Slashdot Mirror
The Report of My Thermal Death Have Been...
A reader writes "Not too long ago, Tom's Hardware posted a video of the grisly events that take place when the heatsink is removed on an AMD Athlon MP 1.2GHz in an attempt to show that the chip has inadequate thermal protection unlike the Pentium 4. Apparently, this is not the case. This new video, which looks like was done by AMD, shows the system continuing to work when the heatsink is removed. Even 9 minutes of Quake3 without the fan operating wasn't enough to destroy the processor. So who is right? It's in AMD's best interest to show that their product doesn't disintigrate under extreme conditions. " Update: 10/30 14:11 GMT by H : Note that it was Terry 'quad3d' Wang that actually did the video - not AMDZone.
Of course AMD is just trying to make their product look good.
This is a moot point, because who is foolish enough to run a system without a heatsink and fan.
Sure a fan can fail, but there are monitoring systems that tell you when this happens...
You can run Quake 3 for 9 minutes on an AMD without a heatsink. AMD was probably behind that test. After all, it is in there best interest to prove their product's reliaility.
Do we speak english here at Slashdot anyore?
I am not less perfect than Lor
Tom's Hardware is generally known for the accurate reporting and trying to keep the playing field level in tests.
This one should be pretty easy to test...Someone just needs to have the resources to risk frying a few dozen processors.
I was a little surprised when I read the results on Tom's Hardware...I would have expected they could see this could be a major issue for AMD and would have run even more tests to make sure their results were accurate and not a bad batch or something like that.
Hate to see Tom's take a credibility hit on this one, so it will be interesting to see how this one unfolds!
The video on Tom's Hardware Guide can be found here.
I'm looking for a HEPA media filter for my TV. I'm alergic to reality shows.
Why don't the chips come with built in heatsinks? The heatsink could be fused right to the core to provide the best possible connection. It would kick up costs, but you never need to worry about heatsink failure. My nightmares would finally stop.
Later,
Affe
I think this post is rather naive, I would put more trust into an independant party (toms hardware) testing equipment than the very people who manufacture the equpment.
the video is a .wmv file .... forget about playing it on linux
I bought an Athlon 1.4, and i screwed up the heatsink (boy, them things are a bitch), and I melted it, just under a little use.
$1.5 grand in mobo, processors, memory down the tubes.
Yeah, I'm an idiot. But Athlon are easier to fuck up than Intel.
I'm not an American, we don't do class action suits here. We have real consumer protection laws
i am endorsed for the carrying of dangerous goods, please be giving me your depleted uranium
I had the heatsink sorta fail. We think that the fan stalled, and that led the processor to heat up. It ended up killing my motherboard, but oddly not my precessor. I bought a new motherboard, and got a big heatsink along with some extra case fans, and now the system works fine.
I'm impressed that that heat could fry the mobo but NOT The processor, it's sort of weird actually. But there are a whole lot of things on most motherboards right next to the processor, and on mine they all looked slightly burned.
...but this new video shows the same results as the TH video! After about 2 seconds with the heatsink off, ths system SHUTS DOWN.
Only when the heatsink is left on, but the fan power is disconnected, does it last 9 minutes.
The thing doesn't last 9 minutes with the heatsink off, crikies....
AMD could have done this with a customized bios. The palmino core contains the PowerNow power safeing technology that could alter the multiplier and core voltage while running. ;)
AMD could have used that technology to emulate pentium 4 like cpu protection.
Or maybe the video is just a fake
Jan
Not true..Pick up a copy of MPlayer (look on freshmeat). Plays WMV/DivX/MPEG/MPEG2/ASF/AVI and others with no problems at all.
Here's the original Tom's article.
Here's the text of the new article direct from the source:
Thanks to the millions of people who e-mailed me about this.
Do we title this: We TOLD You So!
or do we title it: Maybe we were right about him?
or do we title it: AMD Won't Burn a hole in your wallet, or your motherboard?
or do we title it: AMD Slaps Around Little Tom-Tom? as Van Smith put it?
Well whatever you want to call it, there is a new video out on the internet showing what REALLY happens to an AthlonXP (or MP) when your heat sink falls off, or when your fans fails. Now I'm not saying that Tom Pabst is completely wrong, but let's just say this makes his testing methods look a little 'suspect' at best. Here's a quote from Van Smith:
A video has appeared on the Internet countering a huge dose of FUD my former employer inexplicably dumped on his readership a couple of months back. The new video, with AMD credits all over it, is entitled "How an Athlon(tm) MP 1.2GHz Really Copes with Heat Emergencies." The piece demonstrates the AMD Palomino Athlon subjected to brutal circumstances such as heat sink removal while playing Quake III and boot up attempts when a CPU cooler is not attached. In all scenarios, the Palomino comes out unscathed. A similar though much less thorough test came out with unsurprisingly different results at Tom's Hardware. Ouch! Looks like dispensing bad medicine can result in a mouthful of looser teeth. Good job Ben & Joe, perhaps you can also give THG a crash course in analyzing computer technology.
The video also presents the much more realistic situation when the CPU fan fails. In that case the Palomino continued to play Quake III for several minutes before shutting down. Again the chip was undamaged.
This video may look like it's from AMD, but I'm pretty sure it's not, even if it makes a great case. The original download site appears to be down, so I mirrored it onto AMDZone right here.
Other mirrors: Mirror.
Update: One of our readers e-mailed me to give me his first hand experience with the thermal control capabilities of the AthlonXP:
I installed my amp1800 (sic) with the heat sink rotated 180 deg. and after 3 hours of trying, incessantly, and not being able to boot I found the problem, I rotated the heat sink and all is fine. I must say, I was sweating bullets when I found the problem.
No fried chip, no smell of burned silicon. Looks like the thermal diode is working.
(end article)
~Aaron.
student of animation and the fine arts
I can attest to an Athlon 1.4 GHz @266FSB not burning when booted with an improperly fitted heat sink. It failed at / didn't make it to POST. However, fitting a heatsink properly, the machine worked fine. Tests in bios, when I made it, showed the system getting up to 75oC before locking (hard!). Much research into this, prior to refitting my heatsink (using acetone to take off heatsink thermal pasted and replacing it with a good silver compound), showed that Athlon 1.4's do get up to 60oC at operating temperature. Mine operates around 50oC, with an ambient temperature of around 27oC. Indeed, these machines heat the room they are in.
This is utter bullshit. In the first clip, dude pyulls the fan off. OK. Well it cut away in about 2 seconds. What happened then? Most of that answer is obvious, the freakin' game died (and so did the box from the looks of it)
In the second clip, The game died and so did the box.
In the third clip, The game died and so did the box.
you get my point...
Tom's video is by far more realistic...
I'm not a prophet or a stone-age man,
I'm just a mortal with potential of a super man.
I can say that the T-Bird won't always just burn to a crisp like Tom's said it did. I had one running without a fan (the mobo fan plug just up and stopped working) and it wasnt until about 20 minutes later that the machine locked up from the heat. After I fixed the problem the machine went on to work fine again with the same chip (1.4 Ghz T-Bird), with the fan plugged into a different connector.
Last week I was inspecting one of my athlon boxes, which had been running non-stop for quite a while. To my amazement/horror/confusion I found the cpu fan had come unplugged. That explained why the computer had been awfully quiet the last few weeks. Though it didn't explain its uptime.
Please note, do not touch an athlon heat sink when the fan has not been going for weeks, it'll leave burn marks.
I mean, isn't that what you want a system that's malfunctioning to do, shut down? If it didn't shut down, then it would continue to heat up and roast your cat which was using your tower as a space heater.
Maybe I'm missing the point...
F-bacher
James Tiberius Kirk: "Spock, the women on your planet are logical. No other planet in the galaxy can make that claim."
AMD is just karma whoring. I mean, imagine a Beowulf cluster of AMD processors, _without_ heatsinks:
Pros: Could replace my propane furnace if used to heat the house
Cons: The electricity cost would likely be even more than the propane
This space intentionally left blank.
If you read the toms hardware review, it said the athlon MP/XP thermal diode were capable of detecting no more then a 1 degree/celcius per second change and no more ... they've carefully engineered this "test" so the rate of change is below that threshold... toms test was worst-case and thats what you have to plan for.
Free Techno/Jazz/DNB/MI Music by guys obsessed with monkeys!
IIRC, the "exploding Athlon" problem was at least partly to do with a BIOS that could only deal with temperature changes of 1 degree/second, and simply lost the plot when the change was more dramatic.
...
If that was the case, a different BIOS should make a world of difference. Of course, I could be misremembering the details and I can't find the original article now
Wasting your time since 1997.
PIV's heatspreader is a step in this direction. I don't think heatsinks will be completly connect with the cpus because there are too many different markets for heatsinks.
There are the "consumer" PCs where a heat sink should be as cheap as possible, "business" PCs that should be silent and reliable but cost doesn't matter that much and finally heatsinks for overclockers that should be very powerfull.
Jan
Come on, surely this is an open-and-shut case of corporate rebuttal of bad PR. Not that anyone expects their proessor to tick nicely along without any kind of cooling aid, surely?
Must have been a slow news day at /. :)
Security through promiscuity is no better than security through obscurity.
When the Tbird 1ghz came out, my best tech fried two of them. Turns out that if you don't use the AMD recommended make and model cooler, and preferably scrape off the thermal pad and put a drop of professional thermal compound on them, then make sure they are sitting nice and flat on the cpu so one side doesn't get hotter than the other, in less than 1 second they go boom. That was a 1gig, so am I to believe that the 1.4 won't go boom? I don't know. I know that the 1.4 runs a little cooler than those first 1gigs did...
Umm.. excuse me, but how am i supposed to play a ".wmv" file - and what is it?
stuff
here with videos on the 5th or 6th page.
Probably one of the worst mistakes I have made in the past few months was slapping the heatsink from my K6-2 onto my Athlon 1.4 chip because I was too impatient to wait a couple of days for the Alpha heatsink to arrive. Whilst placing it on the chip, I also neglected to put any thermal grease on it, so there was probably poor contact and low heat-transfer. The system didn't even boot. I mean, I switched on the power, and it didn't even output video. After swapping out chips a few days later, it turned out that the Athlon was the only thing wrong with my system, and everything has run smoothly since then (I'm posting from it right now).
I would be suspicious of the assumption that it overheated, given that it had such a small time frame in which to do so, except for the fact that, on closer inspection, the material around the processor core looked like it had melted away from the metal and reformed on the base of the chip. Anyway, it was ugly.
The lesson? It takes these things roughly 3 seconds to overheat without proper cooling, and they don't have any kind of built-in monitoring to shut it down before it goes critical. So take this article over at Tom's with a grain of salt (as you should be doing with any news posted on Slashdot).
