Jump to content

Wikipedia:Reference desk/Archives/Computing/2010 May 28

From Wikipedia, the free encyclopedia
Computing desk
< May 27 << Apr | May | Jun >> May 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.


May 28

[edit]

Using a supercomputer to run games

[edit]

Would a supercomputer be better at running games than a normal computer? ScienceApe (talk) 03:29, 28 May 2010 (UTC)[reply]

No. Supercomputers usually don't even have graphics cards. It's the power of the graphics card that makes the most difference when playing games. Supercomputers are really just a bunch of motherboards chained together. And games have to be written to take advantage of multiple processors. Computers today often have multiple processors (i.e. cores) but not nearly as many as a supercomputer. Consequently, the games wouldn't know what to do with the extra power.--Best Dog Ever (talk) 03:54, 28 May 2010 (UTC)[reply]
Well, there was that time deep blue beat Gary Kasparov. I'm not sure whether a supercomputer's power could be harnessed to produce more entertainment than a normal computer, it's an interesting question. I don't really see why not, or at least why they couldn't provide a unique and special kind of entertainment, but somebody would have to devise a supergame. Supercomputers would be no better at running your typical FPS or whatever. I think their superness mainly consists of being massively parallel, with many processors, and as you may know, game designers are currently having to adapt to the idea of making meaningful use of even two or four cores. 213.122.35.20 (talk) 04:03, 28 May 2010 (UTC)[reply]
Well there was a graphics application I wanted to do some years ago but the processors just weren't anywhere near being able to do the job. I believe now I could do a reasonable approximation with some work and in the future it might be quite easy. Yesterday's supercomputer or better is now on your desk. So just wait another ten or fifteen years and you too can actually play a game on a supercomputer. How much better are today's games compared to those fifteen years ago? Flashes of light on the screen bam bam bam and blood everywhere, um not much different :) Dmcq (talk) 13:30, 28 May 2010 (UTC)[reply]
Probably not, modern super-computers are Massively parallel, even assuming that a game could be re-compiled to work on the super-computer's hardware and operating system, it probably wouldn't take very good advantage of the super-computer's unique hardware.
Game developers seem to have enough trouble porting software from the three-cored X-box 360 to the eight-cored Playstation3, getting them to re-work the game to effectively use 150,000 cores would be a major project.
Not impossible, of course, with time and effort a game could be written specifically FOR that architecture. (Perhaps a game that relies on super-accurate fluid simulation?) However, I think that if some crazy, rich person hired a game company to write a game for a super-computer there would be a long learning curve. It's not something they could just do. (However, as technologies like Nvidia Cuda are bringing this style of programming to the masses, perhaps in the future games will be written that could be moved over to a super-computer without much work.)APL (talk) 14:32, 28 May 2010 (UTC)[reply]
It is possible to adapt a vector-based supercomputer with associative memory to run games if you gave it graphics capability. The PS3 design is similar to the old Cray supercomputers. Of course, the old Crays didn't have graphics capability. The result would a very expensive PS3 clone. It wouldn't really run the games faster. Games don't really run as max CPU speed anymore. For example, if you played Quake on your old 486 computer and you play it on your new quad-core high speed computer, the game runs at the same speed. The framerate increases. -- kainaw 15:38, 28 May 2010 (UTC)[reply]
I think it depends on what you mean by "game". No, a supercomputer isn't going to make World of Warcraft any more enjoyable than on a moderately good desktop (if it even had graphic rendering abilities). As someone pointed out, however, Deep Blue was specifically designed to play a game better. Granted, a modern desktop running a chess engine like Fritz could kick any human player's can, so maybe that's no longer a good example. The best computer Go programs, on the other hand, are only about on par with humans at an "intermediate amateur level". See Computer Go. I imagine that the massively parallel nature of a super computer could be applied to make a better Go playing engine. In fact, one of the references [1] is about a Go playing supercomputer. Buddy431 (talk) 16:13, 28 May 2010 (UTC)[reply]

