot loop. Any help? I'm waiting to hear back from the company I bought the computer from to see if they've kept a record (Microsoft suggests I do that before contacting them), but I'd like to get back-up plans in place, if I can. Matt Deres (talk) 01:37, 19 September 2009 (UTC)[reply]

You might be able to do it by importing the hives. Ie on the machine that is running, import the file from the drive of the machine that won't boot, edit, then export. I have used File, Load Hive to edit another user's registry (on the same PC) before, but that only appears to allow loading of the user hives, not system hives. However File, Import might help - but I've never tried it. Obviously, this has the potential to clobber the registry on the running PC, so try it at your own risk, after taking appropriate precautions. Mitch Ames (talk) 03:38, 19 September 2009 (UTC)[reply]
Yikes. That's just friggin' scary. Besides the basic risks involved, the only completely working version of Windows I have now is Vista anyway; I don't fancy trying to import the XP registry into Vista. :) Matt Deres (talk) 11:53, 19 September 2009 (UTC)[reply]
You don't import the hive, you just mount it on an unused key. It's like mounting a filesystem on an empty subdirectory in Unix. You don't need to do this, though—use MJB Keyfinder instead (see below). -- BenRG (talk) 23:03, 19 September 2009 (UTC)[reply]
I don't know what you need to do with the registry, but for plain editing you might try the Offline NT Password & Registry Editor. --164.67.235.138 (talk) 04:09, 19 September 2009 (UTC)[reply]
That site is for recovering passwords; I need to find the 25-digit security key for XP to install. I need something that will provide me with that key; there are plenty of programs that can hack the registry and do that for a running copy of Windows; I just want one that can extract it from a dormant hard-drive. Matt Deres (talk) 11:50, 19 September 2009 (UTC)[reply]
You might consider talking to Microsoft on the phone before going to this much trouble, unless you have tried already. If you have any documentation at all, pull it together and find out what they say. At worst, they might provide a new set of disks for some amount of money, hopefully less than the full cost of XP. New copies of XP that appear legit can be bought on Ebay for as little as $60. EdJohnston (talk) 15:49, 19 September 2009 (UTC)[reply]
That's my escape plan, if necessary, but online guides I've read have suggested that OEM versions of Windows (even ones like mine that indeed are simply Windows and not Windows + whatever crap HP or whoever feels like squeezing onto a "system restore" disc) are not as easy to get keys for. According to the About.com page, for example, they say getting a regular Windows key costs $10 + time spent on the help line (which is not free), but that OEMs are just more complicated than that. It remains an option. Matt Deres (talk) 16:07, 19 September 2009 (UTC)[reply]
Magical Jelly Bean Keyfinder works with hive files from unbootable Windows installations. It's free and open source. -- BenRG (talk) 23:03, 19 September 2009 (UTC)[reply]
You are my GOD! Thank you very much - that did the trick. For the record, you do need to be either logged in as the administrator or "run as administrator" to get the hive to load. Unhelpfully, the program doesn't tell you what the problem is, it just says that the load didn't work. Many thanks again! Matt Deres (talk) 21:09, 20 September 2009 (UTC)[reply]

BIOS password

[edit]

A long time back (about three years or so), when I didn’t know anything about BIOS and stuff, I set the BIOS password not knowing what I was doing. Now, I’ve forgotten the password, but need to enter BIOS to change a few settings. Is there any way in which I can recover or reset my password? Please help! 117.194.232.75 (talk) 09:18, 19 September 2009 (UTC)[reply]

You can reset your BIOS settings by unplugging it from mains and taking out the battery on the motherboard for 30 seconds. --antilivedT | C | G 11:21, 19 September 2009 (UTC)[reply]
I also believed that worked a few months ago, but now I know that you sometimes need to remove the battery for several minutes (try 15). --Andreas Rejbrand (talk) 12:12, 19 September 2009 (UTC)[reply]
Some motherboards have a jumper that resets the password, without taking out the battery or resetting CMOS. --164.67.235.138 (talk) 19:17, 19 September 2009 (UTC)[reply]
Yea most motherboard have that, but it's a lot easier to tell a user to remove a big shiny silver thing than to fiddle with jumpers. --antilivedT | C | G 03:56, 20 September 2009 (UTC)[reply]
Please don't forget that "reset your BIOS" means it will revert to its factory-default settings, which may or may not be the ones you previously had. If you perform this reset, you may have to go into the BIOS and change the settings so that your computer works correctly. — QuantumEleven 14:15, 22 September 2009 (UTC)[reply]

No. of notebooks in Pakistan (How Many?)