Is your company running tools written by ma
Who opens the codec?
Your answer: Apple
Correct Answer: Sorenson
Who's to blame for not being able to play QuickTime Movies in Linux?
Your answer: Apple
Correct Answer: Soreson.
Licensing is the problem. Now back to your regularly scheduled karmawhore-a-thon.
My only comments on the video are that it's a little hard to tell what's going on. Did the box shut it self down when the heatsink was removed? why? what was the point of showing that it couldnt be turned on sans heatsink? The running with no fan was nice...but show us some CPU temps/speeds there. if i underclock my box and cool the room down nice and frosty, i dont need a fan either...
The opinions in this post are ficticious. Any similarity to actual opinions, real or imagined, is purely coincidental.
All this shows is a motherboard with sufficent onboard temperature monitoring, shutting down because the CPU is over heating. It has nothing to do with the CPU having any sort of built in temperature throttle like the P4 has (or even the 68040 had). Tom's Hardware and this video are showing two completely different and seperate things. One shows how a motherboard reacts in order to save an amd cpu (this current video), and the other shows what happens if the cpu is left to cope with a heat problem on its own (Tom's tests) without the motherboard stepping in to rescue it.
/. just doesn't care some days... or maybe its just on slow days ;).
I wish people would at least FILTER some of these posts. Its seems that
I am definitly one to test the reliability of Intel w/o a HSF.
I ran once (just to test) my processor w/o a HSF and was able to boot up win2k.. I have yet to figure that one out.
My system won't even get to post (I have an Athlon 1.4 and Abit a7m266)
So what does this mean. Does this mean that AMD has better protection that it just doesn't work? Therefor you get the idea that its not working b/c of heat? Whereas when you use intel, it keeps working, setting it self up for failure..
Hrm. I've fucked up my HSF installations.. but I've never smoked an AMD or Intel.
The idea of built in HSF is OK, except for the fact that now you can't replace it with your own monster (Like a Peltier Cooler, or a monster Dragon Orb [loud ass mofo])
Hrm. yup
------------
Sase
"It's the opposite of that."
Besides, alot can be forgiven of the company that gave me a real math co-processor and "fast" math toys on Hercules graphics for my 4/8 mhz(turbo!) 286.
Nice soundtrack.
Morphing Software
It's nice to see AMD investigate the Tom's Hardware story properly and carry out their own tests. However, disproving a problem is a lot different from admitting one exists. If AMD are out to say Tom's wrong, that's a bad thing. However, if their research is a concerted effort to identify and fix the problem, all hats off to them.
I have a 1ghz athlon (yeah i know, but i can't be bothered upgrading it suffices fine, memory seems more of a problem to me, that's why i have 640mb)
;-) ) basically my machine kept locking up randomly especially after intensive cpu work, it took me about a week to realise what was going wrong (the fan still worked but further invertigation showed it was spinning right slow)
;-)
Anway shortly after buying the motherboard and cpu the fan failed (it was free with the motherboard and cpu, you certainly get what you pay for
and the athlon would continue to work fine to temperatures of about 90 deg C, much higher and the system crashed (and as i say for a week all i did was reboot and put up with it)
the cpu still works fine today (with a 1.4ghz rated fan just in case!) and iss back down to reasonable ~45 deg C max (ish) temperatures.
also if ytou want to go back further i used to run my intel 166 mmx at 266 for about a year, which gave great result until the fan died (standard pentium 233 fan) at which pioint it took the funny business of whenever it entered 32-bit mode that it would kill hard drive partitions (and i mean if you needed stuff back you had to endure 48off hours of scandisk) but it did still work fine and no problems if i put it down to the old 166 proper speed. so i then went out and part ex'd it for a amd 400
right must go now GBA doom is calling me!!!!!
+----------------- | What is the question!
My flatmate was fiddling with the fans, trying to create the perfect airflow for his athlon1200, and he was drunk as hell, so he forgot to connect the proc-fan connector, but the heatsink was still on. The temperature rose steadily to 93 celsius, and I sat there laughing, but he didnt catch on until the bluescreen came after a couple of minutes. He just gave the computer a rest while he reconnected the fan, and turned it on, and hey presto it ran flawless...
-By attempting the impossible we can achieve the absurd..
Too bad AMD chose to use a shitty windows only format, since tom's hardware used the multi-platform divx codec i only have their video to go by. The P4 one looked kinda fake though. The frame rate come back like the exact second the heatsink touched the cpu, i donno if i believe that...
Quite possibly that might have something to do with the fact that linux is executing a HLT when it is idle, while windows goes into an endless loop. Since your computer likely spends most of its time being idle, it doesn't generate much heat compared to when Tom runs windows.
-- Cure for Cancer instead of SETI! (only w32 yet - mail and beg)
You don't have the divx codec??
Have you been living in the same cave as osoma bin laden or a different one?
there are 3 kinds of people:
* those who can count
* those who can't
I have Iwill boards KA266-R which, imho, have a very poor physical layout. The heatsink requires the use of a screwdriver and some incredible amounts of force to put it on or take it off. The dimm slot #0 is 3 mm away from screwdriver when removing/installing cpu heatsink. So, which I am leaning on the screwdriver with my whole body weight, I have the cpu on one side, ddr ram 3mm away on the other, and the motherboard with lots of little diodes right below inbetween cpu & memory.
I have already destroyed one MB and 2 dimms in this process, so I decided to just rest the heatsink on the cpu without the screwdriver procedure; 3 Duron 800s, each fried within 5 seconds. They never made it past post. I thought I has seated something wrong since it never posted. Alas, I compared the chips with a fresh one and the burnt ones were several shade darker than the original, looking very like Tom's pics.
A friend of mine forgot to put a heat sink on his 1,4Ghz t-bird. About two seconds after he had pressed the power button the processor caught fire and cracked into two pieces.
Expect the price of these things to continue to fall as the 64 bit machines finally roll in. It's kind of hard to believe that we are still using 32 bit stuff when Alpha has been around at least six years. Oh well, cheap is cool.
DMCA, Hollings, Palladium. What might have sounded like paranoia is now common sense.
I watched this video and found it quite interesting. One difference between this video and tom's video was the motherboard and chipset that was used, that could have something to do with the different outcomes. Tom used some motherboard I've never even seen before, I had to hunt the net to find info about it. The new amd video uses a motherboard with the amd 760 chipset.
My other issues against tom are the benchmarks that he uses now, compared to what he used to do. Last review I just remember seeing the bapco and quake3 tests. Which is interesting because of all the rumors that bapco and intel are the same company. Read here. Then the quake3 benchmarks... umm, wasn't that what the P4 was optimized for? I remember back when the K7's first came out and tom threw a barrage of tests at it... a good FPU one was the 3DStudio one (which I didn't see in the last test).
Lastly, I remember the massive intel ad banners on the site when Tom did the 2Ghz P4 review...
So in my eyes, Tom sold out and I don't trust his reviews anymore.
Question everything that you've accepted without thinking.
actually, i dont run windows on any of my pc's.
stuff
"Note: AMD CPUs must be underclocked to 33MHz in order for Thermal Protection to function."
Patrick Norton (off TSS fame) fried UGM 3.0 by not properly putting the fan on the 1.GHz Athlon.
...and that's all there is to it.
I'm thinking of buying an Athlon XP (when the Asus A7V266-E with the new fast VIA chipset becomes available). I'm not that afraid of loosing the CPU or even the board. The chance that my heatsink falls of it not very high. On the other hand, I am afraid of fire hazards in my house. And I'm not sure whether the chance of an Athlon setting fire to my house is zero. If it's not, it's not really an option for me to buy one and I'll probably go the P4 route.
I wish I had some certainty about the actual danger of these chips. Again: I don't mind the very small risk of loosing a CPU/board . I do mind any risk of a fire.
Don't the newly-released Athlons have a new thermal diode???
At least it has a .wmv extension. Is there an mpg or something somewhere?
I figured I would try to convert this to AVI or some format those /. readers who don't have a dual-boot/win machine could read, but upon trying to open the video in VirtualDub I get a nice message saying "ASF support has been removed at the request of Microsoft."
Damn it!
They were revving 'em up without the heatsink.
"A matter of internal security, the age old cry of the oppressor" - Jean Luc Picard
I work for a local computer shop, and although we are no Dell or Gateway, we've seen a lot of weird hardware stuff over the years.
And the fact of the matter is that AMD CPUs can really burn up just like in Tom's Video.
We've seen more than once, a customer coming back with what he said was a defective AMD CPU, and when we check the CPU, we could see the adhesive barcode that we put on the underside of each, partially burned out !
And don't get me started on the damn fragility of those AMD CPU, we've got dozens of them broken because some guy slipped when installing a fan on them.
Murphy
First of all is why the AMD processor doesn't fail in this video. Tom's Hardware mentioned that while there WAS thermal protection in athlons, the thermal diode responded too slow to be usefull (~1 degree Celcius per second was the max rate of change). Well if AMD switched to a diode that could respond much faster to throttle the CPU/shutdown the PC, things would be like seen in the video. The other two parts to this are they may be useing a thermal sensor on the motherboard that rests against the CPU instead of the onboard diode. The last thing is that I believe that Tom's mentioned that most motherboards don't pay attention to that diode. So it seems to me that either that the new athlons have a better thermal diode, or this rig was well set up so that the motherboard would shut the computer down in the event of an overheat.
While this is still quite a bit better, if you pay attention you may notice that Intel is STILL ahead of AMD. When the heatsinks were removed from Pentium IIIs and IVs, they computer continued to run, but the clockspeed was throttled way down. With the AMD, as shown in this video, the computer shuts down. While this is obviously prefferable to getting a new keychain, I'd rather be able to save my work if I was doing something more important like maybe 3D rendering.
In conclusion, AMD still has a ways to go. But if this is how things will work now, then bravo AMD, for fixing a dangerous problem. Keep up the good work. Personaly, I'd rather have a fast processor that costs $100 and shutsdown my computer when it overheats, to a fast $600 processor that let's me continue to run my PC.
Comment forecast: Bits of genius surrounded by a sea of mediocrity.
They didn't cover that one! I'm sure Tom would've "discovered" that the Intel Pentium 4 CPUs have an internal fresh-bread-seeking missile system while the AMD XP would just make like an oven and bake. Were still talking about the same "Rabbi Tom's Bakery" right?
But I'm sure you already Gnu that.