Let me ask another related question. What about this OnLive thing. Maybe not necessarily them, but could someone make a supergamingcomputer, custom built to run games.. And maybe even have programmers to write games exclusively designed for that machine. Then stream it out to consumers via high speed internet? ScienceApe (talk) 16:15, 28 May 2010 (UTC)[reply]

Sure, there is no technical reason why not. This might be useful if you had some immense simulation that was required in order to generate the super-nuanced physics simulation or graphics pipeline you want. Note that there is a lot of skepticism about OnLive in the game development community, based mostly on the problem of controller-to-screen-event latency, and partially on their business model. The supercomputer rig you describe would not improve the feeling of speed in a first-person shooter compared to playing on a computer that is sitting right next to you; but it could result in a more realistic world. You would have a problem with the business model, because if, say, 100% of a supercomputer is needed to create this super-nuanced physics simulation, that means you have to build 1 of these supercomputers for each online user. If you use 1% of the supercomputer per user, then the reason to use a supercomputer seems to have gone away. Comet Tuttle (talk) 16:48, 28 May 2010 (UTC)[reply]

"Write conservatively, read forgivingly"

[edit]

I thought I heard that quote (or something similar) in the context of interoperability between programs writing and reading data in a given format. I did a web search and was surprised that almost all the hits of the quote were related to etiquette rather than software design. Is there a well-known statement articulating the same philosophy in software design?

Yes, Postel's law. -- BenRG (talk) 06:22, 28 May 2010 (UTC)[reply]
The OP might also like to compare Postel's principle ("be conservative in what you send, liberal in what you accept") with Subtyping_of_functions#Function_types. Briefly: (S1→ S2) <: (T1 → T2) iff T1 <: S1 ("liberal"/contravariant in argument type) and S2 <: T2 ("conservative"/covariant in return type). ForTheNerves (talk) 10:50, 28 May 2010 (UTC)[reply]

Google Chart API - is this a bug or am I somehow mistaken?

[edit]

These two give the same results:

http://chart.apis.google.com/chart?cht=p3&chd=t:1.5,0.5&chs=600x300&chdl=One%7CTwo
http://chart.apis.google.com/chart?cht=p3&chd=t:15,5&chs=600x300&chdl=One%7CTwo

This one, however, looks different:

http://chart.apis.google.com/chart?cht=p3&chd=t:150,50&chs=600x300&chdl=One%7CTwo

And this one even worse:

http://chart.apis.google.com/chart?cht=p3&chd=t:1500,500&chs=600x300&chdl=One%7CTwo

Is this a bug, or did I miss something in the specifications, maybe a rule that values not summing up to 100 will be scaled to proportion, but values adding up to more than 100 are not allowed and will return garbage results?

Or is my logic flawed and 1.5:0.5 truly isn't the same as 1500:500?

Confused, -- 109.193.27.65 (talk) 08:38, 28 May 2010 (UTC)[reply]

I know nothing about the API in question, but a little experimentation with the urls you've given shows me that it's treating all values higher than 100 as exactly 100. APL (talk) 19:15, 28 May 2010 (UTC)[reply]
From here (emphasis mine): "Values are displayed relative to each other: so a chart with values 1,2,3 will look the same as a chart with values 100,200,300. However, when using text format data, values greater than 100 are trimmed to 100, so you will need to use text format with custom scaling to display slices greater than 100 properly (use the chds parameter with min/max values of chds=0,<max-slice-size>). " APL (talk) 19:19, 28 May 2010 (UTC)[reply]
Gah, I missed that one. *selfthwap* Thanks! -- 109.193.27.65 (talk) 16:41, 31 May 2010 (UTC)[reply]

Windows (16 bit) :Technical - Starting up a 16 bit app

[edit]

Following on from a previous question...

In looking up some information on Windows 16 bit lo-level stuff, I came across some mention of two functions called InitTask and InitApp...

InitTask seems to be well documented, but I can't seem to find much information on InitApp (and WINE's implementation proved to be skeletal)...

