- Feature Articles
- CodeSOD
- Error'd
-
Forums
-
Other Articles
- Random Article
- Alex's Soapbox
- Announcements
- Best of…
- Best of Email
- Best of the Sidebar
- Bring Your Own Code
- Coded Smorgasbord
- Mandatory Fun Day
- Off Topic
- Representative Line
- News Roundup
- Editor's Soapbox
- Software on the Rocks
- Souvenir Potpourri
- Sponsor Post
- Tales from the Interview
- The Daily WTF: Live
- Virtudyne
Admin
I had to bite my tongue when visiting my wife's family for the first time. Her uncle (who supposedly worked in tech support) was explaining to her brother (who was trying to get into tech support) how some virus going around would infect the TCP stack in the firmware of the network card. From then on the network card would both attempt to replicate itself on any machines you connected to or who connected to you (apparently being compatible with every NIC on the planet including 802.11b/g cards) and acted as a spam relay zombie.
I just stared at him open-mouthed. He said something to me along the lines of "You really should install a virus scanner to make sure that you can't get infected byt this." I excused myself, went upstairs and laughed my ass off for a full 20 minutes.
Admin
That sounds like my Dad (who is not very computer literate, although he is getting better at using computers). He also attempted to format our hard drive for a re installation of Windows XP by going to Start, Run, typing 'format c', and pressing Enter.
(On a side note, I'm able to play a side-scrolling game designed for Windows XP on Ubuntu with Wine, but I am not able to play it on Windows Vista)
Admin
Thankfully the only really dumb thing I've done in all the years of building my own PC is fry a math-co-pro. Yep, eagerly awaited my Intel 387 in the mail for weeks (here we come flight sim!), popped it in, powered on. Fweeeeeeeeeeeeee! Pop! and the magic smoke escaped.
Back the in day, Intel didn't bother making their chips and sockets idiot proof with bevelled corners, unique pin layouts etc. You could put the chip in any which way. :(
Intel at least was kind enough to issue a replacment ("It must have arrived busted!").
Admin
In hardware terms, a buffer is something you can plug your "weak"/low current voltage signal to get a "strong"/high current version of that signal.
They are used all the time, because your tiny logic CMOS just doesn't have enough current to push up the voltage of the entire wire/trace. So, every pin on an IC is connected to an I/O buffer, which can supply enough current to push the voltage of the wire it is connected to up/down quickly.
(As a side note, what programmers call "buffers", hardware people call "FIFOs" -- first in first out. They are widely used components for interfacing digital circuits running on different clocks.)
Buffers are only output devices. The tech who said "it's ok, they are buffered" was confused about the terminology. (Obviously he was confused, since the board wasn't ok, it got fried).
What he probably meant was "tri-state". Tri-state is when you have several different circuits that all might want to set the voltage at different times sharing a line. For example, on a bus. In that case, the output has three options: voltage hi, voltage lo, or hi-Z. In hi-Z the output is just like an open circuit.
Tri-state also is an output only thing. There is really no protection against putting 12V onto a TTL pin.
Admin
I agree - Short Circuit was an AWESOME movie. Johnny No. 5 is alive.
Admin
I love hanging around the computer departments at Staples and Future Shop for precisely these reasons. They're good for hours of great entertainment. More than once I've had to run out of the store because I knew I was going to fall on the floor laughing.
Admin
tiny pc used to do this and it was the bane of my life fixing peoples systems, but it did earn me easy student $$$
since when was looking at hardware to see if it was fried stupid? there could be burned out chips or leaky capacitors giving an obvious visual clue as to why its not working ...
Admin
Reminds me of the time I guy asked me to fix his coffes cup holder... Ha Ha - He was referring to his CDROM DRIVE!!!
Wot a larf !!!
Admin
3: really isn't a WTF. One way to detect if a video card is a fake is to check the RAM on the video card. Yes, video cards have RAM also.
Admin
Disclaimer: I'm a hardware/software EE working on embedded systems.
When a RAM chip overheats, frequently it will discolor the plastic. It also tends to make the DIMM's label fall off, if there is one.