I have an AMD 1.4GHz in my server, and I noticed when it gets very hot (under the load, and becuase theres no air conditioning in here) the alarm from the motherboard will start beeping and the machine locks. From that point, I shut the machine off and wait about 10 minutes before rebooting it. Since its getting closer to winter, this isn't happening anymore. But, it still worrys me.
I belive what Toms Hardware posted, and the video appears to be very acurate. I just need to watch things better. This doesn't mean I will stop buying AMD. I am a long time supporter of AMD, they, IMO, have the better quality chip over Intel.
until (succeed) try { again(); }
you should stay an a anonymous coward if you were dumb enough to have your RAM seated when the socket is extremely close to it. Not to mentioning burning a cpu three times in a row.
Since both videos are talking about the Athlon XPs which have thermal protection, and not the durons, which don't, your claim is meaningless.
Any poster who claims to present benchmarks based on gcc 3.1 is either lying or incompetent. GCC 3.0.2 is the current release. GCC 3.1 is scheduled for release on 15 Apr 2002.
In high school, I used to check Tom's Hardware Guide regularly. His posting on over clocking seemed fascinating. He seemed to have a good understanding of hardware issues.
Then I went to college, still a reader.
Taking a few courses on hardware however, made reading his reviews painful.
He doesn't understand hardware, and it shows. Sure, he is able to run tests, but his reasons are completely flawed.
I don't accuse Tom of being on the take... However, I think that Tom should stick to testing, and not give his "engineering" insights that are based upon made up terminology and without an engineering basis.
This is a more reasonable review, and the first one that I viewed of his in years.
Alex
Durons are very cheap right now. I took out my Duron from my wooden box and thought I could check if the Duron could also make me hot coffee.
.. the computer shut down. I thought I had sent my Duron to the grave, but I waited 10 minutes just to make sure it was really dead. It was still alive.
What I did was to remove the heatsink from the motherboard (just an ASUS A7V board..) and then just turn the computer on. The Duron was a cheap 750 MHz Duron that switched place with an Athlon @ 1.4 GHz.
Anyway, I turned my computer on, and what happens? Holy shit, this baby is hot. The stench I could smell reminded me of burnt boots or something. Funny thing was
I powered the computer up again. What happens? Well, it works fine. I wait just a few secs, and then the processor stops working in that point that the monitor goes black. The mobo won't power down though. Tom's Hardware had a sleezy white smoke trail from his CPU. I suggest this is the burning thermal compound in this case. My smoke was black..
Anyway, I sliced the Duron in two pieces to see how deep it went. Unfortunately, I got no digicam, so no pics. The CPU seemed the be melted halfway down and rather black.
Regarding Tom's position to Intel, I tend to agree with some people. He seems to be paid by Intel, yes. Another weird example would be: why the heck does he show three Intel CPU:s and only two AMD CPU:s in that heartbreaking movie..?
I'm so glad that I'm not the only person who worries about this. the little server I leave running 24/7 is using a 466 celeron clocked down to 333 (on a slotket.) this is specifically so the tiny box will run with minimal noise and heat, so I won't have to worry about some sort of massively hot system short circuiting and roasting my living room.
:)
I love big computing equipment, but not when it's sitting in my house making noise and sucking power
EOM
Relax dude? :(
:) Funny. Well, I got more room in my wooden box now, hehe.
What was I gonna do with that processor anyway? I got a job that pays me $5000 a month so I don't really care about a few bucks.
I had no idea about the pre-Morgan thing though
Oh, you might also consider taking a pill or two with some calming stuff inside. You seem to express anger at me for no reason. Tell me, comrade, did I harm you in any way?
The original movie from Toms Hardware was on an Athlon. The AMD movies are on an AthlonXP, which if you look at what has changed added a thermal diode. So the Toms Hardware blew up, and the AMD didn't. What else would you expect?
I'm sure I am -1 redundant, I'm too lazy to look at the lower scored comments
HOWEVER,
YOU DON'T SEE THIS HAPPEN WITH LINUX ADMINS; PEOPLE INSTALLING SHITTY HEATSINKS WHATNOT AND PULLING THEM OFF WHEN THE SYSTEM IS ON.
Only Rabii Tom's Hardware.com would be stupid enough to give an example by running MS Windows and remove remove the Heatsink.
x86 is for retards
You don't see them showing examples of this with PA-RISC and Alphas. This justifies the bias of rabii Tom's Hardware.
But I'm sure you already Gnu that.
I object to this type of video trash. They provide an unrealistic, degrading portrayals of CPUs that take off their heat sinks/cooling units for no apparent reason! Like that sort of things happen all the time in real life.
I don't care who is making them, it must be stopped. Will someone think of the children!
Codeala - Just another mindless drone
When did the U.S.S. Heat sink?
No, but some people would read your post and think "those damn AMD processors suck and Intel rulez" because you painted an incomplete picture. Nothing ticks me off faster than someone spouting off when they obviously don't know all the facts behind the scene.
And just why did you want to kill a Duron, comrade? Had it offended you?
Tom's video shows what happens when you remove the heatsink from Athlons on current and available motherboards that DO NOT support the Athlon XP's onboard thermal diode. His video accurately reflects what will happen if you go out and buy an Athlon XP today along with any motherboard available in retail that supports Athlon CPUs.
This other video that's been released does NOT disclose what motherboard they used to test the Athlon XPs. For all we know this board is an internal only product that actually supports reading the onboard thermal diode of the Athlon XPs.
The fact is, you can have thermal protection on both the motherboard and CPU. P4s have onboard clock throttling that saves the CPU in the event of overheating. Athlons depend on the motherboard maker to implement this feature. Being that the Athlon XP is the first generation of Athlon CPUs to have a thermal diode, it's no real surprise that current motherboards do not support that feature. In 6 months to a year I'd expect that we'll start seeing that feature supported but not yet.
Bottom line: The new video shows what should happen if you remove a heatsink while a CPU is running. The Tom's Hardware video shows what WILL happen to you if you remove the heatsink on any current Athlon/motherboard combo.
But they're using a separate temperature probe and modified bios! The problem is most mobo manufacturers don't include the bios code to shut the system down... or cheap mobo's don't include a thermal diode at all. What they have demonstrated isn't implemented on 99% of the Athlon systems out there, Intel is somewhat better, this isn't going to save Joe Blogg's chip.
The new Athlon XP+ range now includes an internal diode like most Intel chips, by the time external sensors beneath the ZIF reacted it was too late, fried chip. So an internal diode, great you may think, but basically nobody has implemented the code to even query the sensor let alone set up the board to auto-shutdown. Tom used a board that implemented reading the internal sensor, it did just that, but the auto-off functionality wasn't there, again, fried chip. If AMD have to use an older Athlon with an external diode then it pretty much proves the functionality for reading Athlon XP sensors isn't on any board, yet.
Also... this thing crashes, certainly better than a fried chip but remember the P4 automatically scales clockspeed to temperate and doesn't crash, even if it means running at 100mhz (no data loss).
I think it's the motherboard that can't properly throttle, i think the Good Dr was just mislead with his testing.
Non impediti ratione cogitationus.
Huh?
If it ticks you off so much, keep it to yourself, dude. It's really not our problem to read your whining. Just write that I made a mistake and that this test does not apply to pre-Morgan processors. It's really that simple. Do you smash your keyboard when you read this? Buy a new one. I recommend Microsoft keyboards.
And just why did you want to kill a Duron, comrade? Had it offended you? -- How can a Duron offend me? I stated that I got a lot of money and I'd rather experiment with hardware than selling it or whatever. That Duron 750 was aged. Btw, Durons are not living beings. You sound like you think that.
I killed a Tbird by running it without a heatsink for a whole 4 seconds. It got very hot, VERY fast. I know several other people that have done similar things. 9 minutes? I think not.
"shows the system continuing to work when the heatsink is removed."
Actually, it only runs when the FAN is turned off. It still crashed when the heatsink is removed. Watch the video.
Look closely. Notice the AMD chip looks toasted the instant the HSF is removed? Looks like a 'lil make up has been applied.
Also. Might they have been shorting the AMD chips out as they clumsily removed the fans? (Remember, the intel chips don have nearly as much top side to get in the way)
Frankly, I'll believe Tom's first. He at least can claim to be an independant agent (though I'm sure I'm about to get 50 replies telling me about his secret AMD funding), so I think he might be a bit more impartial than the "AMDzone".
I can hardly imagine how there could be this large a difference. Either the systems were NOT the same (hardware-wise), or the burn-out chip was poor quality, or AMDzone is lying, or Tom's is mistaken/lying, or this has been misreported.
I see no way in which you could "mistake" as to whether or not a processor burst into flames upon cooler removal. That sort of thing is pretty much an either-or that anyone with eyes can determine.
-Kasreyn
Kasreyn: Cheerfully playing the part of Devil's Advocate to hairtrigger
I propose a new test: strap the mobo+Athlon to a wall.
.45 into the heart of the processor. Run performance test.
Scientifically fire a
Tests may show that the Athlon does not hold up under impact of a projectile. A video of this process may be necessary to prove the point to the skeptical.
Naysayers and Athlon proponents may argue that this test does not reflect real-world operating conditions, but who cares -- it's a great video.
my friend heard his fan and heatsink fall off, and by the time he turned it off - it was already done for.
help out.
I really dont want to badmouth Tom or anything but after watching the videos It kinda looked like the damage was done to the CPU before the tests. I know the Tom's is always objective and usually leans towards AMD's side but this is what it looked like to me - I dont know why someone would do that either.
My little Universe is cool for the people who can fit inside it (being 250 6'4" there aren't that many who can)
OK -- AMD has my vote for no other reason than using my current "heads-down" development soundtrack: the Matrix.
-- @rjamestaylor on Ello
If you look at both videos, the thing to notice is that Tom's tests were done using VIA chipsets, whereas AMD's tests were done with AMD chipsets. There is NO discrepency here. VIA's chipset is to blame because it doesn't shut down the system!
Go back and watch again. I think this makes far more sense. AMD would not publish a video that was outright forged, they would be silent on the matter. Tom would not gain anything by publishing misleading data.
THE PROBLEM IS THE DIFFERENT CHIPSETS USED.
XeoMage
The new athlons have a thermal Diode (as did the ones that tom tested) However at the moment there is no motherboard that properly takes advantage of this.
So what this video probably means is that AMD are about to start releasing this new feature and that it seems to work.
Rather than AMD being DODGEY its probably that they have just been testing implementation of this feature.
In any case if athlons stop burning themselves out its better for everyone as this chip is (in my opinion) cheaper and better than a p4.
ta
NankstaH
sure, a couple of ounces of moderate-sized heatsink is fine. but im sure he was talking about those oversized ones that stick out a few inches from the cpu - it becomes more of a problem in proportion to how "tall" the sink is.
