A GTX560 has a TDP of 150 watts. It might work but the PSU wouldn't be happy when the card was under high graphical load.

RacoonRider wrote:

What we received later was an OEM InWin system with a separate boxed GTX560. I don't know if my estimate was wrong in the end, I quit the year after this happened and never looked back. Looking at GTX560 specs though, I see "minimum 450W power supply", so I might have been right all the way.

https://www.techpowerup.com/reviews/Palit/GeF … 60_2_GB/25.html

The GTX560 itself can consume a little over 200W, but that's a maximum measurement, in an unrealistic synthetic benchmark scenario. Realistic peak consumption is ~160W. So, with a real, good 300W PSU, it would probably be fine. The problem is that many PSUs are junk and can sometimes provide only 50% (or less!) of their stated specification. For this reason, video card manufacturers always state inflated power requirements.

What's strange though is that they even bothered to order a separate GPU, when there is one integrated in the CPU. For what seems to be typical office/university work scenarios, it seems hardly necessary.

dr_st wrote:

https://www.techpowerup.com/reviews/Palit/GeF … 60_2_GB/25.html […]
Show full quote

RacoonRider wrote:

What we received later was an OEM InWin system with a separate boxed GTX560. I don't know if my estimate was wrong in the end, I quit the year after this happened and never looked back. Looking at GTX560 specs though, I see "minimum 450W power supply", so I might have been right all the way.

https://www.techpowerup.com/reviews/Palit/GeF … 60_2_GB/25.html

The GTX560 itself can consume a little over 200W, but that's a maximum measurement, in an unrealistic synthetic benchmark scenario. Realistic peak consumption is ~160W. So, with a real, good 300W PSU, it would probably be fine. The problem is that many PSUs are junk and can sometimes provide only 50% (or less!) of their stated specification. For this reason, video card manufacturers always state inflated power requirements.

What's strange though is that they even bothered to order a separate GPU, when there is one integrated in the CPU. For what seems to be typical office/university work scenarios, it seems hardly necessary.

Well, the head of our department insisted on a separate high-end GPU, which he believed AutoCAD and Kompas3D would take advantage of. They don't, as far as I know. And btw, this situation is very typical for Russia. Budget funds are funds that need to be spent regardless of wether it is nesessary or not...

smeezekitty wrote:

But what was that smell? It became a chemical stench seconds later when I knew it was the CPU that was getting fried 😮
Switched it off and of course I wanted to know what the problem was.
Well, after a short inspection apparently the problem was that I had forgotten to use any TIM (I had installed the HSF directly onto the CPU with nothing in between). The funny thing was that the print on the Athlon XP was burned onto the heat sink, but mirrored

I am really surprised it got hot enough to fry just by leaving the paste off. I guess the heatsink and/or CPU had a poor surface? I have run my Athlon 64 X2 with a heatsink just laying on it and no paste to speak off. Of course it ran warm but didn't fry.

The only really dumb thing I can think of right now is accidentally spilling liquid on a hard drive circuitboard while it was on. There was smoke...

Knowing myself, I probably did clean the surfaces of the CPU and HSF with some rubbing alcohol, but I suppose it still was bad enough to make the CPU hot enough so it could fry itself.

I had the same thing happen to an Athlon Thunderbird 900 which I've accidentally overclocked to 1200MHz at least twice (no TIM) and it posted just fine and never fried.

It's possible the Palomino was overclocked by accident or something, but I'll never know as the thing never posted (it never got past a black screen and it started smelling as I was waiting for the screen to come on). Perhaps the heatsink wasn't completely level, but I doubt that's the case as I'm usually very thorough when builing computers (doublechecking and triplechecking every step of the way).

And I remember another mishap of mine.

In my dark ages of computer building...err I mean computer messing around, I was I think busy installing Windows or something (the computer I was working on was turned on and I was probably busy installing something as I was waiting for it to finish whatever it was doing).
I was drinking a glass of wine and (I kid you not) I had put the glass on top of my CRT monitor.

Yup...as I tried to pick up the glass, I accidentally tipped the glass over and the entire content of the glass spilled into the CRT which was displaying the progress of the computer I was busy with in a software way. The screen became all wobbly and I kinda panicked and unplugged the power cord of the entire rig, instantly swithcing off not only the CRT, but the entire system.

Then I just stood there for a minute or so I think...kinda in disbelief.