RAM tends to be one of the least forgiving chips when you have voltage surges or power supply failures.
Aside from the fact that I've never heard of a "YES gate" (maybe you mean "wired-AND"?), I think you're thinking of digital inputs. This is a case of a power supply being connected to analog inputs. The buffers are likely unity-gain op amps, or a similar analog conditioning circuit that cleans up the signal for the analog-digital converters on the sound card.
What the tech got wrong was the fact that sound cards can't stand 12 V on their inputs, and might not be able to deal with 5 V either. For reference, audio signals are generally 1 V peak. There are some protection diodes that can dissipate a little energy from a static charge, but not an overload from a direct short to a power supply. That 12V probably went through the protection diodes to the power rail, (almost) shorting the 12 and 5 V rails together until something burned out and broke the circuit.
Introductory EE classes tend to leave out details like power supply rails limiting just what you can get into or out of a chip. They're weed-out classes. Once someone is past those classes, then we spend the time teaching how things really work... and break.
Admin
You're new to the internet, aren't you? Assuming is what we do. Makes things more interesting/fun. Besides, allows you to make an even bigger ass of yourself when you post. :)
Admin
I had a great one from a 'tech' company i got out of Yellow Pages once - the conversation went something like this:
Me: Hi, i need to buy a firewall Them: A what? Me: A firewall, you know? Them: Ah yes, sorry, we sell IT equipment - you should try Tommy's House of Fires for something like that.
I'm not sure whether they're still in business...
Admin
(2) Maybe the encoder automatically switched to low quality because there was not enough RAM in the machine and swapping was not an option.
Admin
Line level audio inputs are usually AC-coupled and have a ~10 k resistor in series, good luck trying to get through that with 12 V DC.
Admin
Geez... That must be why my sound doesn't work. I wasn't using high voltage AC......
Admin
Was your uncle was using a KillerNIC (http://www.killernic.com/products/nic.aspx) with a hacked in old 2.4 kernel that could be exploited or something?
Admin
Woah. Scathing comment there. Some of us do have hardware backgrounds and it still took a while to work out what he meant. However, regardless, it seems that this card was not buffered, nor, like newer ones opto-isolated. Not a nice thing to do to a sound card considering it would likely have been receiving +12v through the Left channel and +5 through the right...
Admin
Having been a tech for longer than is healthy, I can honestly say I have:
Ah, the joys of techdom!
Admin
Actually, in #3; He was probably looking for hairline fractures around the RAM on the video card.
Years ago it wasn't uncommon to take a motherboard or card and hold it to a light to look for fractures or "opens".
He probably just didn't communicate it effectively.
Admin
I don't know the pin outs of the sound card cable. Is it possible that +5 or +12 was connected to the cable's ground and it blew out a trace?
Admin
It has left and right signal on the ends and 2 ground pins in the middle. The card probably had only the range on the analog input to only account for the correct input (the sound input). Without some protection, everything gets fried, instead of simply rejected and reported as the highest value at that bit resolution.
Admin
There are two ways to look stupid:
Admin
I once worked with a guy who bought an upgraded hard drive for one of the office computers (this was in the days of MFM and RLL drives -- ancient history).
He noticed that the colors on the drive connector did not match those of the molex connector. So what's he do? He cuts the keying off the molex connector with his pocketknife, and jams the connector in anyway. Yellow wire has to go in the part of the socket with yellow paint, right?
He returned the drive as "defective" to the store.
Admin
It's not a 'useless certificate' if he's making more money than you because of it. It serves its purpose of impressing HR and getting you more $$$. =)
Admin
I once worked in a smal computer shop, a woman brought her computer back because it was 'unreliable and kept crashing'. As she turned the mini tower case around and placed in on the shop counter we saw that one side of it was entirely covered with colorful fridge magnets.
Admin
Original poster here. I could have been a little more complete. The buffered guy had just started his education, he thought he knew a lot more than he actually did. As I recall, he didn't work long with us, he wasn't the sharpest tool in the shed.
Admin
Admin
As I recall, it was a Sound Blaster 16 plugged into an ISA slot. Not the original SB16s, but the ones that supported Plug and Play in the Win95 era.