Dont ask me...Im just the bass player.
I've never seen a fan that weighed more than a couple of ounces at the most
I have a large desktop fan pointed into my system right now (don't laugh, it reduces the CPU temperature by 15-20 degrees). I would guess its weight at 3-5 pounds.
I pledge allegiance to the flag...
of the Corporate States of America...
Either the systems were NOT the same (hardware-wise)
/. article is a joke.
do you not understand?
If the tests weren't run on the same frickin' hardware then they have no relation to each other, and this entire
-Kasreyn
Kasreyn: Cheerfully playing the part of Devil's Advocate to hairtrigger
I think you are a little boy-man. You like boys 'cause thats what ma taught ya. Go back to the shower, Kevin, and learn some more Visual Basic Programming.
I had a window fan pointed at my open tower case during the summer. It definitely reduced cpu temps by at least 20 degrees....but I had to go re-comb my hair every time I used my PC.
It can play DivX/MPEG/MPEG2/ASF/AVI with open-source codecs, but for WMV/WMA it uses a Windows DLL and WINE.
I was using a OCZ Gladiator all copper hsf. It weighs in at somewhere near half a pound. The one little plastic tab dealie didnt support it
The opinions in this post are ficticious. Any similarity to actual opinions, real or imagined, is purely coincidental.
As my former employer used to say:
"There are three things:
Good Quality, Fast Service, and Fair Price. You only get to pick two."
If AMD can keep their processors a few hundred cheaper than Intel by leaving out a thermal protection system, that's fine by me.
After reading the last article about this subject, I tried to upgrade the heat sink on my Athelon system, but I no matter how I tried I couldn't get the old heat sink off the motherboard. Those things hold on tight! The only way I can imagine them falling off is if they were installed improperly in the first place. I decided to leave it alone, and it has been working fine.
It's very easy to proof that a processor can't work for so long. Just put it under midday sun in Rio de Janeiro(+100F), and you see your CPU goes to the CPU's heaven.
But we can also put it to work in alaska and we don't even need the coller :o)
Whatever, these tests are just like the others. Tom's test was prepared to see when the processor crashes, and the AMD test's was made so the processor survives for a long time. It's a matter of statistics.
A joke now:
The mathematics made his calculations and said tha 2.351 liters of water was enough to solve the problem
The physics drawed some schemes in the paper and said that 2.500 liters was enough to solve the problem
The engeneer loked up, stayed quiet some secondes and said that 4.000 liters would solve the problem The statistic saw the problem and aske the fireman, "How much water do you want to spend?"
And that's what have happened in both tests.
-=-=-=-=
I know life isn't fair, but why can't it ever be un-fair in MY favor!?
Now I would certainly like to see a ~20W Athlon (which I have heard about, can't say where) some day, but even if they don't get that low and if they do burn up with no thermal solution, I don't care...
If you run it without any thermal protection you should know better... Even a P100 needs some kind of cooling (albeit not as much as anything now-a-days), so if you're running without, tough luck... If you thermal solution fails, you better have made sure in advance that the chip was under warranty for such things or that your thermal solution provider had a warranty in case their product failed or perhaps never works in the 1st place. I hope the Tyan Tiger + AthlonXP 1800+ MPs come down in price and then I plan to buy them and have a nice SMP system, and I could care less if it runs kind of hot, as long as it does what it's supposed to do and as long as the heatsink/fan/whatever I have does what it's supposed to do.
Now if when we eventually have fiber optic processors &/or quantum computers in people's homes and they burn themselves up, yeah then that'd bug me, but face it metal conducts, it's gonna heat up, plain and simple.
And if you totally disagree with me, that's kewl, I don't mind, you're free to care about such things if you want, I just felt like saying it doesn't seem to be that critical of an issue to me.
I have a hard time believing AMD was able to run Quake for several minutes without one. I can't imagine the architecture has changed that much since the thunderbird.
Looking for a computer support specialist for your small business? Check out
yep, works great ^_^ i'm now a mplayer fanatic, after all what other media player can play movies in ascii?
mplayer -vo aa -aaextended -gui
Aren't these the new AMD Pinto processors?
I have a windows ac unit pointed at the open case of my server. Condensation is my biggest problem, but once you get past that, it works fine. Cools it down much better than the ole fan ever did.
I recently (two months ago) bought myself a Athlon 1.4 Ghz CPU and because the fan that was supplied with it was so noisly, I switched it for another fan. However, I was careless and didn't mount the heatsink/fan properly onto the CPU (basically they weren't touching).
Let me assure you that the Athlon 1.4 Ghz is more than capable of destroying itself within 5 seconds if it doesn't have adequate cooling.
I am now Aus$400 poorer and a little more careful.
oh and i forgot to mention, hold down 2 while playing until the slider goes to around 75, and it'll even work in console! (don't put in -gui)
So it took about a month for AMD to correct the problem. Tom must be proud of his help.
The person playing Quake 3 doesn't have a soundcard in on the motherboard.
Does the name Pavlov ring a bell?
The Intel chip is still better, because the mobo didn't matter. The chip underclocked itself, or shut itself down, without any additional assistance. This is superior to any sort of motherboard dependency.
My 2 1.3 ghz athlon mp's were roased in about 10 seconds by soeone stupid enought to try it on my stuff. No; way allways have a heat sink.
and that was with the fan and heatsink sitting on top, just not sinched down.
Show us a free quicktime codec that is better than Sorenson and i'm sure people will start to use it. Your accusation that the "problem" is because its not free, is an invalid excuse. It was used because the output video is better (best video for a given file size, given play time, and given resolution) than any other codec, not because it cost money or wasn't free.
The comparison is irrelevant, since they aren't testing the same processors. T-bird, Palamino v. MP? no.
I just wasted 8 minutes on disinformation.
Ok it was cool to see the processors litterally burn up.
Funny, Maxim's first app note for the MAX6512 is Simple Circuit Activates Fan When Processor Heats Up.
So it is apparently a mobo feature that uses a thermal diode integrated into the processor. Is the thermal circuit in the Pentium totally self-contained?
+5:offtopic,but anti-American
That statement is misleading. It uses WineLib to read the dll, you do not need to have wine installed for it to work.
"We obviously need a new moderation category: (-1, Woo-fucking-hoo)" --Mr. AC
I built a Thunderbird 900 system for a friend and the heat sink that we bought, Thermaltake Chrome orb, for some reason appeared to float 1mm above the chip, which we didn't find out until the system started locking up. While trying to figure out first hand what was going on, I removed the heat sink and powered on. The system ran for 5 seconds and shutdown and never came back up. We took the CPU and fan back to the store, and the guy there said that it was our fault for buying the "wrong" heatsink. Hello! The label on the front said "Athlon", and the salesman didn't stop us from buying it when we were getting the parts. Finally they agreed to replace the chip and then sold us a Coolermaster heatsink. Might I also add that after removing the heatsink the first time, appx 1/2mm of core flaked off the processor (still ran). I think AMD has more than a few design problems to worry about
Okay, so on two different systems, both with the MSI-k7 master motherboard, I have seen chips fry within seconds of being booted without proper cooling. One of the chips actually exploded so violently that it broke the entire chip in half, and blew the little processor off the rest of the chip. The other one just put out a little smoke and never booted again. So, I believe Tom.
Sam
They are two different Flash files; both are archived on http://www.stileproject.com a few days back. Go hunting and you will find them.
This whole thing is a big "who cares" in my book. Sure, it's fun to watch video of CPU's blowing smoke. But get serious. My car can't run long without water in the radiator, either, though it'd be possible to build one that would.
The solution is simple: Don't do that.
I don't see how the heatsink could fall off a TBird unless it wasn't properly installed in the first place. Damn, I can barely get them off when I try; you'd have to drop the machine off a building or something to jar it that hard.
I have a 1.2GHz AMD Athlon and my CPU fan did indeed fail on me. The processor reached ~120 degrees Celsius after maybe fifteen minutes. Waiting for it to cool to room temperature and turning my box back on revealed that it would rise about two degrees every five seconds.
The only way you can keep that sucker running without a CPU fan is by leaving your case open in a wind tunnel.
From an earlier HardOCP article...
"One thing that has been worth looking forward to is the addition of a thermal diode inside the core of the AXP CPU.This will for allow you to read the core temperature when the right hardware and software are present. You might have seen a famous hardware site make the statement on a video recently that even while their Palomino has a thermal diode in place, it burned up anyway when they removed the heatsink. Well, this was no fault of the AMD CPU, just the fault of the misinformed operator. You must have a board that has the ability to utilize the diode and also have the proper BIOS to facilitate a system shutdown in a high core temperature situation. Of course, making it work cooler now allows us to speed it up and make it work hotter later!"
Tom's is usually pretty accurate but he too has had some knocks on his credibility from time to time. I think the best thing we can take from this whole incident is never take any site's opinion as gospel. Always read several site reviews before making a decision about a product.
G. Washington on Government "it is force. Like fire, it is a dangerous servant and a fearful master."
Firstly, I need to make my opinion on tomshardware absolutely clear -- the articles are clearly written by kids with half a clue, mostly picking up fancy words from specs documents and running the occasional benchmarking routine. I take everything they say with a pinch of salt.
Secondly, what the fuck does anyone need to run a ~2GHz AMD CPU on the desktop for? I was moving a few machines around today, and my brother happened to notice that the fan/heatsink had come loose off my dual P-II 233MHz. Yes, it was hotter than the other CPU, and no, I have no idea how long ago this had happened, but it was probably earlier in the day. But no damage caused.
Yes, that's right, 2x 233MHz, it's quick enough for me, and I have no worries about overheating my PC or myself for being close to it, nor do I panic when the electric bill comes.
Not that I would have worried that much, because the motherboard has clock throttling and saving that, a critical CPU temperature setting. This, to me, is more important than the ability to pay the latest Power Bash Pro 3D Plus Quake at 3000fps.
Now, to extoll the virtues of the Alpha box on my right, with more sensors inside it than I can throw a stick at... even keeps a log in non-volatile memory of temp/fan status over time.
A little effort in buying quality not quantity (more MHz != more value) goes a long way, boys...
DivX is better.
It is APPLE's fault that there are so many bloated QT Sorensen movies out there. They accepted the licensing terms, tricked vendors into buying the qt movie producing software, and bundle it with all the crApple computers.
As the age of broadband nears, more and more people are going to be pissed off at that stupid "Upgrade to Quicktime Pro Now" popup nag screen because those motherfuckers are the only one with a compatible player.
Choices:
Smaller same quality divx
Sorensen quicktime with proprietary player
Somewhat larger mpg (remember broadband here..)
