Slashdot Mirror
Pushing a CPU to Heat Death, Intentionally
sdougal writes "This site is showing a Pico-ITX board running Ubuntu with no cooling whatsoever. They even let the public guess how long it would last: 'Last week thousands of you placed bets on how long the new Pico-ITX board from VIA, the VIA EPIA PX5000EG, can last without any cooling whatsoever. An ARTiGO Builder Kit was offered as the grand prize. Yesterday afternoon the voting stopped and the Naked Pico Challenge started in earnest. We simply loaded up Ubuntu 8.04, set it to work playing an mpeg-4 video and then removed the heatsink, leaving the CPU and VX700 chipset bare to the world. We recorded the event here in this video and set up a live video stream so you punters can keep a watchful eye on the PX5000EG as it works away.'"
It was an expensive lesson in the importance of the heatsink.
Of course, many of us can remember back when CPU's didn't even need heatsinks. My first build was a 486SX with a zif chip slot and no CPU cooling--hard to believe now.
SJW: Someone who has run out of real oppression, and has to fake it.
The rule of thumb among engineers is: One square inch of flat aluminum surface will dissipate one watt at room temperature and rise about 20 degrees Farenheit.
A CPU chip with 900+ pins run a bit cooler as it's a it more than one square inch if you an include the substrate, and a certain percentage of the heat will conduct itself down the pins.
Ah, the point is to demonstrate how efficient the CPU is. Fair enough, I thought this was just breaking stuff for no reason.
Destroying things is fun, especially done with unorthodox methods.
As our way of thanking you for your positive contributions to Slashdot, you are eligible to disable Slashdot 2.0.
They should make the mp4 hours of video of hardcore pornography, and we can all make bets on what the final frame that it shows before locking up and shutting down will be about. Blowjobs, anal, AtM, Bukkake, fem domination, tentacle sex, etc. It will bring more people to RTFA and WTFS (Watch The Fucking Stream).
slashvertisement. There I said it.
VIA showing off their board, offering a VIA-equipped toy to someone, disguising the entire thing as a geek event and plastering it on geeky sites. Gee, that sure is great news for nerds, stuff that (doesn't) matter...
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
They've been doing this with aircooled VW engines for probably 50 years at shows and races. Pull the fan belt, drain the oil, and put a brick on the accelerator. Everyone pays a buck to bet on the time, and with any luck the engine explodes spectacularly, much to the crowd's pleasure.
Yet again, "on the internet" somehow makes it original...
"As God is my witness, I thought turkeys could fly." A. Carlson
If a CPU is going to crash or go up in smoke after heatsink removal under load it will do so within 30 seconds. Since it hasn't done so yet and considering it's a 1W energy efficient CPU the only effect should be a reduction in its longterm lifespan (maybe it will only run 2 years rather than 8). I don't see the excitement here, until they take a hairdryer to it which they say they will do after two weeks. That should be interesting.
https://dalgamotor.wordpress.com/ - Elektronik beyinlere ozgurluk asisi (Turkish)
A server with all heat sinks removed, and then linked to on the front page of Slashdot. ;nspb
Will it melt?
The meme is dead, long live the meme!
Like the ole Timex watch that "took a licking and kept on ticking" my desktop box, an ancient AMD Sempron 2600+ with a VIA chipset, unknown to me, lost its power connector to the CPU fan, which I only discovered by accident when replacing a hard disk drive. The CPU was hot enough to scald my finger, but neither its performance nor its stability has suffered one bit.
Of course, the heatsink was still connected. But the Sempron was IIRC most definitely NOT a low-power cpu.
Yes, I reconnected the CPU fan. But at least I know my sh*t can take the heat.
No video is available ;o{ .
that this is an experiment. They already know that the device will run indefinitely. No company would do a media event like this that would shed bad publicity on their product- except Microsoft, LOL.
Sorry to nitpick, but doesn't the term "heat death" usually mean death by maximum entropy (i.e. no heat), and not death by heat?
I read Usenet for the articles.
We had a headless linux server that one day started beeping constantly for no apparent reason. With every intention on fixing it, after a couple of weeks of it still running ok, we just assumed the speaker had died so just ignored it (the server room being sealed away as it was). Then one day we had to move the servers to another room, went to pick the machine up, and "Jesus! This thing is boiling!".
:)
It was some ancient AMD chip that we literally couldn't buy new fans for any more, so we just snipped the speaker cable and let it carry on.
Naturally, the Linux guys claimed if it had been Windows, we'd be looking at a dead server at this point in time
throw new NoSignatureException();
How long before we see this up on www.microsoft.com/getthefacts/ with the headline:
"Linux will set your computer on fire."
You have been warned.
Genesis 1:32 And God typed
I mean, I can run for several hours without a heat sink or a fan.
The Via is a VERY low power processor.
Since its one of the 1 GHz processors in the board, TDP is 5W.
Depending on what power-feedback is involved, the processor might actually just go "I'm overheating, throttle back" and drop down to say 500 MHz at 2.5W or so. The MPEG decoding shouldn't even take too much power, since the CN700 chipset includes hardware MPEG2 decoding.