So does anyone here have a detailed spec/explanation of what InitApp and related functions do ? (BTW I have checked MSDN with no luck)

10:35, 28 May 2010 (UTC)

If you're really interested in this stuff, I'd recommend you pick up copies of Undocumented Windows by Schulman, Maxey, and Pietrek, and Windows Internals (first ed.) by Pietrek. WI is still published (it's now by Russinovich et al. and published by Microsoft themselves), but you'll want the oldest edition you can get. I can see both on Amazon Marketplace for trivial sums. It's been a decade since I knew where my copies were, but they were both interesting. Although the first one is mostly about undocumented functions, it has a lot about figuring out how the Windows API really works, which seems to be the trail of tears you've set yourself upon :) -- Finlay McWalterTalk 13:34, 28 May 2010 (UTC)[reply]
I once rather rashly made a comment on #reactos that it would be a nice idea if someone wrote a Windows layer for FreeDOS,

That's partly why I am looking into 'this stuff' .

Once I know what I'm letting myself in for I might try and see if there's any interest in actually getting such a project started. (I mean writing a 16 bit Kernel for a 286 can't be THAT hard ... <insane laughter> ) 212.225.120.224 (talk) 21:45, 28 May 2010 (UTC)[reply]

Domain name owned by domain registrar?

[edit]

Hi, I'm sort of new to the whole domain name business, so I'm sorry if this question sounds dumb. I checked out a certain domain name, and whois says the registrant is leasedomains.com, a domain registrar. I went to that page, and it tells me that the domain is not available to acquire, but it says again that the owner of the domain is that same page, leasedomains.com! What does that mean? Could I acquire that domain from them or not? (I can give the name of the domain if needed, but I'd rather not if it's not necessary.) Thanks in advance. 200.118.156.9 (talk) 15:52, 28 May 2010 (UTC)[reply]

Probably, this registrant registered the domain name under its own name to keep the real owner anonymous. In this case, they are not planning to sell it. Otherwise, they would advertise it.--Mr.K. (talk) 16:31, 28 May 2010 (UTC)[reply]
Some registrars, when anonymising their customers in this way, make it clear that they're doing so. Dreamhost, for example, list such entries like: somethingorother.com Private Registrant, A Happy DreamHost Customer, dreamhost's own address -- Finlay McWalterTalk 16:38, 28 May 2010 (UTC)[reply]

I would say this is the explanation. I forgot to mention that the domain is currently parked, and I thought it meant that it was available for anyone to get. After reading the article about it, I realized it's anything but. Thanks for your replies. 200.118.156.9 (talk) 16:43, 28 May 2010 (UTC)[reply]

old programs, new computer

[edit]

So, I have a stack of old computer games from back when I had windows 95 and 98, now I have windows 7, and many of them don't work. Some claim my computer is 64 and they need a 32 computer to run, but others from the same time, or even older, work quite well. Others, the very oldest, claim that they or the program, they're a little vague on which, doesn't support full screen mode. And some just refuse to install, or to work afterward, or the colours are all messed up. Is there anything I can do to get more of them to run, without having to go out and buy a new old computer to play them on?

148.197.114.158 (talk) 16:01, 28 May 2010 (UTC)[reply]

You can use a virtualization program such as VMware or Windows Virtual PC to run an old version of Windows. If you're lucky, the game will run without problems within the virtual machine. decltype (talk) 16:09, 28 May 2010 (UTC)[reply]
Before doing that, try Compatibility mode. Right-click the .exe and click the "Compatibility" tab. You can make Windows 7 pretend to the old .exe that it is Windows 95, Windows 98, Windows XP, etc. This has let me play some old edutainment software discs we have around the house. Comet Tuttle (talk) 16:40, 28 May 2010 (UTC)[reply]
Keyword is if you're lucky. If the software runs "without problems" it means merely that software has not crashed yet. My favorite crashed after nearly three months of regular use on Win7 - neeeded certain "file save" conditions to show up. East of Borschov (talk) 16:41, 28 May 2010 (UTC)[reply]
If they are that old they might also work under DOS so Dosbox is worth a shot for any DOS compatible titles. Exxolon (talk) 21:41, 28 May 2010 (UTC)[reply]
VirtualBox is a great free/open-source virtualization platform for Mac, Windows, Linux, try that for running these (and i find that Virtualbox is less headache then VMware or VirtualPC)--rocketrye12 talk/contribs 15:21, 29 May 2010 (UTC)[reply]