Larger other codec avi
Sorensen is dying
... people doesn't have to replace their processors every five months or so. AMD has a strategic initiative to manufacture unstable products because that helps to drive up the number of purchases. Just wonder what kind of mistake Intel made when they put that cooling system on their chips (being called a MHz whore, seeing a rapid dip in revenue, ...).
Seems a lot of people THG as biased.
I don't know how to put it. But given that you can reproduce all their results, and they did run a lot of testing under different situations using different tests, you may disagree on what they conclude, but you can't disagree with the test results they have. It sounds unbias to me.
And given there are a lot of tests they have run, the human bias of choosing particularly one type of tests to favor the results could also be reduced.
Maybe THG is really biased, but it's not more than the views of those who attack THG.
Now we have two situations, that THG produce results of heat death of Athlon and AMDZONE produce complete workable Athlon results when heatsinks are removed, by using two different chipsets. I just wonder when people would stop argue which one is accurate, and really put some thoughts on the results.
A sig is redundant.
Today's giga-plus processors really, really, run hot. They need to be protected from thermal runaway upon startup, with or without heatsink. Are they adequately protected? Apparently not. Is this the fault of the bios, or of the processor? The answer is not clear from the preceeding discussion.
Before we progress much further though, we will need to address more fundamental question: How do we minimize chip-generated heat? Transmeta seems to be a one-word answer, are there any others?
For my purposes, I prefer a more efficient, if slower processor. 100 watts seems a reasonable amount of power to have running in my home continuously--it's the power of a reasonably-bright lightbulb. Anything more and I'm reminded of my extravagance in my monthly electric bill. This also seems to be the point where I start resenting my cooling system noise.
Your homework for tonight, chipsters, is how to squeeze more computing power from less and less electrical power. It's a compulsory question that must be answered, if you want my consumer dollar.
In my case, Asus K7M, AMD 1.4GhZ processor. Built a bum fan tail, fan died, CPU death resulted within seconds. My all-copper heatsink turned all sorts of funny rainbow colors from the intense heat that resulted, and the top of the processor looks half-melted.
I'll agree with the posters saying it's a motherboard's thermal sensor issue, not necessarily an AMD problem. But I and my $142 credit card charge promise you, you lose your fan and/or heatsink and you've let the magic smoke out.
-- If we were in any other industry they would've shot us a long time ago.
Licenced the technology from those "smokeless" cigarettes you kept hearing about but have never seen. It is also possible Truth got to them and wanted them to quit smoking - it's setting a bad example and Intel might start from peer pressure.
Pity, I liked knowing my high-performence processor could "smoke" the competition.
---
DRM is like antifreeze, to the MPAA/RIAA it's sweet, to the consumers it's poison.
when I got my duron 700 a while back the idiot from the computer store forgot to take off the plastic piece covering the heat compound on the heat sinc. well when I had it off I decided to turn it on with my finger on top of the center of the chip. well was that a misteak! within 2 seconds I had the Duron Logo burned into my finger for about a week. it was pretty cool to be branded like that but it hurt! I dont believe this. I would believe Toms Hardware before this. and what the hell was that blinking light thing? I dident read anything I just watched the video.
Daddy would you like some sausage?
Can anyone tell me what music is playing at the end of the video after the heatsink is turned off but still on the chip all the way to the end? I have this same music in this skate video but there are no words so I can't search for it.
When I watched that TH video, it struck me as odd that the intel chip had no heatsink compound and the athlon did. Personally, I think it was that compound burning, not the silicon or housing...
I posted about this article already under a slashdot story about heatsinks not too long ago.
;)
I thought it was funny how they removed the heat sink to see which CPU fried. In any actual event, if the CPU fried, it would be the heatsink's fault, because the CPU and heatsink are designed to go together in actual operation.
I think its back to better heatsink design before we lay any blame on AMD.
Perhaps they could return to the old CPU/HEATSINK combo that I remember on old 286's. It seems logical. Why don't they sell a heatsink(+fan) cpu combo, that is designed to go together, and make it a standard package? I guess it would seem logical, but less price=more potential buyers. AMD: Forget it, let them fry their CPU so they will buy another.
If you make a product too good, it probably won't sell as much
Well.. maybe not heat sink but fan.
I have a 1.33Ghz Athlon and had the cpu fan plugged into the motherboard, the motherboard (MSI K7T266 with Raid) decided to quit providing the necessary juice to spin the fan. I thought that the defaults in cmos would have been fine, so I hadn't changed the temperature warning/shutdown thresholds.
After running for 10 or so minutes (sometimes up to an hour or two) I'd start getting lockups/crashes/etc. Finally popped the case to see what was going on and found that the fan wasn't spinning.. plugged it into the power supply and everything has been fine since.
I will say I am never getting another MSI motherboard, besides the power for the fan failing and the system never noticing the chip was overheating, the mic on the onboard sound doesn't work either.
Yeah, I could get it RMA'd, but who wants to live a couple weeks without their motherboard?
kenny
If the heatsink is attached but the fan is not plugged in or the heat sink is attached backwards or lastly if no heat sink is installed the chip (AMD Athalon) will burn up with in seconds...I know...I've done them all...
The lab where I work purchased an Athlon system for testing (this was about 3 weeks ago). In shipping the heat sink fell off. When it was booted, POOF!
Order up, one well done Athlon.
Rock on, AMD! AMD's can handle anything. To prove it, I've got some videos and stuff running right now (loading the CPU), and I will take the side off the case and remove the heatsink... (ouch! it's hot!..)
See!
In particular, the 1 degree/second claim from Siemens is a little suspect. I find it hard to believe that any thermal diode would react that slowly to temperature change. If the thermal diode was that bad, how could Siemens even implement any thermal protection? The cpu would be dead before the motherboard even knew what happened. Instead, I think it was the thermal protection on the Siemens motherboard that was defective, and then Siemens tried to place the blame on AMD.
I studied the end of the clip, around when it says "Over 9 minutes later". They show the same 3-4 sec sequence of playing before and after.
See for yourselves!
Is the power down function built into the chip itself or into the mobo? The AMD video was reasuring after seeing the video from toms hardware a few weeks ago, but not as fun to watch :P.
Carpe meam simiam!
I had wished he would have posted the article earlier. About a week before it was posted, my CPU fan died. It friend the chip and appears to have damaged the mainboard so not to be usable. I was highly disappointed because I stuck a lot of money into a system that was supposed to be the "hot" ticket. It didn't last long. Shortly thereafter, I switched to Intel. Not only do I not fear the thermal problems as much, but I have had a MUCH more stable system with ZERO compatibility problems with any of my devices. Having been a long time AMD customer, I can't remember the last time this was true.
Remind me again why I purchased AMD in the first place? Oh yeah, the price. Well a fried motherboard and CPU is hardly cheap...
I'm curious what you all have to offer as far as recommendations for a good heatsink/fan for an Athlon 1333. Right now, if I remember correctly it runs at about 86 C steady, with a simple CoolerMaster.
Also, at what temps does performance take a hit, and at what temps does it improve? Since we're talking about thermal tech. here, I think these are very relevant questions.
What?
Never tried it with no heatsink, but when building a system with an AthlonXP I installed Windows 2000 without the fan ever operating. The heatsink got HOT, but the system never crashed nor did the chip die.
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
I have to go with Tom's Hardware on this one. Just this summer I bought a 1ghz and was really pumped to have a go with my new machine, but after trying to get the darn heatsink on for ten minutes I gave up and I thought I would satisfy myself with watching the large RAM number come up. I did get to see the number, but I also managed to burn a hole through my rug as the little thing on top flew off. Which smelled up my room for two days in that taunting sort of way that says "HAHA you just wasted one hundred dollars" and since the company said it was my fault and I couldn't get a replacement I did waste a hundred dollars. On a more positive note the split processor looks cool on my wall.
I remember reading that current generation motherboards (pre KT266A, anyway) do not support the Palomino core's internal thermister, and therefore that would cause the temperature monitoring hardware not to kick in and underclock the processor.
Running a CPU without a heatsink is like testing a car with no wheels - the results are meaningless
It shouldn't matter since the DeltaTRAK ThermoTrace thermometers he used were infrared, non-contact thermometers. I'm not sure, but, I'm guessing (makes me an expert here :) ) that it uses some EM wave magic to determine the temp from the reflected wave of IR light. Perhaps someone else could, ahem, shed some more light on this matter?
***OT***
While I'm at it... Mitnick, don't go into acting. consulting for tech-related television/movies, maybe, but not acting. It was a nice chance to work on the set with Jennifer Garner, though, so not a total waste of time.
***/OT***
Withdrawal before climax is very ineffective and those who try this are usually called "parents."
Maybe you should stop putting unicorn and heart stickers on your customers CPU's Douche!
The current Slashdot moderation system is made by gay communists!
I forgot to remove the square of paper from the bottom of the heatsink/fan. It covered the thermal grease. I thought I had a bad processor, because my system would boot and then crash about 20 seconds later. I noticed in the BIOS that the temperature was too high.
I took it back to the shop, the techs peered at it, made ridiculous suggestions (turn down the heat in your house, point a fan at it, etc, etc).
Took it back to the sales desk, and a sales guy noticed the square of paper, and pulled it off.
Now, there were no instructions with the heat sink and fan indicating that this paper should be removed. Regardless, I felt a bit dumb.
The processor was fine, but.... I would suspect that the extra heat generated may have shortened the life of my CPU. Will only last 5 years instead of 10. Of course, I will upgrade in about 6 months, so who cares.
Working at a big time chip manufacuring company myself... I know how this happened. Basically, you have AMD pushing the envelope because they tend to struggle with their manufacturing process. So when chips bin out at certain speeds, some are better than others. What Tom came across was the typical chip, probably just barely passed as a 1.2 GHz. AMD obviously knowing how to test their own chips, found some that were 1.2, but tested very high. Hence, they don't fry in milliseconds.
I think the best answer to this debate is to read another source, one who has a P4 but still believes that AMD makes great chips. I dont think anyone has posted this yet, but hey, post it twice, twice the chance youll take the time to read it. http://www.64bits.org/cpuheat/cpuheat1.htm the last intel proc I had was a pentium (pro?) 100, ive had a 266, 400, 450, 750, 850, and now my 1.4, (all AMD) since then, so rather then be biased like the oh-so-many-amounts of other people, I think I'll just go with "both videos have problems, I'm not believing one or the other untill my heatsink falls off (maybe an earthquake or something like that will make it happen?) and my processor reaches many times the necessary heat to melt the aluminum interconnects that supply power to the core" theory. but, I do think tom's video was entertaining, and isnt that the purpose of the internet?