As a bonus, the box is OPEN, which improves the cooling.
Test your net with Netalyzr
A few years back, I was troubleshooting a problem on my desktop. It had a Duron 800 in it. I got tired of putting the heat sink and CPU fan back on every time I made a change, so I figured, what the hell, how hot can it get in the time it takes to try and boot. It made it through the boot fine. I mused "Works great! I bet it doessn't even get that hot. Wonder how hot it is?" With that thought, I reached in and touched the top of the CPU. It was so hot that it instantaneously branded the text and logo etched in the top of the chip onto my thumbtip, before I could react and yank my hand back. For a few weeks, until it sloughed off, it was readable in reverse on my thumb...taught me new respect for the current consumption & heat generation capabilities of CPUs.
There is no God, and Dirac is his prophet.
In open air, with no fans blowing air PAST a hot object, it will cool much slower than inside an enclosure where air is brought to the object and is actively exhausted.
This isn't readily apparent in most modern equipment because hot components have their own active cooling, and the ambient air is cooler outside the case.
However, if I turn up my 3-speed 120mm case fans to Max, as opposed to Min, my CPU temperature will drop below what I am able to achieve outside.
But that is only possible when the wiring has been carefully managed to avoid heat traps.
If I re-encode a movie I get: Do I care? Not really. Been like that for 3 years now. When it dies I'll swap it for a less powerful CPU and go totally silent.
How long will that be though? 2 years? 3 years?
which is totally what she said
ew... I can't quite see why anyone would ever want to watch that. I mean, I know some people are weird and would generally count myself as one of them.. but that video sounds almost as bad as having to judge your country's possible entries into the Eurovision song contest..
which is totally what she said
and no fan I had an Athlon 1000 about four years ago almost catch on fire if I just didn't happen to come home for lunch. The fan failed for whatever reason and the CPU got smoking hot and started to burn all the dust around the MB. The only reason I notice is that the whole house smelt like burnt dust/hair. After unplugging the power I touched the fan about 5 min later and burnt the hell out my finger as it accidentally touched the heat sink. I had a red burn mark on my finger. I can imagine who hot it really was when it was still powered up.
by TheSpoom (715771) Uncaring Linux user here. I have nothing to add to this but please continue. *munches popcorn*
The case fan is on the heatsink. So closing THIS case would greatly reduce the cooling, as hot air would be trapped in the case.
Test your net with Netalyzr
But the kermit the frog reaction video has to be one of the funniest to come out of it.
Since when was running a 1 Watt CPU without a heatsink regarded as a challenge?
http://www.genesi-usa.com/efika.php - plug plug
That system runs at 1W@400MHz, although has no video-accelerating northbridge to add to the heat, it can play that MPEG4 video just fine (I am playing something similar now). We've designed it so the 2.5" hard disk actually sits about 5mm from the top of the CPU - if anything we're making cooling harder, and there is NO heatsink. The CPU does NOT power manage into SpeedStep style states - it just runs at 400MHz or "standby" (where it cannot run code until an external interrupt).
It runs fine. Mine's been on 24/7 for nearly a year, barring moving it around and connecting it up to things like new hard disks, changing power strips or measuring the power it uses. It never overheats.
What's the challenge meant to be? Just how crappy Via's chip needs to be that it CAN'T run at 500MHz on a 90nm process, and do without a heatsink of some kind?
Back in my Transmeta days, I set up a demo doing exactly this...one of our CPUs playing movies without a heatsink, head-to-head with a comparable Intel and it's (hot) heatsink. It lasted all day, and only got slightly warm. Still, I always expected to get burned every time I stuck my finger on the die top for the reporters. Poor, poor Transmeta. :)
-g.
Hardhack is short for hardware hack.
And I've had my both of my pumps in my water cooling loop die at the same time (they where shitty Thermaltake Bigwater ones) on a P3 1.2GHz Tualatin based home file server.
Still as it was unattended, the sever was left on the whole afternoon. I only realised it wasn't responding in the evening. The power was still on.
The heat of the processor evaporated the cooling liquid, and melted the plexi top of the CPU block.
And I still burned my finger when detaching the remaining copper block from the CPU even after a couple of minute after shutting down the power.
But even after all that cooking, once I replaced the cooling bloc and installed redundant pumps from a real brand (2x Lain DDC), the same CPU and motherboard started happily without complaining.
Now that's something that you won't be seing with more recent and fragile CPUs.
But given that VIA CPUs don't eat big amount of watts and don't generate high amounts of heat, it is still possible the their pico-ITX board will similarily survive heatsink-less cooking. The system will crash, but the CPU & ITX may be recoverable after letting cool down.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Actually, the newer CPUs will handle it BETTER than that P3 did.
Anything older than that P3 would have cooked. That P3 went into a frozen state to save itself. The P4 and newer underclock themselves until they're running cool enough, and freeze if that's still not cool enough.
Exploding CPU