[edit]

How many people in Pakistan have and/or using notebooks and laptops? What is the market size and growth potential?116.0.61.115 (talk) 09:37, 19 September 2009 (UTC)[reply]

I can't find any data for pakistan, though there are some numbers available for india (not complete). Would assuming the same ownership per capita be reasonable - you could derive a very rough figure?83.100.251.196 (talk) 12:47, 19 September 2009 (UTC)[reply]

Tiled images

[edit]

Hi! I am looking for help with these question (especially the first). bamse (talk) 11:54, 19 September 2009 (UTC)[reply]

In that case, the images appear to be in a simple x,y format (e.g. [1] = x02, y14). The minimum (top left) is 1,1; the maximum is 28,19, at the maximum zoom, 4). You'd need to use a tool that let you specify the maximum x or y images and download them to a big folder (basically any language can do this pretty easily). Then, more complicatedly, you'd need something that could stitch those together based on their coordinates (ImageMagick+whatever language you originally used, is probably best for this). I'm not sure there's a one-stop-shop program that will do all of those automatically. --98.217.14.211 (talk) 13:30, 19 September 2009 (UTC)[reply]
Thanks for the reply. I was mainly interested in the first part, getting the tiles on my computer. Is there a tool that can do that, or would I have to code something myself? Stitching is not a problem and it seems to work. bamse (talk) 14:11, 19 September 2009 (UTC)[reply]
I gave into the temptation and wrote a PHP script - see my reply at the Graphic Lab question. (Never sure of cross-posting etiquette, but it seems to be the more specific channel) - IMSoP (talk) 14:59, 19 September 2009 (UTC)[reply]

Pingler

[edit]

What is Pingler for? Apparently it's got something to do with blogs but the site itself doesn't explain much? Help appreciated. Cheers, Jeffrey Mall (talkcontribs) - 14:44, 19 September 2009 (UTC)[reply]

It tells search engines that your blog has been updated, so then any new content you added is indexed faster. Xenon54 / talk / 14:48, 19 September 2009 (UTC)[reply]
Ahh thank you! Jeffrey Mall (talkcontribs) - 14:49, 19 September 2009 (UTC)[reply]
Actually I don't think it "tells" search engines anything. I believe they simply perform ping, which is a network diagnostic tool (which you can run for yourself, for free). I'd call this a "scam site"; it looks like they want to charge you for a service which has no impact on your search-engine ranking. Nimur (talk) 16:19, 19 September 2009 (UTC)[reply]
(Twitter agrees, and has suspended their account for suspicious activity... sounds like spamming! Nimur (talk) 16:20, 19 September 2009 (UTC)[reply]
Hmm.. Jeffrey Mall (talkcontribs) - 17:06, 19 September 2009 (UTC)[reply]

Algorithmn to space cars along road segments?

[edit]

I'm trying to write an algorithmn that will space static cars (just points or dots really) evenly along a road. The road consists of a number of straight-line road-segments that join up to make the road. For simplicity I am first of all trying to write an algorithmn that can do this for a one-dimensional road, and then hopefully modifying it to deal with a more realistic two-dimensional road. Even when the road goes around a bend, I want the cars spaced at every X metres along the length of the road. I'm stuck trying to create an algorithmn that can cope with all circumstances: for example where the road-segment lengths are less than the distance between cars.