I ended up putting the CRT in the shed and bought another (secondhand 14.5 inch HP-branded CRT monitor) and a couple years later I felt confident enough to open up that CRT, cleaned the insides of all spots of wine (I think it was red wine as the spots were easy to identify), closed it, attached it to a working rig, switched the PC on and the CRT worked flawlessly 🤣!

Still have that CRT though I haven't tried it in years

Sorry for double post, but I just reread the title of the thread, it includes things you see others do.

Well, this does open up a can of worms 😁

I was once working with 2 colleges and one of them was at the final stages of building a "new" rig (build using gifted old parts from schools for us to give away to poor people who couldn't afford a computer at the time). They were busy setting up Windows and were already at the desktop (installing drivers and such) when the guy (he was a former bike repairguy...yes 🤣 ) noticed the Creative Labs sound card wasn't properly set into the motherboard (it wasn't pressed into the slot and wasn't being detected by Windows, the board was on the table and not build into a computer case) and right after he noticed, he pressed the sound card into the motherboard while Windows was busy installing stuff..instant BSoD and the motherboard was toast!

The guy really should never have bothered building computers, he also stole stuff (to sell on secondhand websites) and had an ego bigger then the room we used to build the rigs we got donated. He was a pain to work with, I ended up working someplace else after a while.

dr_st wrote:

Well, when I was a dumb kid, I had a glitch that corrupted the file system (FAT) on one of my drives, which caused a directory inside \TEMP to point back at the root. Instead of fixing it, I decided to delete it, which of course deleted everything on the drive. 😜

That's weird. I remember on my 386 back when, there was a similar issue where I had an infinitely recursive directory. Can't remember what happened.

133MHz wrote:

• Melted a bunch of keys on a laptop keyboard by going over them with a hair dryer a bit too near (trying to dry it out).

I did this last week. I spent about 2 hours dismantling a grubby old Mac Pro keyboard and individually cleaned each key. I had a genius idea to dry them - put them in a colander and take a hair dryer to them. Worked perfectly. Except for bending the space bar.

I now have a pristine, but completely useless, keyboard.

Ah I'll jump in.

Dumbest thing I've done on a computer has easily by far and away been my naive donation of personal information to corporations and governments. Mostly over the last 10 years I would say.

Otherwise, hardware wise, I don't think I've ever done anything dumb. Maybe applied thermal paste with a credit card?

I work in IT, so if it's been done to a computer, I've seen it done. 😀

Buuuut one time when I was a kid I put some SDRAM in backwards (don't ask. DON'T) and fried it. And the board.

AidanExamineer wrote:

I work in IT, so if it's been done to a computer, I've seen it done. 😀

Buuuut one time when I was a kid I put some SDRAM in backwards (don't ask. DON'T) and fried it. And the board.

It's hammer time? 😁

Tetrium wrote:

I just reread the title of the thread, it includes things you see others do.

Yup, I included the "someone else do" for a good reason - my younger sister's husband.

Back in the 90's and early 2000's (when he was still single) this guy fancied himself as a computer aficionado; never mind the fact that his knowledge and experience was about the same as mine. Because of this he would get himself in hot water constantly:

- Tried to overclock his Pentium 133 PC to 233MHz. All he got was a loud bang, lots of smoke, and killed his 430VX motherboard, CPU and AT PSU.

- Attempted to flash his work-issued laptop with a RAR compressed .BIN file. The laptop went berzerk and he tried to recover with a backup; about half-way through the battery ran out (he forgot to plug in the power brick). Dead laptop. His workplace gave him another one with a note - "Please retrain in portable PC use and care".

- While configuring the dial-up connection for his new external modem he entered his own phone number instead of the ISP number. The result was a busy signal for a few hours.

- After buying a new power strip for his computer room he somehow managed to plug in the power strip into itself and a desktop lamp on the outlet. Of course only the lamp worked.

At least he settled down after marrying my sister and has stayed out of trouble for quite some time.

I worked for 5.5 years for a local mom&pop chain of computer stores (now out of business) starting as a tech guy about a year after this forum started, plus another several years as an IT sysadmin, so I'm pretty sure I've seen it all.

However, the ones that scarred me for life have mostly to do with bricking.