Security with Ventrillo (or other VoIP software)

[edit]

Like many folks who play MMOs, I use Ventrillo to communicate with other players. However, a persistent problem with Vent is someone deciding to be a troll and posting the vent login info (including the server password) to some place like 4chan. The result is a bunch of trolls logging into the Vent server and doing obnoxious things. My question is: is there some additional layer of security I could implement that would prevent unwanted or unauthorized users from logging into a specified Vent server? I was thinking of something along the lines of a whitelist, whereby only people with usernames on the whitelist could log in, but I'm not sure if Vent offers this functionality. I'd also be open to exploring other VoIP programs if they provide additional security features. Thanks for your suggestions. Dgcopter (talk) 16:35, 28 May 2010 (UTC)[reply]


I play WoW and once during a raid we had a person who got fed up with the group and started playing a "remixed" version of the "Slap Chop" commercial through it (it was rather hilarious though). He was kicked from the group and the vent owner kicked him from vent and also did set it to auto-kick his username whenever it tried to keep coming in. When he came back on other usernames the vent owner then said that she had blocked his IP from connecting. How? I don't know. I know that she did though. Otherwise all you can do is keep booting them, or say move yourself and others to what is otherwise a restriced channel. Sorry I couldn't help more. Shotgun5559 (Talk) (Contrib) 17:50, 28 May 2010 (UTC)[reply]
Yeah, the server admin tried blocking the IPs, too. The problem with that method is basically the same as the problem with trying to block IPs from editing Wikipedia. That is, a determined troll will find ways around an IP block by connecting via a proxy server or whatnot. And, in fact, the trolls we're dealing with have done exactly that. I was hoping for a bit less of a "whack-a-mole" type solution. Thanks for the suggestion, tho. Dgcopter (talk) 18:45, 28 May 2010 (UTC)[reply]
Only other thing I can think of is to change the password, or to change the info entirely.... Shotgun5559 (Talk) (Contrib) 19:10, 28 May 2010 (UTC)[reply]
You could try contacting the users ISP if you know that (which you may if you have their IP from before they started using proxies). There's no guarantee it's going to work, but a properly worded e-mail particularly from a server admin can do (people do file abuse reports on persistent abusers for wikipedia and they do sometimes work although a common response is, we need to be contacted by the server admin).
More generally, depending on how Ventrillo's IP blacklist is implemented you may be able to use it as a whilelist. I've seen people do it before with the ipfilter.dat used by many P2P programs and there are IP managers which will help you set this up. Obviously if the blacklist doesn't allow IP ranges this definitely won't work.
Nil Einne (talk) 23:39, 28 May 2010 (UTC)[reply]
The User Access Rights feature allows the server administrator to create more advanced security schemes. For instance, restricting guest accounts from entering certain channels, or prohibiting them from transmitting. Users are considered guests unless they login with a username and password created by the server administrator. decltype (talk) 00:26, 29 May 2010 (UTC)[reply]

Port forwarding in Ubuntu

[edit]

I am trying to access a Cpanel behind the ports 2083 and 2082 through Ubuntu, working behind a proxy which cannot be changed and blocks everything which doesn't go through the ports 80 or 8080. How can I accomplish that? --Mr.K. (talk) 17:35, 28 May 2010 (UTC)[reply]

With ssh's tunneling feature, probably. Unfortunately I didn't really understand your question well enough to give you the exact syntax. --Sean 20:44, 28 May 2010 (UTC)[reply]
Yes, if you could be more specific, we might be able to pose an alternate solution. But off the bat, it definitely sounds like a tunnel down port 80 is your best bet. --rocketrye12 talk/contribs 15:22, 29 May 2010 (UTC)[reply]
thanks so far. It seems that this is exactly what I need. I am trying to reach the address http://www.heliohost.org:2082/login/ but my current internet connection (at work) doesn´t allow any traffic through this port (in the same way that it does not allow peer to peer connections).--Mr.K. (talk) 15:57, 29 May 2010 (UTC)[reply]
Are you able to establish a connection to a server/another computer running SSH? --rocketrye12 talk/contribs 01:03, 30 May 2010 (UTC)[reply]