Is there any forum where I could get help creating such as algorithmn please? 89.243.181.104 (talk) 16:28, 19 September 2009 (UTC)[reply]

You can ask here - as far as I can tell this is easy:
1. Sum the length of the road segments to get the road length.
2. Divide the total road length by the number of cars to get the spacing between cars.
3. Starting at the first road segment add the road segments until the length is greater than the car spacing.
4. At this point place a car at the point on that segment where the distance from the previous car is correct. Then start counting the distance again from that point. Repeat the process from this point until you run out of cars.
Did that makes sense? Ask about any part you get stuck on.83.100.251.196 (talk) 17:07, 19 September 2009 (UTC)[reply]

Good, thanks, I had begun thinking along those lines myself, but what about when the segment is much longer than the car-gap? Or where you have segments of various lengths which may be much longer or much shorter than the gap? I'm trying to think what variables you would need to keep and update. 78.144.252.66 (talk) 19:01, 19 September 2009 (UTC)[reply]

You need to count the number of segments that have already been used, and also the position along the segment the last point was placed.
A general proceedure would be to first check if the point can fit into the current segment, and if so do so, and update the most recent point
and if not then to increment the segment number (remembering compute the increment or decrement of gap length by the correct ammount).
It might be easier to show some pseudocode - here one that finds the segment, and possition along the segment of the Qth point. If you use this as a proceedure you could just call it for each position - it's not the most efficient way - but should always work.
Note that l[n] is the length of the nth segment - an array - so here you can vary the segment length if you want.
n=number of segments
p=number of points
l[n]=length of segment n
Lengthtotal=sum (1 to n) of l[n] self explanatory
pointseparation = Lengthtotal/p
computer Qth point distance
distancetopointnumberQ=Q*pointseparation

x=1    x counts the number of segments
while ( l[x]<distancetopointnumberQ ) do
  {x=x+1 
   distancetopointnumberQ=distancetopointnumberQ-l[x] }
  
Output "Qth point on segment ",x," distance along that segment=",distancetopointnumberQ 
stop or return 

Note I've let the points go right to the end of the line, and there is no point at the beginning (technically this would be the 0th point) - to change this behaviour minor tweeks are needed

Hopefully I haven't made some blindingly obvious algorhtylmic error...83.100.251.196 (talk) 20:55, 19 September 2009 (UTC)[reply]

Thanks, I've studied the above. I think - could be wrong - that it would only work where the road-segment-length is much less than the distance between the cars. It does not have any mechanism for placing several cars within a long length of road-segment. Having such a mechanism would also entail remembering how much of the road segment was left over since the last car, and taking that into account when positioning the next car (which might be in the next road-segment, or could be several road-segments further on). 78.149.186.66 (talk) 19:39, 20 September 2009 (UTC)[reply]

The above method places one car at the Qth point, thus to place all the cars you would need to run it in a for/next loop with Q going from 1 to p (number of points or cars)
It's possible to modify it so that it finds each point in turn more efficiently, but as it stands it should work to place all the cars as long as you include the for/next loop.
I'd suggest trying it, excluding any typos it works for all combinations of road segment length, and inter-car-distance both ways. If it doesn't work you could post your code, if it does work it's a step to a better version.
NOTE it only does one car at a time - I think that is the source of your confusion.83.100.251.196 (talk) 22:10, 20 September 2009 (UTC)[reply]

Thanks, with your help I think I've got how to do it. I have a while-wend that moves along the road-segments (as you showed), and a nested while-wend inside that that moves within the segment if required. The main variables are the current position as I jump from car to car, and the starting position and end position of the current road-segment. Your approach of measuring positions from the start of the road helped a lot. Thanks again. 78.144.198.168 (talk) 13:29, 21 September 2009 (UTC)[reply]

Good. Note that my example always starts "counting" from the beginning of the road (road section 1, offset 0) - it would be quite simple to modify so that it starts counting from another position (alter the initial value of x, and add a new variable called offset) eg
while ( l[x]-offset<distancetopointnumberQ ) do
   {x=x+1 
   distancetopointnumberQ=distancetopointnumberQ-l[x] 
   offset=0}

