Wikipedia:Reference desk/Archives/Computing/2011 August 17

Computing desk
< August 16 << Jul | August | Sep >> August 18 >
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.


August 17

edit

Most likely thing to die in a PC from a power surge

edit

Of course the cause of death is the power surge, but in a case where the thing now does absolutely nothing when you press the power button, what is the most common thing or set of things that that surge has done to kill the PC? Is it most often frying of traces on the PCB, unviewable damage to the CPU, the chipset, or what? 20.137.18.50 (talk) 13:11, 17 August 2011 (UTC)[reply]

Power hits the power supply first. The power supply is designed specifically to kill itself in order to save the rest of the computer (otherwise, we wouldn't use power supplies - we'd just run power right into the motherboard). So, the power supply is the most likely thing to die in a power surge. -- kainaw 13:49, 17 August 2011 (UTC)[reply]
That is some hopeful news, Kainaw, but one thing I noticed while looking at the motherboard and plugging and unplugging the power cord is one really tiny square green light that still comes on when it's plugged in but goes out when I pull the plug. Would that fact take away the possibility that it's just the power supply, or can power supplies too damaged to let the thing turn on right still provide some power to the motherboard? 20.137.18.50 (talk) 13:55, 17 August 2011 (UTC)[reply]
It could still be the power supply. It could also be that the power supply failed to protect the computer and there are more bad parts. I always start by swapping out the power supply. If it has a separate network card (which is rare now), I swap that out too because almost nobody puts a surge protector on their network cable. Then, I swap out the motherboard if it still doesn't work. Then, if there are still issues, I keep swapping out parts. -- kainaw 14:19, 17 August 2011 (UTC)[reply]
Something quick to check is the power cord itself, including any power bars involved. It's less likely than the power supply, but it's a simple check and a cheap fix if it turns out to have been fried. Matt Deres (talk) 19:57, 18 August 2011 (UTC)[reply]
I agree with all of the above. You can also check your physical switch and the wire that connects it to the motherboard. I've had this switch melting before. Power surges are one of the banes of modern man. Sandman30s (talk) 15:28, 19 August 2011 (UTC)[reply]

Someone sending emails from my Outlook

edit

Tonight I wrote and sent a single email. It was the only email in my outbox but the send/receive process seemed longer than it should be. The notification in my systray said "sending message 3 of 3". I definitely did not have 3 emails to send and my 'sent' folder shows only the email I knew about. Could someone be using my Outlook to send spam emails and also be hiding the evidence so I can't see them? If so, how can I see what has been sent and how can I prevent it happening again?

I hope you can help. Gurumaister (talk) 17:27, 17 August 2011 (UTC)[reply]

If you're using Outlook Express, it has an option to "Break apart messages larger than" that lets you specify a size in kilobytes. When you send a large e-mail, this option will split it behind the scenes into separate e-mails under the size limit. When this happens, you'll see something like "Sending message 3 of 3" when sending, but the Sent Items folder only shows the 1 message in its original form. Maybe an option like this is the cause of what you saw. --Bavi H (talk) 01:56, 18 August 2011 (UTC)[reply]

Thanks Bavi H - I hope that was it. Gurumaister (talk) 18:30, 19 August 2011 (UTC)[reply]

Reversi Bug

edit

Hello. On occasion, at the end of my Reversi gameplay, the last pebble placed on the board switches colour between black and white. How can I fix this bug? You may access my source code at http://sites.google.com/site/superaec and download the files that come with the package (i.e., Clap.au, Hi.au, Lo.au, Manual.pdf, Oops.au, and Reversi.java). If you need more time to answer this question than it can last on the reference desk, please email me through my talk page. Thanks in advance. --Mayfare (talk) 18:10, 17 August 2011 (UTC)[reply]

It appears that you only set the color in one place: line 139,
  cell [i * DIR [b] + box] = me;
Is that line of code always valid? In other words, under what circumstances would i * DIR[b] + box be invalid? Under what circumstances would me be opposite what you expect? If you can define the answers to those questions, you can set a conditional breakpoint to automate the process of trapping the error-case.
Debugging your code for you is sort of out of scope at the Ref Desk; but I can recommend this tutorial on using the Eclipse Java IDE for debugging: The Eclipse debugger and the Debug view, from IBM. Use the step-debugger to trace your code execution; use the variable watch window to investigate the values of variables at runtime. Debugging computer code is a very valuable, very difficult-to-acquire skill. Nimur (talk) 18:46, 17 August 2011 (UTC)[reply]