Admin
That kind of stuff was standard practice for the old packard bells right?
Admin
Hard to say - we told her to take them off and sent her on her way, and she never came back. I would have thought that the case would have kept some of the magnetism though, but maybe not enough to cause issues.
I remember once putting a wooden DIY external disk enclosure containing two 5.25" 157Mb ESDI harddisks on top of the PCs PSU with the case open (ok it was a temporary measure), and that caused all sorts of odd disk issues until I moved them away a bit (I specialy made really long IDC cables). I only did that because they were both in the case but they got so hot my single speed mitsumi CDROM drive (with propietary ISA controller card) which was above them would frequenty refuse to read disks.
Oh the joys of DOS 6.22 and MSCDEX.exe...
Admin
A friend of mine was surprised to see that her internet did not work anymore. After hearing her explaining the case, it simply added up to:
I could not convince her about the fact that it was very unlikely her setup would've every worked.
Admin
I think that some here are forgetting that it's not just an incorrect voltage that fries parts, amperage can be just as much of a problem.
Admin
Voltage and current are related. You can't have one without the other.
If you have a "safe" high voltage, the voltage will drop instantly (say 10,000V of static from your finger to a doornob - that 10,000V zap only lasts for a tiny fraction of a second).
So, to maintain an ongoing voltage of a certain amount you just need to supply a certain level of current. If you push a certain amount of current through a circuit on an ongoing basis, you will develop a certain voltage.
Admin
Wow! That is a pretty shallow viewpoint. On one hand, sure web development isn't as powerful as low-level machine languages and hardware manipulation but without the "web" where would we be right now? I guess I wouldn't be replying to your post in my wonderful Firefox web browser which was written by some really stupid non-computer scientists to interact with some protocols and languages that don't deserve to be in the same sentence with computer science. Hmmm, I'm sure the Mozilla foundation or Google or IBM or Microsoft or Yahoo or... would agree with you.
I'm fairly confident that there are some pretty damn intelligent computer scientists that have focused on web development. Am I way off base here? I bet your reply really stroked your ego "thanks to the pull-up resistor which prevents the ground from being directly connected to the Vcc."
You may say "This guy is just a stupid web developer that doesn't know a resistor from a capacitor." Well, to be honest I don't like web programming that much and I'm not into hardware but I thought your perspective sucked so I felt compelled to reply. Have a nice day.
Admin
I'll bet that the horse would have told you it's buffered: http://www.youtube.com/watch?v=TfZCF22BrKw
Admin
Wow, way to be an asshole about it. By the way, you're wrong.
Admin
His statement about them being buffered is perfectly reasonable. Sound cards typically have hardware buffers between their working guts and their inputs and outputs.
They're designed for precisely this purpose -- to protect the guts from bad things that might come in over the wires. A more typical example might be static electricity from someone touching the end of a connector.
One would typically expect the buffers to absorb transients and perhaps to blow if connected to power. Components in the buffers (such as series resistors or resistors to ground) can overheat and fail. But one wouldn't expect a damage to one input to affect either the working guts or the output circuits.
If you thought he meant software buffers, the WTF is that you don't know anything about hardware and are working on computer hardware.
Admin
that reply was for dams not Mopper
Admin
Actually, now that I think about, a likely scenario would be that the inputs were AC-coupled but diode clamped. Connecting the 12V supply to the diode clamps could have raised the supply voltage to nearby chips from its normal 5V up to much closer to 12V. That could have fried any number of things.
Diode clamping should usually be done through a series resistor, but it isn't always. Designing input buffers is a bit of a black art.
Admin
Um...
Hate to break this to you guys, but the EE wasn't too far off. AFAIK, "buffered" is a fairly common hardware term meaning that the inputs to a circuit are protected and/or isolated from dangerous voltages/currents in some other part of the circuit.
Obviously in this case they weren't protected enough, but they probably should've been.
The WTFs here are the guy bringing in the PC and you implying the EE is an idiot.