Human Memory

[edit]

What is the limit for a regular grown up human to remember numbers? —Preceding unsigned comment added by 83.183.172.203 (talk) 18:50, 28 May 2010 (UTC)[reply]

This page lists people who have been able to recite many digits of pi from memory. The world record holder apparently recited 67,890 digits from memory in just over 24 hours. We have an article on him: Lu Chao. (I am not claiming this number is a "limit".) Comet Tuttle (talk) 18:53, 28 May 2010 (UTC)[reply]
For a "regular" person, it is tied to what they practice. For example, most Americans can easily remember seven digits easily. Why seven? They regularly memorize 7-digit phone numbers. If, instead, phone numbers were 8 digits, it would be common for them to easily remember eight digits. As with everything else, it isn't a real limit. It is just what the average person is used to. -- kainaw 18:57, 28 May 2010 (UTC)[reply]
Actually I was taught that, per some of the material in our Chunking (psychology) article, the idea of having people memorize 7 numbers (plus or minus 2) does have some neurological basis. (PS: Actually The Magical Number Seven, Plus or Minus Two is the better link.) Comet Tuttle (talk) 19:00, 28 May 2010 (UTC)[reply]

What is a number? I think if you can recall (ie picture right now) ANYTHING iconic, from a classic iPod, to a cooper mini, to a big mac, or the Coca Cola bottle... anything at all: then each such icon ought to count for hundreds of numbers. I would even say it is recall on some level if you can't recall it, but can tell if it's "off". The exact number of bits of entropy where if you show a picture deviating by that many bits of information, the person will MISTAKENLY say it is the famous icon, gives you an idea of how much information is in the icon. Faces are similar. Now I am going to go out and say that remembering a face is no different from remembering numbers - they are just bits of data. So let's imagine a hypothetical "photographic memory" person, who spends their WHOLE life, 20 hours awake, meeting new people for - let's say 3-10 seconds each. So, they live 80 years this way. That's 10^9 seconds (1,000,000,000). For an upper bounds, let's say 1 second per face, and for extra entropy the person is to be listening to something that they will associate with that face. And they will be in a specific geographic place when they meet the person for a second - say a few days to weeks in each large city in the world. Anyway you space the 80 years so that all the big cities come up. Now you see the problem: it is quite easy to recall, perhaps after a bit of thinking, just WHERE you met someone. ie to associate every single face with the geographical area you were. It is not a stretch to associate with the time of day, morning or night. It is not a stretch to associate it with whether you were in severe pain, whether it was when you were in a period of recuperating from an excrutiating skiing accident in which you broke 85% of the bones in your body. I think no one who met someone for five minutes while in such a painful recuperation would forget that they met the person under such conditions - I am assuming they are not so medicated they don't recall the meeting at all. But now you see the problem: the brain really IS associative! There is HARDLY ANY limit on the number of things that can be associated with a meeting of 5 minutes or even just a second, if it really grabs you - or if you have a photographic memory. Where is the storage limit of an associative brain, in which neurons are connected with other ones. The answer comes from graph theory, and by now the reader will see exactly where I'm going. I don't need to whip out the scientific notation and start multiplying exponents, and raising exponents to exponents. Suffice to say, that I maintain that the only realistic answer is infinite, in the word's non-technical definition. In the word's technical definition, we can't use infinite, since there aren't enough brain states to put them in 1:1 correspendence with even the natural numbers - you run out of brain states. That means they can't be even the smallest kind of infinite. So, in a mathematical sense I would give the number as ten to the bajillion to the bajillion to the bajillion. That should satisfy the mathematically trained among my readers, who would balk at the more honest ansswer: infinite.82.113.121.38 (talk) 21:09, 28 May 2010 (UTC)[reply]

Please cite sources; this is a Reference Desk. We have an article, Exceptional memory, about extraordinary general (not numbers only) memory recall. Comet Tuttle (talk) 21:27, 28 May 2010 (UTC)[reply]
I don't need references, foo' - I AM the reference! 82.113.121.126 (talk) 13:44, 29 May 2010 (UTC)[reply]