A computer without a chassis?

edit

As I've assembled two computers from individual components inside chassises (chasses?), I got to think, would it be possible to assemble a fully working computer entirely without a chassis? Just the components connected to each other, lying on the desk or on the floor. What would be the advantages and disadvantages of such a thing? I can think of two advantages - better airflow and saving the cost of a chassis - and one disadvantage - the system takes up more room and is harder to transport. Are there any other things to take into account? JIP | Talk 20:41, 17 August 2011 (UTC)[reply]

When Google started, they laid bare motherboards on baking sheets (I think with a rubber mat between the MB and the sheet). They then racked the boards in dense racks, and ventilated them wholesale from the side. They were mostly doing this for reasons of density. Having just a few, which you intend to transport, it'd be bonkers to not have a case. Cases are relatively cheap, and they protect the delicate PC components from ESD, thermal, and mechanical insults. And don't assume that the case hampers cooling - often the opposite is true. When electronic devices are designed the heat dissipation is modelled on a program like FlowTherm, which tracks where the air from fans goes, and analyses how the heat from components dissipates. The case is part of that equation; if you remove it, you risk the fans blowing air straight off into free space, rather than over all the components they're supposed to be cooling. -- Finlay McWalterTalk 20:54, 17 August 2011 (UTC)[reply]

Yes, you should watch the film π ¦ Reisio (talk) 21:15, 17 August 2011 (UTC)[reply]

You lose proper cooling air flow circulation. As well as protection from dust, falling objects, and spilled food and drinks. Likely shortened lifetime of hardware, due to pieces of your flaked dead skin and other such gunk accumulating on delicate circuit boards and causing erratic currents. 88.112.59.31 (talk) 00:38, 18 August 2011 (UTC)[reply]
I'm not sure cooling is as much of a problem when it's not in a plastic box? I had a friend who was into the whole "totally silent computer" fad awhile back (or maybe it is still going on?) and he had a whole, non-chassis computer laid out in such a way that he didn't need any fans, or something like that. So yeah, you could do it. But there are obvious disadvantages to having all that stuff exposed. --Mr.98 (talk) 01:26, 18 August 2011 (UTC)[reply]
Over the years i've gotten used to working with open-chassis machines, and by my experience they do cool quicker and better than if they are still in the chassis. However, if i were to set my machine on the floor, it'd gather way more dust than it does on the desk. Still have to be careful with the drinks and all, but unplugging it and dusting it once in a while does the trick. Removing the chassis altogether seemed pointless for me, so I just removed the top cover and side panels as much as was possible without hassle. --Ouro (blah blah) 09:44, 18 August 2011 (UTC)[reply]

Another disadvantage comes to mind. If you accidentally kick a computer chassis over, the computer inside it will probably still keep on working. But if you accidentally step on exposed computer internals, you risk damaging them beyond repair. JIP | Talk 19:23, 18 August 2011 (UTC)[reply]

A friend of mine has his home computer zip-tied to a peg board and attached to the bottom of his computer desk. It is all spread out on the board and the fans are still there blowing over the main components. The disc drives are mounted at the front of the desk right next to his keyboard tray. It creates a very clean environment. Since the pegboard is hinged to the front of the desk, all he has to do is unplug his monitors and the board swings down so he can work on the components. The only issue he has had was he used a wide ribbon cable to extend his video card off his PCIE slot and at full speed he would get video problems. So, he lowered the video card speed (which is fine since he doesn't do any gaming) and all has worked fine. -- kainaw 19:30, 18 August 2011 (UTC)[reply]

Finding and configuring certain hotkeys

edit

Is there a way to find out what key combinations do what? For example, whenever I press AltGr+Shift+E, whatever window is currently in focus becomes out of focus. The key combination was once used to open up ASUS WebStorage, which I have since uninstalled. However, it seems that Windows is still expecting WebStorage to run or something of that sort, which might be why the window blurs. 141.153.216.54 (talk) 21:04, 17 August 2011 (UTC)[reply]