Admin
I must say I, like the subject of the WTF, would have expected that the inputs on a genuine SoundBlaster to have been well enough buffered not to have been damaged by that. I am really surprised that it also took out the ISA slot! Still, 12V, at a very low impedance, is a lot for an internal line-level audio interface to have handled. Seems like they ass-u-me'd that someone handling that input would do so with a little knowledge....
Admin
Card glued in with thermic gunk This is actually common practice. Open a lot of PC's and you'll find some where different components have been glued in with thermal. It's a cheap trick when you have unreliable connectors... Makes it hell when you have to swap out one of those components though.
overheard "get more ram to compress better"? Well, if you didn't hear the whole conversation, maybe the guy was complaining that when he wanted to compress at better quality, the machine would slow badly, and that he was forces to compress at low quality. A RAM upgrade COULD be a good option...
Holding card against light to check if RAM chips are ok. Quite often when a card (ram stick, motherboard, graphics-sound card, whatever) "dies", the culprit will be that one of the onboard chips itself has died. Sometimes, when this happens, the chip short-circuits internally and itself burns. This burning MAY show through the plastic casing with some cracking, or some sort of speck which when examined attentively indicates the faulty chip. A good way to see this is to hold it in such a way that the light "bounces" off. It certainly won't allow you to fix it, as long gone are the days where you pry out the chip, but if you SUSPECT that the card is dead, and then you actually see that it has a VISIBLY dead onboard chip, it will give you peace of mind that you're changing the right component. So that tech actually did know what he was doing.
Admin
I was a lab assistant for Introduction to Computer Engineering at my university for a few semesters. Early on in the course the lab project involved wiring up CMOS gate chips on a breadboard to make a basic circuit (and then a very complex, object lesson in "This is why we use Verilog from here out" circuit)
Each lab station several molex socket which they'd get a plug for with alligator-clipped wires on the 5V and ground wires to make powering the circuits convenient.
One of the groups couldn't get their circuit to work, though they were sure the wiring was right. A few quick tests showed me that they had one of the most common problems- a bad row on their breadboard- specifically the one where the connected everything to the 5V line.
I adjusted the wiring to a good row and clipped the power back on for them.
Then was nearly hit in the face (at 6' 4" that's no mean distance from the bench surface) by a fireball shooting out of the CMOS chip they they were trying to connect.
Sure enough, further inspection showed that they'd plugged the molex connector in upside down. The first time around the breadboard had saved them by acting like a fuse. This time it didn't fail sat enough. (Though that row was bad afterwards aw well)
I managed to hold onto the chip (with a big crater in it now) for a couple years before it got lost somewhere when I was moving.
Admin
What part of performing a visual inspection do you find out of the ordinary? You can see blown capacitors, you can see arc traces from shorts, you can see melted solder, and you can see burn spots on overheated ram chips. Sometimes you can even see the gouge marks where some dweebs magnetic screwdriver hit the board. Next time you see something that looks funny, pay attention, you might learn something.
Admin
I had an ethernet card fail after a PSTN line was plugged in to it. Pins were fine, but it never worked again.
Admin
I would think that high-pass filters in the card would take care of the ringing voltage, which is 40V AC at ~15-70HZ.
The loop supervisory voltage present on an FXS port, however, might be a bit too much for it (an additional 48V), when combined with the ringing voltage, but simply plugging an ethernet port into an FXS port won't kill an ethernet card, if the port is single-line and wired properly for loop-start signaling (generally what you find on an FXS port).
Ever held on to a telephone wire pair while it's ringing? You might be shocked at the amount of fun you'll have performing that experiment...
Now I'm going to have to test out the whole ringing voltage killing an ethernet card theory, to add to my database of possible troubleshooting situations. Thanks. :)
Admin
Just curious, do you have car bodies up on blocks in your front yard? :-).
Admin
Admin
Am I missing... an eyebrow?
Seriously, CMOS chips can do that much?
Admin
True, as a graduated EE myself, i probably would have wondered the same thing, wether the inputs on the card are protected against overvoltage...
I dont really see a WTF here, assembling a PC largely boils down to trying to figure out which plug fits where (and you could build a pc largely just by trial and error), and if the guy was an EE student, then his remark is perfectly logical