Norton Internet Access Control Alerts

[edit]

I have Norton/Symantic security and I keep getting pop-ups which ask me what I want to do when some program is attempting to access the internet. It says "Automatically update security settings for programs that Symantic has identified as safe." but it doesn't say how to do this. I have Live Update set to automatic already but that only starts Live Update as soon as you click on it, it doesn't update my security settings every fixed amount of time. How can I stop geting these pop-ups?98.77.194.6 (talk) 19:13, 28 May 2010 (UTC)[reply]

Nobody has helped me in two days. I deleted Google toolbar and I haven't seen a pop-up since.74.233.118.147 (talk) 18:46, 30 May 2010 (UTC)[reply]

HELP I'm seeing new pop-ups again. How can I stop them, they are driving me crazy!74.233.68.2 (talk) 20:43, 2 June 2010 (UTC)[reply]

Reading .docx

[edit]

What's the smallest, free, non-time-limited, downloadable piece of software that can read docx files? Suitable for downloading on a dialup-speed connection. Additional criterea: the software must be any good. I'm trying out OpenXML Document Viewer right now, as a firefox plugin, and I've got a bunch of newlines where there shouldn't be any, and generally sub-optimal formatting. 213.122.46.240 (talk) 21:06, 28 May 2010 (UTC)[reply]

If you just want to view the document, you can upload it to Scribd or Google Documents, both of which read .docx files. You can also use them to convert the .docx to something else, which you might have a reader for (e.g. PDF). -- Finlay McWalterTalk 21:50, 28 May 2010 (UTC)[reply]

http://www.zamzar.com is an excellent, fast and efficient conversion website. ╟─TreasuryTagUK EYES ONLY─╢ 21:51, 28 May 2010 (UTC)[reply]

Thanks, but I did mention "downloadable". Oh, and Windows XP, I should have mentioned. I just tried TextMaker Viewer, which got the newlines right but didn't display some equations. Perhaps there's a python module for displaying docx? 213.122.46.240 (talk) 21:55, 28 May 2010 (UTC)[reply]

OpenOffice isn't very small, but it does read .docx files completely. The latest version is pretty competitive with the full Microsoft Office product line, and is totally free and open-source. Nimur (talk) 23:38, 28 May 2010 (UTC)[reply]
Personally I wouldn't want to regularly download it on a dialup modem though although I did download CD size stuff one or twice in my dialup days I think and of course Windows SPs. Nil Einne (talk) 23:45, 28 May 2010 (UTC)[reply]
You can also buy copies of OpenOffice on CD. Not quite free, but I think 10-20 USD rather than 100+ for Microsoft Office. Amazon appears to have them for $15 (plus shipping, of course). Buddy431 (talk) 01:07, 29 May 2010 (UTC)[reply]

(OP here) I can't physically download 150Mb (which seems to be the size of the OpenOffice installer) - I'll get halfway through, which will take four hours, and then I'll get cut off. Oh well, never mind. Got my document in .pdf now anyway. And I guess the Firefox add-on isn't that bad. It's only quite bad. 81.131.23.148 (talk) 03:47, 29 May 2010 (UTC)[reply]

Unrelated to the initial topic, but...on dial-up you must do yourself a favor and acquire a download manager. Few things more frustrating than a failed download 4 hours in. FileHippo provides a few choices. I don't actually use one myself anymore, but I think I used GetRight back in my dial-up days. -Amordea (talk) 05:17, 29 May 2010 (UTC)[reply]
I use Free Download Manager. You could also try using torrents, they would handle disconnections automatically although your download speed could be limited Nil Einne (talk) 20:07, 29 May 2010 (UTC)[reply]
As it seems you just want to view the document, and not edit it, I'm surprised no one has mentioned Microsoft Word Viewer -- 174.24.200.38 (talk) 21:27, 29 May 2010 (UTC)[reply]
Notepad works well. Just change the file extension from .docx to .zip, double-click on it, and drag out the document.xml file from the \word folder. You can then open it in Notepad. It comes with Windows, so it'll save you some download time. Also, the version of Wordpad that comes with Windows 7 can read .docx files.--Best Dog Ever (talk) 20:18, 31 May 2010 (UTC)[reply]

can an EEE PC 1005H act as a wireless access point?