That could be used as a proceedure or subroutine that starts with (road section x, offset) as inputs and returns the same sort of values as (newsection = x, newoffsetposition = distancetopointnumberQ) when the "while wend" loop finishes. Same answer but prevents beginning the count from zero if you are calculating the positions sequentially..83.100.251.196 (talk) 16:30, 21 September 2009 (UTC)[reply]

Facebook moble texts

[edit]

i just added someone as a friend, and asked facebook for the sms mobile text sevice with them, so why isnt it sending me mobile texts about their updates?Accdude92 (talk) (sign) 17:09, 19 September 2009 (UTC)[reply]

There are several answers to this question...you need to give us more details. Does it work for other people? Is this the first person for which you have subscribed to SMS updates? Have you set the notifications up correctly (correct carrier, phone number, etc.)? Xenon54 / talk / 18:05, 19 September 2009 (UTC)[reply]
Does it work for other people?
Yes
Is this the first person for which you have subscribed to SMS updates?
No
Have you set the notifications up correctly (correct carrier, phone number, etc.)?
Yes

Accdude92 (talk) (sign) 18:25, 19 September 2009 (UTC)[reply]

So it is evident that the problem is only with this one particular person. Perhaps the addition of the person to your notification list "didn't take" yet (so wait for another status update, then report back here if the problem continues). If what you're telling us is correct, then there is probably nothing wrong on your end. Xenon54 / talk / 18:58, 19 September 2009 (UTC)[reply]
It's more likely that Facebook is having trouble, I have this too but I usually get no texts for a while and then they all catch up, sometimes up to 12 hours later. I think it's down to the fact it's a new service and probably still under some kind of construction to help cope with the sheer amount of messages being delivered. --195.49.180.146 (talk) 13:27, 21 September 2009 (UTC)[reply]

Ship of Theseus

[edit]

Is there any consensus on how to determine the age of a desktop computer in which, as with the Ship of Theseus, every component has been replaced on a separate occasion? NeonMerlin[2] 19:15, 19 September 2009 (UTC)[reply]

No, just as there isn't with the Ship of Theseus. It doesn't matter whether its a desktop computer an automobile, it's a philosophical question, not a technical one. --98.217.14.211 (talk) 21:53, 19 September 2009 (UTC)[reply]
One clue is to look for the oldest files on the system. Though on my desk top computer, the files go back to 1980, they are more of a reincarnation from an old Apple II pascal disk I had. Graeme Bartlett (talk) 22:11, 19 September 2009 (UTC)[reply]
Right, file creation dates have nothing to do with the age of the machine they are on at all, unless you can prove they were created by that individual machine. Even then, it doesn't really solve the Theseus problem, unless you are just interested in, "when was the oldest hardware component of this machine put together," but that's not the Theseus problem (which is a question of identity, not a simply question of forensics). --98.217.14.211 (talk) 03:20, 20 September 2009 (UTC)[reply]

"Beep" command makes my computer case beep!

[edit]

I've got Windows XP and I've downloaded the GWBasic interpreter onto my desktop, where it runs in a DOS window very well. When I put my headphones on and typed the beep command, the beep noise came from the case of my tower PC, not the headphones. What is it in the case that is beeping? Could I use it for other sounds too? 78.144.252.66 (talk) 19:21, 19 September 2009 (UTC)[reply]

Have a look at PC speaker. Old DOS PC games (e.g. Doom) usually had an option to play the sound effects/music through it in the setup program. It sounded, as you might imagine, all bleepy. --Kurt Shaped Box (talk) 19:26, 19 September 2009 (UTC)[reply]
If you use DOSBox, you can redirect the beeps to your headphones. NeonMerlin[3] 19:40, 19 September 2009 (UTC)[reply]
Oh my gawrsh, this question makes me feel old. :-( --98.217.14.211 (talk) 21:57, 19 September 2009 (UTC)[reply]
Any Windows programmer can use the computer's speaker to produce a sine tone. The function beep in Windows API (kernel32.dll) can be used for this:
beep(400, 500)
will use the computer's speaker to produce a 400 Hz sine tone for 500 milliseconds (that is, 1/2 second). --Andreas Rejbrand (talk) 12:41, 20 September 2009 (UTC)[reply]