Yep. That MicroSoftRam. Old 16mb pc64 simms renamed MSRam selling for $350 a stick. Damn Bill Gate's taking advantage of the average consumer.
One thing is removing the heatshink while the CPU is running. If this happens accidentally (ex., the clip breaks and the heatsink falls off), it'll probably hit graphics card, short-circuiting it. So even if the CPU survives, your system still crashes. But the chances of this happenning are very, very slim.
A different thing (rather more likey) is the fan stopping while the heatsink remains on the CPU. This has happened to me once, with an Athlon @ 1GHz, when a cable got stuck in the fan (whoever designed those Titan Majesty coolers didn't remember there are cables inside the case). After about 1 minute the motherboard started beeping and the system froze (I don't know if it this was a safety measure taken by the board - an Asus A7Pro - or a consequence of some error). I turned the power off, pulled the cable from the fan, let it cool for a couple of minutes and turned it back on. It still beeped for a while (it was at about 90 C) but booted normally, and it's been working fine since.
Anyway, an appropriate heatsink is a requirement of the CPU, just like a proper electrical supply. If you feed a CPU 220 volts instead of 1.7, it probably won't last long either...
Well, the P4 mobo was an Intel chipset, probably the P3 also. I guess this means we should ditch 3rd party chipsets so our cpu's dont fry when our hsf falls off. Oh wait, this almost never happens! :)
A person I built a system for didn't realize you can't handle a computer in the same manner as a football, she broke off one of the tabs that holds the heatsink on to the processor. The heatsink came halfway off, a lot of heat was generated. The plastic clips that hold the fan on the heatsink melted off.
At this point the computer was making a really loud grinding noise(since the processor fan was now resting upside down on the video card). Of course, this person didn't realize that the grinding noise was a bad thing! So she went on using it for a good day and a half. Eventually it shutdown and never started again.
When she finally called me the entire processor was charred black. It was an AMD 1.2ghz. I've still got it lying around here somewhere.
Watch out, or I'll have the penguins eat you.
Oh...and, I'm liquid talent
I have found that hair is more of a pain in the but than anything so I remove mine on a semi-regular basis. It falls out into the keyboard, you have to brush it, you have to wash it, you have to make sure you don't get car grease in it because it ruins it...it can catch on fire...if it gets caught in a table saw your fucked...
Besides, mine is gonna be gone for good soon so I figure I should just get used to it.
jik-
After 9 minutes, the video camera stopped working because the room all of this took place in was cooled to -50 degrees C...
Somebody's already noted that it might be a chipset difference that makes this one "not fry" but I'd like to point out that, while *almost* identical, Tom's Hardware tested an Athlon XP whereas this movie claims to test an Athlon MP.
Could this be a difference?
I tried that once for a few minutes until I saw the flicker of the monitor slowly build and realized that the fan is producing some rather strong magnetic fields that would be oh-so-healthy for any data on magnetic disks...
_____
If you can't hear the voices in my head, then you're just not listening hard enough.
I just got a new Athlon 1800+ CPU with an ABIT KG7-RAID motherboard. It had a thermisistor for the CPU with good default settings, and it has a CPU FAN RPM detector. In the BIOS you can turn on a feature to shut the system down if the CPU fan fails, and you can also set alarms for CPU temp.
So, given that, I'd rather have the better performing and cheaper Athlon system. The risk seems minimal and by the time time you pay slightly more for the Intel CPU, the Intel mobo, and the Intel Rambus RAM, you're paying a lot more. My personal opinion.
- Twid
- "When you want something with all your heart, the entire universe conspires to give it to you" -Paulo Coelho
my athlon (a thunderbird 1Ghz) is working with a dual fan at approx 80 C, every once a while my cooler has little problems, the cpu starts to overheat and i get a warning. normally i turn off my system and fix the cooler, but a few days ago i wasn't home when it happened, ans i must say that my computer continued to poerate for about 3 hours then it crashed (i saw this in the logs). what is more important - > the cpu is still functional. although amd produces more heat than the pentium, i don't thinkt that it will melt at once when the heatsink is away
".Sig Stealer" was here
But Tom was testing on a system that DID NOT SUPPORT PowerNow. Yes. It had support for reading the thermal diode. But, IIRC, simple support for reading the diode doesn't equate to PowerNow support.
So yeah. If you run on an older system without PowerNow, it's the same as running an Athlon without a thermistor.
On a setup that actually supports the thermistor, the system will shut down in time.
Chas - The one, the only.
THANK GOD!!!
I also work for a computer shop! In the past month we had about 50% of our Athlon chips come back to the store burned....this was with the "retail heatsink & fan" on mind you (we now ship them with modified thermoengine coolers). Intel chips? Only 5% returned. Also, for RMA, let me just say that AMD blows. Took us over 1 month to get a replacement Athlon, and we had to pay shipping, whereas Intel shipped us the replacement first within 2 days, and paid for the shipping of the defective part!
I think we will continue to sell a few Athlons to devoted AMD fans, but we recommend everyone to move to P3/P4 Intel Chips....
I have owned a classic athlon for nearly 2 years. The processor was equipped with a dual Fan heatsink. One of the fans failed 8 months ago (it was not a amd heatsink / fan combo/ but was made by a third party), and me being the cheapo I am didn't replace it. About 2 weeks ago, the second fan died. The Athlon in a linux box with 200+ days uptime rebooted itself magically. I didn't open the case because I was busy at the time, and it stayed up for about a week. It rebooted again, when I opened up the case, I noticed my poor athlon didn't have a single fan to cool it. It didn't fry, and it actually ran for a week with NO heatsink.
I'm glad to say that she has a new heatsink / fan and is back up and running stable as ever.
Remember that you are unique, just like everybody else.
Also. Go back and take a close look at the 1st part of the AMD section of the Toms Hardware video. When the heatsink is taken off, the processor already looks burned. Look around the center of the processor, and you can see what looks like burn marks.
Did anyone else notice that?
That's all.
If the mainboard is compatible & is able to read the intenal thermister inside the AMD CPU then the CPU will just throttle back more 'n more as it gets hotter & hotter, eventually when it hits the maximum barable temp it will shutdown..
If the motherboard is not compatible, then either your computer will shutdown straight away or your CPU will fizzle (depending how your bios is setup & how your CPU fansink's fan is wired - if your fansink's fan is plugged into the motherboard's CPU fansink header connector & your bios is set to 'auto-shutdown on CPU fan failure' or something like that, then your computer will just shutdown straight away when it detects the failed fan, if not your CPU will fizzle)
The famous German computer magazine c't reported the same as Tom in a recent article. (It was about coolers, IIRC.) They said that you have to be very careful to mount the cooler correctly. If it doesn't sit correctly, i.e. it doesn't contact the CPU flatly and there is a millimeter or so of air between cooler and CPU, then the CPU burns and is destroyed "within seconds" - the machine won't even boot anymore. They said so, because it happened to them during the tests. It was an 1.2 MP, IIRC.
Sorensen is better than Divx (try it), but unfortunately the target market of porn rippers can't afford the $30 for the encoder, so crappy MS technology will win again.
i love my new athlon 1.2 but one thing is sure, if my fan stops working is bye bye cpu, it is a hot cpu with a cooler without one it is not a cpu its the sun
...AMD wouldn't replace my CPU when it went up in flames when the fan failed.
If this is a feature that needs MB support, it should be explitly advertised as such and then we may see fully XP supporting boards.
See my journal, I write things there
Well, I have been a long time user with AMD CPUS since the days of K6 - I have built quite a few Athlons and I did manage to fried a Athlon 750 by not installing the heatsink properly - and the dealer in Hong Kong happily replace it (and give me a Athlon 850 wheepppy) I do agree with a lot of people that AMD processor aren't as strong in terms of installation - but then aren't we all suppose to handle our toy carefully? I mean would you drive your Porsche like a car you stolen (hmm.. actually people do! ;-p)
;-p
I also have to agree with people that Tom is rather bias at times but then hey like CNN aren't?! Oh well, I will still recommend AMD CPU to everyone - except I will make sure that their CPU cooling solution (what a technical word!) is chosen wisely!
Oh and Athlon MP with Tiger MP is way cool
The cameras on the TM videos were held over the proc and after about five seconds smoke began to appear form the AMD chip, the camera on this video was turned off only a couple of seconds after the heatsink was taken off and was held far enough away from the proc that you couldnt see it. I wuold say this video is dubious at best. looks like marketing or defence of AMDs honour to me, hey I've got one and it rocks - but I aint gonna take the heatsink off.
2 wrongs dont make a right - but 3 lefts do
Maybe it's just because of the new internal thermal diode of the palomino core in combination with a mobo that supports it? And Tom's Hardware used Athlons with old cores?
Just a thought.
We did a little round of computer shooting about a year ago.
We placed a 486 CPU chip on a board and fired a 0.22 into it. Also shot up laptops, monitors and other computer gear.
As expected, the 486 shattered. I shot a video and encoded it in QuickTime MOV format. We couldn't run any benchmarks on the remains though. The whole article starts here.
So, lets get to the point: According to the video an all-aluminum would be *poof* a very bad idea?
I personally know someone who had a 1.2GHz T-bird
that he started up just after buying it. Without a heatsink
or fan. It melted in four seconds.
I received a customer return where the customer had decided that the cooling fan was a bit too noisy for him, so he removed it. Then when the 1.2 Athalon had welded it self to the socket and died, he sent it back. It was dead in the water. These babies really 'cook'. They really do need better thermal characteristics. It was a Compaq system and yes the fan was noisy, but...
The SCO box had been going kind of erratic, killing the odd user process. This was bad because their entire business (four vet clinics, maybe 20-25 workstations) was running from it.
I couldn't touch the case bare-handed to get it out, it was too hot. I used some convenient price lists as insulators to get the (still running) box out far enough to circulate air around it, at which point I aimed a fan-heater (set at blow-only-no-heat). After a few minutes, it was cool enough to touch, albeit briefly, so I unbolted the lid and angled the blower at the PSU. This kept it happy (ie no more dead processes) until after closing time about nine hours later, when the clinic's director replaced the PSU.
The box kept running flawlessly for some months until they replaced the P-II-350 SCO box using 32M of RAM and about 800M of its 2GB hard disk for Telnet clients with a monster multi-CPU gigabytes-of-RAM acres-of-disk Terminal Server box supporting similarly high-grunt Win2k client boxen, each doing nothing more than an X server would be.
Got time? Spend some of it coding or testing
Mind you, the no-heatsink test does smack of running a half-track into a passenger sedan.
