[…] one that I bought in September. If you have been following along, you might remember that last year I blew a video card in my old rig. I managed to squeeze maybe six more months of use out of that old rig by putting in […]

[…] old gaming rig has started to fall apart lately. Last year my video card blew up and took one of the PCIe slots with it. It took me a few weeks to sort it out. I purchased a […]

[…] PC has died once again. This has happened before but it turned out to be a video card failure. I was able to identify the problem by listening to […]

I had a problem with my ATI Radeon card recently. It was badly overheating and the fan controller was not working. After ruling out other ways of repairing, I decided to change it for a newer Radeon. It was a very nic failure. The new card did not recognize my fairly standard Samsung screen properly. So I went back to the shop and changed it. I took an Nvidia GForce. I had never used Nvidia before and I am sorry I did not switch earlier. It is quite a breeze to use in comparison. Now, all I need to play Skyrim nicely is a new CPU… But finding a Core2 Quad 9200 these days is nigh impossible…

@ Victoria:

Ah, I forget that Apple stores are not as ubiquitous around the world as they are here in the tristate area. :P

@ Luchs:

Uh, this is actually my concern too, especially since my machine was born as a Dell and they do have a nasty habit of building their machines in such a way that it is very hard to upgrade anything. That said, the XPS case seems somewhat spacious and there seems to be lots of space above and below the PCIe slots so hopefully all will be well.

@ Eric:

Yeah, it’s kinda silly to stop POST if the video card is not working. I guess I can see some logic here – if the card is dead you won’t get any output to the screen ever, which might be hard to diagnose. Especially if something else goes wrong. For example, if a headless machine loses networking and you don’t know it also has a bad video card, then it becomes a mysterious black box that seems to boot all the way up but doesn’t.

So a “fail early and fail hard” approach helps to identify the issue more readily.

@ MrPete:

Oh wow, I have never seen a fan with a broken axis either. You’d figure it had to be a factory defect or something. Doesn’t seem like this would be a common wear-and-tear type issue.

@ Robert:

Wow… That’s… Priceless. How could presence of a PS/2 mouse affect specific network transfers on a hardware level is completely beyond me. That’s some next level engineering right there. lol

@ Tj:

Well, most people that probably wouldn’t mind dragging over their computer to my house, or letting me barge and canibalize their devices are non-gamers who switched to laptops long time ago or they live kinda far these days. I hate to impose and ask people I don’t regularly hang out with (other than occasional interweb exchange or game of DOTA/TF2) for random favors like this. :P

Do you have a friend you can swap parts with to isolate the malfunctioning one? Maybe it’s the motherboard?

I once had a server lose three of its four drives, the raid controller, backplane, and backplane cable, all at once. Fun night. (I think a flaw in the cable took down everything else. Took a week and four trips for Dell to get all the bad parts replaced.)

_Weirdest_ hardware failure I’ve encountered was a bug. Back in 2002 a friend and I had both built PCs around the same AMD multi-processor board (a rarity in those days). I had no trouble with mine, but he had a weird problem with his. If he copied a lot of files off it over the network, it would blue screen, every time. Only with a big copy, and only if the copy were initiated from the other computer. (Sitting at computer A, he could copy anything he wanted to computer B. Sitting at computer B, though, copying the same files from A would cause A to crash.) He tried replacing the network card, updating drivers, reinstalling Windows, all the usual stuff. Finally just had to live with it.

A few months later, I got a new laptop and in the process of setting it up I tried to pull a bunch of MP3s off my desktop. BLAM: blue screen. Once is a fluke, twice is a problem, right? I tried it again, with the same result. Fiddled with a few things, then remembered that my friend had had the same problem. Called him up. Had he ever gotten it fixed? Nope.

No help there. I turned to a web forum where a lot of people used that particular board and explained the problem. Anyone run into that before? Got some ‘yeah, same here’ replies, then a “What kind of mouse are you using? PS/2 or USB?”

WTF? USB, of course. But that shouldn’t matter….

Try plugging in a PS/2 mouse, the mysterious stranger suggested. I said something like, “You’re shitting me,” tried it, and said, “I will be god damned.” It worked. I called up my friend, asked him what kind of mouse he was using. USB, of course. What does that have to do with– “Try plugging in a PS/2 mouse.” You’re shitting me, he said, but he tried it and then said, I’ll be god damned. (The PS/2 mouse ended up as a dongle hanging off the back of the computer.)

Large network copies would cause the host PC to crash, but only if initiated from the remote computer and only if a PS/2 mouse was NOT plugged into the host. I don’t know how AMD managed that one.

Oh, hardware failures!
Luckily my actual system is not suffering from those (until now).
The last one I had was my previous gaming rig starting to slow down. At the time I was playing GTA San Andreas and NFS Underground (meanwhile very unusual genres for me). With GTA I could play for 1.5-3h before anything unusual happend, but with NFS it was a mere 30min. About 30min later everything was fine so I started thinking.
Looking at both games I then came to the conclusion that the difference was graphics.
So I took the usual precautions (disconnect power, always touch the metal casing and so on) and had a look inside.
I was very impressed, both by the failure and the sturdiness my graphics card demonstrated: The fan axis was broken, the little motor vainly trying to turn the blades which lay sideways in the heat sink.
I was fortunate in that a new cooler solved all problems (except for a single line near the bottom of the screen with a slight bit offset) so it’s not really qualifying for a disastrous failure.
Although, it was the first fan I ever saw with a broken axis and not an engine failure…

I also find beep counting (and distinguishing short vs long) difficult. I find the best way is to gather about 5 highly trained computer exports around to listen together. It inevitability leads to arguments in how many beeps there are, which involves everyone beeping at each other.

Also I don’t like that you can’t POST a computer without a video card? Seems like a bad reason to halt the whole boot process.

My previous desktop PC which was a relatively cheap prebuilt computer suffered hardware failure after the first TF2 Halloween event. After playing all week it started showing colorful lines. A few days later, the graphics card was completely broken. I didn’t have any key to the motherboard beep codes, but it was pretty clear – the card didn’t even have its own fan. Even the heat sink was so small that I was seriously surprised it didn’t break earlier.

I thought “no problem – some average new card will do the job”, so I bought a rather cheap one for about 100€. When it arrived a few days later, it turned out that the card wasn’t cheap enough for my PC – its motherboard was one of the few BTX-type boards which are attached to the left side of the case. This means that the card’s fan is on the side of the CPU instead of the side of the other expansion ports. Unfortunately, the PCIe port was still the first port after the CPU, so my new two ports wide card had no chance to fit in.

In the end, the only component I could keep was the hard drive, even the DVD drives weren’t usable because the new motherboard didn’t have an IDE connector.