- The very first optical drive firmware flash I ever tried, it did not recover after reset, and I don't think there was any way to restore it other than replacing the interface board and/or drive. It was my father's HP 8250i. I documented the story here back in 2001. Sadly, it never recovered but happily I sourced an exact replacement from work so that he never found out. 😀

- there was this one guy who brought in an MSI AMD Athlon 64 motherboard into work sometime in 2005 or 2006 inside his custom ATX computer case, he was having problems identifying his PCI sound card. I tested the sound card on another motherboard and it detected fine, but I was able to duplicate his results in his motherboard: apparently it was getting quasi-random results in the PCI ID string - a rather odd behavior I've never really read about anywhere else. "It must have a BIOS problem" I cried, and popped in a BIOS flash program (yes, the correct one, I triple-checked) before the customer had even signed the liability service form. After reboot, it never POSTed again, nor could I get any sort of BIOS recovery to work. Ended up splitting the hardware bill for a new motherboard personally with the customer, sourcing an exact new-in-box replacement off eBay, because not having properly processed in the customer for service (and wait times) and doing it on-the-fly on the bench that Saturday while he waited (he was a nice kid and I'm an old softie for sob stories) meant I was on the hook... oops. Learned that lesson the hard way.

Other than that, the stupidest stories of customers are starting to fade. One that stands out in memory is one teenager opening the case of a display model desktop and ripping out a stick of SDRAM WHILE IT WAS RUNNING. I'm not sure what his explanation was anymore but we were all pretty sure he was either trying to steal it or was clinically stupid. 😀

Stiletto wrote:

After reboot, it never POSTed again, nor could I get any sort of BIOS recovery to work.

Oh God that sinking feeling you must have had is like no other -ruining someone else's gear... right there in front of them! Your face must have been on fire! I'm surprised you weren't sick.

brassicGamer wrote:

Stiletto wrote:

After reboot, it never POSTed again, nor could I get any sort of BIOS recovery to work.

Oh God that sinking feeling you must have had is like no other -ruining someone else's gear... right there in front of them! Your face must have been on fire! I'm surprised you weren't sick.

I don't remember exactly how I felt but we routinely would go Office Space on our own broken equipment behind the shop after hours to blow off steam 😁
Furthermore I am a bullshit artist so I probably spun some excuse that it was actually his fault but we'd do him this favor this one time, just gotta wait for the parts to come in 😉

I'm still mystified as to the actual problem. How do PCI cards start misidentifying? What exactly is going on there? Weird.

Stiletto wrote:

I'm still mystified as to the actual problem. How do PCI cards start misidentifying? What exactly is going on there? Weird.

I have a crappy laptop whose mini-PCI slot consistently messes up the device ID number in a deterministic way, like a stuck bit or two (i.e. 0x2222 becomes 0x2220, if I remember correctly), so all cards become "Unknown device". It's from the RoHS era so I'm inclined to think it could be bad solder balls under the northbridge or something like that.

Touching a bnc network cable while traffic was transmitted through the network.
Plugging an Amiga joystick into a serial port on a PC (dead machine).
Accidentally encrypting an entire hard drive on a 286 and forgetting the pass phrase.
Finally pushing reset button while boot-n-nuke was securely wiping a 80 gig drive,
resulting in the master password in it's chips being set.
Took me 4 years to figure out how to unlock the drive.

- Helping someone to install a sound card. Open the case. The sound card is screwed in but sitting above the PCI slot, not in it.

- The CPU had burned out on a system I was working on. The heat sink was one of these. The pins were pushed tightly into the motherboard. But prying it off revealed an untouched swirl of thermal grease. I still don't understand how that's possible.

- About once a year my father will have me look at his computer because it's acting funny. I move whatever he has sitting on the keyboard and it's goes back to working fine.

133MHz wrote:

I have a crappy laptop whose mini-PCI slot consistently messes up the device ID number in a deterministic way, like a stuck bit or two (i.e. 0x2222 becomes 0x2220, if I remember correctly), so all cards become "Unknown device".

I have a CardBus eSATA card which suffers from the same problem, except it's not even deterministic. Sometimes one of two bits may flip, sometimes it's right. In my case it's a faulty card, though, not a faulty adapter.

If it's truly just the device ID, you can work around it by force-installing a driver with a non-matching ID. Windows will let you do that, if you insist. In Linux it should be possible to do by recompiling some module. However, usually the problems will not end there. In my case, the CardBus card fails to work properly even when it's recognized correctly. It gives abysmal transfer rates or locks up entirely.

JayCeeBee64,

Oh, your poor sister... I hope he's at least handsome and caring.

SKARDAVNELNATE wrote:

- About once a year my father will have me look at his computer because it's acting funny. I move whatever he has sitting on the keyboard and it's goes back to working fine.

Clearly a case of "Error-40". That error sitting 40 cm's in front of the computer. 🤣