Got time? Spend some of it coding or testing
The videos are definately eye opening... especially the one from Tom's Hardware. I was wondering what the major difference between Slot and Socket chipsets were? The video was all Socket Chips... I personally have a motherboard set up for Slot. I was debating whether or not I should change it. I personally do not know the difference except the difference in apprearance, and the fact that I can not get a Slot Card (AMD) above 900 MHz. Is there a reason for this? Thanks in advance for your worldly computer knowledge folks!
Linuxrunner
www.slightlycrewed.com - Because aren't we all?
I don't want to have an electric fan in my
CPU heatsink. Is there a totally passive
heatsink for the Duron which will cool adequately with
no fan at all?
You're forgetting one very important detail. The CPU presented in that little video clip is the Athlon MP, the new Palomino core that supports SMP. I wouldn't be surprised at all to find that AMD made their new MP processors resistant to heatsink/fan failure, following on the heels of Intel's P4.
The processor that Tom Pabst had fried is an older model, using the T-Bird core. It is only fair to assume that the old Athlons didn't have this overheat protection built in (which becomes obvious for all of us who have fallen victim to those stupid rubber spacers on early heatsinks).
-Billco, Fnarg.com
My cat knocked over my PC with a AMD1.2 MP with thermal protection software and sensors that worked and within 5 seconds, there was a horrible smell and a fried CPU. My Pentium 733s fan failed and it ran for 2 hours until the computer shut down. I like AMD but for the love of God, Don't let your cpu's fry so fast!
Welcome to the Entropy Bar, may I take your order?
We toasted an Athalon 1400 in about 5 seconds, cracking the CPU ceramic nearly in half. In the process of switching the heatsink we failed to completely remove a thin layer of plastic protecting a preinstalled layer of thermal paste. The top layer of protective plastic was removed but a second layer (which should have been removed) remained on the heatsink. This 0.5mm thick piece of plastic was enough to cook it. The MB survived so I imagine the CPU took the majority of the heat load.
Yup, I managed to fry a 1GHz Athlon when testings fans.
Recently I was working in my system and, in the process of putting everything back together, I managed to shift one of the drive-power cable harnesses down onto the CPU fan. This blocked the fan when I powered up. The system froze within about a minute after booting. Fortunately, I found the problem and the CPU worked after unblocking the fan. This was is a 1Ghz CPU, so I don't how different that is from the 1.2.
It's from personal experience, at this rate the mere act of me powering up my 1.2 tbird would have fried it (I've had it foolishly powered up without a sink before) and the chip still runs, even though the damned KT-133A chipsetted board won't let me do 266 FSB. Anyone got a fix for a Tyan Trinity KT-A so it'll do 266 without crashing at the BIOS boot screen?
Don't call my crazy, that's what they called me back in the home!
I downloaded those videos from toms. Does nobody else notice that when that hand pulls the heatsink off of the 2 Athlons that the Heat Damage is ALREADY on the chips? So either they got damaged with the heatsink on it, or they were pre-damaged. Either way I don't like it.
Some of you may be surprised, but I duplicated Tom's experiment... uhh... accidentally. Running my Athlon with an improperly mounted poor heatsink (ie not in contact) fried it up good. Little black scorch marks and a processor that doesn't do anything were the result. Luckilly, AMD processors are cheap enough that I didnt cry for more than a day.
A perfect example as to what it's like to have good competition in the market place. We have AMD and Intel thrashing it out ending in us getting what we deserve, a better product. The fact that we see two different vendors of processor in the high street is lovely.
Wouldn't things be great if we had a choice of operating systems in this way?
Didn't read the article, did you? The motherboard was specifically verified with its manufacturer that it was compatible. However, the thermal diodes response time was not designed for catastrophic heatsink failure. It is unable to respond in time to catch the failure, since the temperature of the processor skyrocketed in just a few seconds. He did, in fact, use one of two motherboards he was aware of that support AMD's thermal diode in this configuration. Also, after the burnout, he checked with the motherboard's manufacturer and found that yes, this was a realistic failure scenario, since the diode can only respond at about 1C/s
This shouldn't surprise anyone. Tom has been questionable in the past and is arrogant to an extreme.
AMD must have pissed him off.
And it isn't like the MB manufacturer would have any reason to blame AMD instead of itself. After all, every company in the industry takes responsibility when its product is at fault, instead of trying to pass the buck to someone else. Tom was perfectly justified in taking the MB manufacturer at its word.
luser
i personaly own an AMD Athalon 750, its not even 1GH and its a space heater. you people think there arent going to be problems if you start grilling with a space heater? yes there will be problems if you take off the heatsink, its not a question if your CPU of any make or brand will live after 7min. of Quke III without a heatsink. the question is "are you stupid enough to take off your heatsink?"
Careful what you say around me.. I will assume you mean it.
I'd like to comment on both the heat issues and the fragility issues. A couple months ago my friends and I went to the local MarketPro computer show and picked up a couple Asus A7M motherboards and Athlon 1.333 and 1.4GHz CPUs.
:oops:) and turned it on. Immediately some text appeared (BIOS junk) so we turned it off.
:( I took it back over to my friends, stuck it in there (with a heat sink this time) and tried it. No dice. All it took to fry my CPU was around 1 or 2 seconds w/o a heatsink. AMD was kind enough to ship me off another one a week later and it's been running great since. Moral of the story? Always use a heatsink. Second moral of the story: read the manual, since in big bold letters AMD says 'Under no circumstances should you ever - ever! run this CPU without a heatsink.' (paraphrased).
When we got them to our respective homes mine, for some reason, kept beeping (no RAM or invalid memory type). Grabbed my friend's known-good RAM and still had the error. Thought it was my CPU, so I took it over to my friends house. All I wanted to see was it POST since it couldn't do that on my mobo @ my place. I popped it in (sans heatsink
Got a new motherboard in the mail 3 days later and I pop everything in. No worky!
In regards to fragility somehow in the shuffling of CPUs my friend managed to chip the die on the front - rendering his CPU useless until AMD, once again, cheerfully sent him a new one.
Thanks,
--
Matt
I fried a 1.33 in about 10 seconds without the fan attached, cause I was being a moron. But really, the smell told me it was toast seconds after I mistakenly hit the power switch.
We have an athlon 1.2GHz, on a motherboard that although allows monitor & protection for CPU Fan Off, comes with that set at default as OFF! Before we knew the fan had failed, we actually were able to run the system, from a cold state, for about 10-20 minutes without any trouble, and on pretty intensive applications. (No Quake, but a lot of Windoze processor intensive stuff...do you know how many cycles it takes just to shuffle stuff around in its brain-damaged memory scheme?) After that amount of time, it would BSOD and halt. Anyway, when we put a new fan on the system, it worked fine. I'm typing on it right now, and we haven't seen even one problem that would suggest anything has happened to the processor. Point is, make sure you have your motherboard protection turned on if it is provided, and also, Athlons are much more durable than some give them credit for. (This problem was happening for about 3 or 4 days.)
Most Socket A heatsinks have a thermister inside the middle of the socket. But they basically do buggerall unless you have some utility running in the system tray all the time.
However if the CPU's fansink is plugged into the motherboard's CPU fansink header connector & your bios is set to 'auto-shutdown on CPU fan failure' or something like that, then your computer will just shutdown straight away when it detects the failed fan.
What Tom was I assume testing was the internal thermistor that all AMD Palomino cores have (Athlon MP, Athlon XP, & Morgan cored Durons).
In which case, if the mainboard is compatible & is able to read the intenal thermister inside the AMD CPU then the CPU will just throttle back more 'n more as it gets hotter & hotter, eventually when it hits the maximum barable temp it will shutdown
As far as I know, so far the only Socket A boards compatible with the Palomino's intenal resister are the latest Seimens boards, they of course still have thermisters inside the socket for compatibility with older T'bird Athlon & Spifire Duron cores, too.
I have a friend who bought a crappy heatsink and fan that was impossible to attach, so he thought he'd see if his 1.2Ghz Athlon would POST. Unfortunately it did not and his CPU started smoking before he got it turned off. Anyway, I'd take my friends word and Tom's over AMD's
"about a burnt Intel Celeron(!!) after a cooling failure"
This is a Water cooled Celeron ?
Oh, an Overclocked Celeron...
Well, as long as you don't have redudant cooling systems (think SERVER) you should know that an OC CPU is a potential Frying Pan 8)
"Please shoot me when I start pumping water into my computer!"
You know, water cooled is only a term...
Die Hard (or is it Hot Die 8) Gamers with an OCed CPU are using Mineral Oil when serious about the whole deal...
Non Conductive, better heat dissipation ratio, only problem is viscosity...=> stronger pump.
So, Please reconsider. When Overclocking, "Waterr Cooled" is the only way to go above the natural limitations... And please remember some people use Nitrogen to Overcool their sysmes. It's not condensation they have, its Ice on the Mobo 8)))))
It takes 40+ muscles to frown, but only four to extend your arm and bitchslap the motherfucker
I also work in a local shop. I see pretty much the same thing. I've replaced about 5 AMD K6-2 400's that just up and died after 2 years (these were various Compaq systems... so i'm guessing not faulty install) and the Intel counterparts(same brand) of the same era are still running fine.
Also, I've done a few upgrades with putting Duron/Althon's in people's systems.. I've personally burned 2 because I didnt have the heatsink on correctly. The machines didnt even BOOT! They burned up almost instantly... and I mean BURNED, the warranty sticker was charred!
That's not to say I blame AMD for tourched the new chips though:) that was clearly my bad. But the do burn pretty quick... i've done the same with P2/3 installs, and I was able to save the chip.
But, I'm stil buying AMD. The chips are great, fast, and cheap. SO hey.. i dont mind that much
Ford showed grisly pictures of what happens to a Chevy motor when it is run without coolant at 6000 RPM until it fails...
Before you jump down Tom's throat maybe you should re-read his article. He was surprised that the AthlonMP fried, and contacted Siemens (the maker of the motherboard) and they were the ones that told him the chips probe was not good enough for such a problem. In fact it appears Siemens was wrong and it was the motherboard, or the motherboard chipset that wasn't polling the probe enough to catch the problem.
What this means is you should make sure that your AthlonXP motherboards support the use of the added temp probe, and that they are actually good enough to catch the error and pull the plug before the processor dies.
Intels solution is still the better solution because the chip has an internal circuit that handles the temp situation so no matter what the motherboard is doing it will not fry the chip.
That being said...don't be an idiot and make sure your heatsink is installed properly, and check your fan every once in a while and you will be fine with the AthlonXP chip.