[edit]

short and simple: some laptops can act as a wireless access point, some can't: point in case, macbook pro can, macbook can't.

My question is: can my eee pc 1005H netbook act as a wireless access point or can't it?

(details: I am thinking of purchasing an iPad, but already pay a monthly cellular charge for a 3G enabled netbook - the model mentioned above. I'd rather not pay a second one to enable the iPad to function. One option would be to chisel down the sim card from the netbook into a microsim size, but I'd rather prefer to keep them both connected. Therefore the best solution would be to share the internet connection over wifi with a wifi-only iPad (also saving $100 in the process) - but of course this is only possible if it is possible. So, with the eee pc 1005h, is it possible? If not, I guess I have to buy a myfi or something. Thanks.)82.113.121.38 (talk) 21:42, 28 May 2010 (UTC)[reply]

I have had success configuring my netbook as a wireless DHCP server. I use a Lenovo S10E, with a Broadcom BCM4300 wireless chipset, running Ubuntu. Setting up the DHCP server is "non-trivial", but not beyond the reach of a technical user. (Alternately, you can try some "internet connection sharing" features on either Windows or Ubuntu or other systems, but I have had less success with that method). These forum posts seem to indicate that some EEEPC wireless cards are not configurable to act as wireless access points, but you can give a shot with your model. If you can post the wireless card manufacturer and model, we might be able to help you find a more complete answer (I couldn't find the exact wireless card for the EEEPC 1005H on the Asus website or elsewhere). Nimur (talk) 00:46, 29 May 2010 (UTC)[reply]
I believe that you are a bit mistaken a bit here --- this idea that some laptops can act as APs and others can not is a hardware generalization that I believe you are misplacing with the presence of software functionality: more often then not the ability to use a computer as an access point depends on two things: (1) Originating internet connection, and its tolerance to having DCHP run in front of it and (2) Software ability to switch the connections. Macs are known for having very easy internet sharing capabilities, that is both Macbook Pro, Macbook, Mac Mini; virtually all models. In the case of your eee pc, see this how-to article for instructions...definitely should be possible. Let us know if you run into any issues.--rocketrye12 talk/contribs 14:56, 29 May 2010 (UTC)[reply]
In order for a laptop to operate as access point, its wireless card must support master mode (if operating under linux)(in windows there may be a special application). I have seen wireless cards with such apps in windows, but i have never seen a wireless card with such a functionality in linux. Most wireless cards supports ad-hoc mode and in many cases it is the only useable solution. Network operating in ad-hoc mode, however will have lower throughput (because of less efficient protocol). -Yyy (talk) 08:13, 31 May 2010 (UTC)[reply]

Vista laptop randomly shutting down

[edit]

For some reason, my 4-year old Gateway laptop shuts down on its own. The problem started a few months ago. From time to time, the computer shuts itself down, no warning, nothing. And it's not like I face an immediate black screen or crash - the shutdown fades out, exactly like if someone had hit the Hibernate or Sleep function. When I start my laptop back up again, it proceeds as if nothing ever happened; no error message, not even an entry in the Event Viewer. So, my question is, what the hell is going on with my laptop, and how do I fix it? Could this be an overheating issue? What tools would I need to further diagnose the problem? Please help, and thanks, 141.153.216.29 (talk) 22:35, 28 May 2010 (UTC)[reply]

Does this happen when the power is plugged in? Nil Einne (talk) 23:30, 28 May 2010 (UTC)[reply]
Yes. 141.153.216.29 (talk) 23:47, 28 May 2010 (UTC)[reply]
It goes into hibernation specifically? If it was heat, it would likely just cut the power immediately after reaching a critical point (although certain software might be set to hibernate at critical temp instead if you have such a thing installed). You may want to check your power settings to see if anything has been altered (Control Panel -> Power Options) and see if that correlates to the symptoms you're experiencing. If not, it certainly couldn't hurt to give the computer a nice dusting. They need them about once a year. -Amordea (talk) 00:36, 29 May 2010 (UTC)[reply]
No, it actually shuts down (if it hibernated, I would have seen a screen to log back in, and my files would still be open). My power settings seem fine; nothing's different. Do you know any good tutorials/tips on dusting a laptop, or should I just refer to Google? Thanks, 141.153.216.29 (talk) 03:21, 29 May 2010 (UTC)[reply]
While not an ideal guide it's a decent general guide. Major points to follow: be sure to disconnect ALL power sources (adapter and batteries) and press the power button before touching the insides of the laptop (this actually does more than makes sure there is no more power connected to the laptop, it also discharges the capacitors of any residual charge). Be sure to discharge yourself too by touching a metal surface to get rid of any static electricity you may have built up.
You do not need alcohol or glass cleaner (I don't even know that I'd use glass cleaner around electronics), a moist cloth will do (moist, not wet) and you don't need compressed air. Your god-given blower will do just fine (just be sure to dry your mouth before blowing and if you do get any saliva on the electronics, wipe it off before turning on the laptop again). Do not use a vacuum cleaner! If you do use compressed air, stick a pen or toothpick or something in the fan to keep it from damaging the motor by spinning it too fast.
If you're having trouble finding your fans, refer to your model's service manual for more detailed information specific to your model of laptop. -Amordea (talk) 04:02, 29 May 2010 (UTC)[reply]
What do you mean by "the shutdown fades out"? You say it proceeds as if nothing happened - are your programs and files all still open and as you left them? --Tango (talk) 01:11, 29 May 2010 (UTC)[reply]
On a Vista machine, when a typical shut down sequence is performed, the screen will display "Shutting down", and then fade to black. When the hibernation sequence is initiated, the screen will merely fade to black. Here, the screen merely fades to black, but when I turn it back on, it goes through the normal cold boot sequence (whereas I would see a login screen if it were hibernating). When I said "nothing happened", I meant to say that the OS acts like as if I had shut down the computer myself. 141.153.216.29 (talk) 03:21, 29 May 2010 (UTC)[reply]
Most likely an issue with overheating. You will not see an event in the event viewer, as this is a much lower level issue. I would recommend gently vacuuming out the fans at a safe distance --- putting the vacuum tube onto a couple of extenders works well, but it seems like others here have recommended against this; try at your own risk! Unfortunately, not much you can do in cases like this --- as repair on a 4-year-old machine, while probably not possible, would certainly be the cost of a brand new more powerful computer. --rocketrye12 talk/contribs 15:01, 29 May 2010 (UTC)[reply]
Truth be told, I've used a shop vac to clean PC's out plenty of times. But I always read other places that you shouldn't, from nearly every source I've read about cleaning PC's--supposedly has something to do with static electricity (I don't touch the hose to the electronics anyway, but it seems wiser just to advise someone against it entirely than to let them engage in unnecessary risk). I've never had a bad experience, but I figure it'd be just like Murphy's Law to crop up the day I don't advise someone against it. Plus I had some horror playing in my head of the unwitting person taking a vacuum with one of those bristle bars to the laptop... -Amordea (talk) 15:41, 29 May 2010 (UTC)[reply]
I agree, it sounds like overheating. It isn't shutting down or hibernating, it is just turning itself off. Before worrying about dust in the vents, you should check they aren't being blocked by wherever you put your computer. If the computer is on soft furnishings or your lap then the vents may be blocked. Try using the laptop on a hard surface like a table, desk or one of those lap tray things. --Tango (talk) 15:19, 29 May 2010 (UTC)[reply]

MonoDevelop vs. SharpDevelop con't

[edit]

In reference to this question here[2] (which is now archived):

Don't know. Most .NET developers use Visual Studio. In any case, AFAIK, both SharpDevelop and MonoDevelop are free. Why not try them both and decide which you like better? Alternatively, you can download Visual Studio 2010 Express for free[3] (although it's not open source). A Quest For Knowledge (talk) 23:52, 28 May 2010 (UTC)[reply]