I just had a good friend buy that exact chip (AMD Athlon MP 1.2GHz) and toasted it 3 weeks ago.
He calls me up afterwards all pissed off and explains that he wanted to see if it worked before putting the heatsink/fan on. So he powered up the machine and in less than 15 seconds it is smoking. By the time he is able to shut it off, it had trashed the board also.
After several minutes of trying not to laugh my butt off as he tells the story, I calmly explained how stupid that was and why.
That's funny. Tantric Linux?
Our shop here too has seen it's share of AMD chips go south. I personally am afraid to buy one. I can't say anything about the new XP chips, and that wasn't the chip Tom was reviewing either, but the older chips are just seconds from being useless garbage without serious cooling. I had a 1200 fry just into the BIOS, because I didn't put enough heatsink compound. Yes, there was a cooler and fan, but the compound was only covering about half the chip. Booted, a few seconds later, sizzle sizzle, garbage. We run several servers, ONE has a athlon in it and it sits right beside us with a thermal probe and temp readout on the front. Even with a $70 heatsink/fan (hedgehog) I don't want to run the thing with the sides of the case on. It's just too damn hot.
Maybe the XPs are better, I sure hope AMD fixed that problem because it sucks. As for the naming scheme.. well that's a whole other subject. (*Cough* Cyrix *Cough*).
Is the a record of the temperature in the rooms where these two experiments were done? Did Toms Hardware do this is a room at room temperature, and did AMD do there in a air conditioned environment? These would explain the differences? Especially if AMDs test lab is kept cool.
At the next eco-hypocrisy-meeting, count the private jets used to get to the meeting. Should be interesting to see that
A coworker of mine built an Athlon 1400 that he ordered with a heatsink online. As it turns out, he was sent the wrong heatsink and the processor fried (almost instantaneously). According to him, the ceramic package even cracked. He isnow running a "huge" all-copper heatsink. Who would I believe? Tom's website.
But it's that 5 HP motor he has attached to the fan that weighs so much.
I used to work with all kinds of custom ICs and trust me, 700F is perfectly possible. I've even seen circuits that worked pretty well with components at about 300F. When I was a beginner HW engineer I was looking into problems with a board we were making, and I wanted to see if some components were overheating. Stupidly I touched them with my finger. Ouch!. My finger touched the component for far less than a second and I was severely burned. Think "touching a soldering iron tip with your finger" type burned.
This isn't to mention all the times when a layout error resulted in powering on a system and watching components explode, fire, smoke, etc.
An electric burner doesn't heat up very quickly compared to fried electronics. The reason why is simple. An electric burner dissipates a few hundred watts over an area that's at least 5cm by 5cm. When a chip fries you can dissipate 300W over an area that's less than 1cm by 1cm. That's a lot more energy in that small area.
I've seen many components that were fried and the solder disappeared. Seeing as it takes a temp of about 400F to melt solder, it would take a lot more to flash it into nothingness.
But don't trust me. Miswire some electronics and see for yourself.
I rather suspect that OEMs heard about the article and started going back to AMD for reassurances that the consumer-grade PCs wouldn't see a higher rate of return when the bottom-end clips on the lowest-bidder cooling units failed resulting in a cheap heat sink putting a minor dent in the bottom of a bulk-purchased case.
So AMD says "We can run without a Fan." But Tom's said "AMD fries if the Heat Sink falls off." Apples vs. UltraSparcs, anybody? This isn't about actually proving the test wrong, it's about proving the chips to be safe. Change the test a bit, and AMD can do that. And then tell the OEMs that they "still give the best bang/buck so buy from us because we don't have a massive amount of revenue to tide us over through this recession..."
Why is it that everybody thinks that big corporations care about the individual geek-hobbyist consumers? _This AMD demonstration was not for you._ (Nor was the Blue Man Group, nor "NetBurst", nor the 845 chipset, nor installing your own SoundBlaster on a VIA chipset...)
A $12 Cooler Master HS/FAN will adequately keep your CPU cool. These things are supposed to get hot because they have millions of little transistors inside. Why does anyone need to use a fan or a frickin' air conditioning.
Really. Some people take cooling way too seriously. AMD claims that the Athlon can work fine up to 90^C or so. As long as you don't exceed 60^C or so under full load, it should be plenty cool. So what if a little extra heat can kill your CPU more quickly. What does an 1.4GHz T-bird cost now? $115?
Man I need to get some of that. Take a look at the video again right before. The player respawns and is about to fall into the ditch with 123 bullets left. (according to the video, 9 minutes later). Then when the picture comes back he's in a corridor with 124 bullets.
I'm not saying that the amd is telling the truth about it being 9 minutes, but it isn't the same sequence.
Was that a standard Athlon (Thunderbird), an Athlon MP or an Athlon XP? We already know that the old Athlons fry pretty nicely (and fast)...
I had an original Athlon 1gig, and was installing it in a new motherboard with some new RAM. Something wasn't working right with the memory on the new motherboard, so I put the processor and RAM back in the old motherboard to check and make sure the new RAM was OK but didn't want to go through the hassle of installing the HSF (clean, reapply ASII, place, clamp on with screwdriver) on the old motherboard just to immediately have to take it off and put it back on the new motherboard. When I booted without the HSF, the Athlon immediately burned up and started smoking without even getting to POST. I realize it was probably dumb to try and boot without a heatsink. I am an electrical engineer, and I actually have grown to expect complex or costly chips to have a thermal shutdown in them. For the love of god even a lot of $0.50 dropout regulators have a thermal shutdown feature, why wouldn't my expensive state of the art microprocessor? Seems like PC microprocessors are a fairly mature technology at this point, and I am aghast at the lack of a reasonable thermal shutdown which prevents an idiot user from easily destroying an expensive part. How much extra die space would this feature take up? I would wager it is miniscule. I love AMD, but this seems like a blatant oversight. Maybe its fixed in the new XP processors.
See title
-- We all get heavier as we get older because there's a lot more information in our heads.
Regarding about the video showing AMD's CPU did not burned up in a simulated HSF-failure tests. I'm responsible for getting that video, from a reliable source, and posted on the net. First with HardOCP's forum and then Anandtech's Forum...
It is simple as this... both CPU and motherobard (AMD platforms) reference design has internal thermal diode. Chipsets/BIOS on the motherboard has to correctly detect what kind of CPUs it is (let it be Duron, Thunderbird, MP, XP... etc.), so it can correctly calibrate and use the full functions of thermal diode.
Tom's tests most likely did not had the right motherboard to support the CPUs and therefore, CPU burned up as in his tests. Which I'm not surprised... (and no... I'm not working for AMD and I'm not a spokeperson for anyone on this issue)
People been questioning the authenticity of the video, honestly... I can't tell you much more than this... I did get this video from a reliable source and it is *made* by AMD. If Mr.Tom Pabst wasn't so famous, I probably can say something about his video (not that it isn't authentic) too... such as asking "Is he bought out by Intel? Just to to discredit AMD's technology?!"
I'm not here to prove Tom is wrong, just he could've be more thoroughly on his tests. Knowing the fact that Intel's R&D department is much much more bigger and better than AMDs, plus Intel's standardization among computer hardware is incredible. Also noting that he uses a motherboard, which it claims, to support the thermal diode of the Palominos (Chipsets not developed by AMD and board isn't made of AMD's reference design).
Again, you people have my word that the video did came from AMD... and you can believe whatever you want to believe. This is costing too much confusion and it certainly isn't necessary. A simple fact that I've point out and I hope you guys can get it as it is... Like HardOCP said... "DO NOT TRY THIS AT HOME!" Have a nice day!
(Only reason I removed the main download link is because Apache, most likely, had a bug that crashed my OpenBSD server. For it not to happen again, I removed the link. =)
Terry "quad3d" Wang
http://www.fastclocking.com
I just went to tomshardware.com, and he (well, actually Frank Vöelkel) has an explanation as to what is in the newer video, and how it can be explained that their test blew the Athlon.
In short, because the Siemens motherboard (as all other motherboards on the market now) do not protect the CPU from thermal death, but that AMD has some new design guidelines which can protect it with the addition of a chip.
Go read it yourself here.
http://www.tomshardware.com/column/01q4/011029/
There needs to be an external hookup to have the CPU not fry.. all it does is basically shuts off the system abruptly causing you to lose whatever you were working on(equivalent to a power outage).
--iamnotayam
There is a response to AMD's video at tomshardware.com here.
-AC
About a year ago I had the pleasure of experiencing my one and only fan/heatsink failure in all the thousands of computers I've dealt with. The failure, in this case, came from my own stupidity in letting the college morons who worked in the back of the local computer store install the motherboard, cpu, and fan/heatsink since I was too lazy to do it myself. Never again.
It turns out that although they placed the fan/heatsink atop the cpu, they forgot to actually secure the clips in place. So my computer was chugging merrily along when I heard this 'whump' as the fan/heatsink assembly dropped off the CPU and onto the bottom of the case. I was laying on the couch watching TV at the time and it took me a few moments to match the probable cause to the sound itself. About 55 seconds, I'd guess; then another 5 seconds where I yelled "oh shit!" and dived for the computer to shut it down.
This was an AMD 1.2 ghz chip, supposedly very hot; and yet it took no damage whatsoever from going 60 seconds without being cooled. ASUS probe, a software monitoring program that worked with my old (then new) motherboard, didn't even sound an alarm, which it would've done had the CPU temperature exceeded critical limits (about 85 C). And even had the CPU passed the critical point ASUS would've shut it down anyway.
In any event, no damage at all from going completely uncooled for a minute. The temperature didn't spike uncontrollably, the CPU didn't melt, all the doom that anti-AMD folks warned of didn't come to pass. Everything was fine.
I'd still recommend monitoring software that'll allow you to turn your computer off or halt the CPU if things go bad. But I'd recommend this for any machine regardless of CPU brand. And I now know for a fact that my CPU can live completely uncooled through boot until ASUS loads and evaluates it, so I have no worries about 'sudden heat death'.
Max
My god carries a hammer. Your god died nailed to a tree. Any questions?
I don't see why it isn't the case manufacturers responsibility. Dedicated hardware and a small sensor set up next to the processor (or even as a small bimetallic strip between the processor and heatsync to measure temp. would be capable of reading a bad tempeture in a fraction of a second (basically, the moment the strip heats up), and could cut power to the entire PC. With some clever design on the part of the motherboard manufacturers(just placing a sensor on the bottom of the CPU in the ZIF slots middle), high temps. could be detected even faster.
:)
...I have to agree with every post which says "heatsyncs don't fall off!", since any decent motherboard and any decent heatsync should be designed to not fall off in the first place(cheap plastic or cheap tin is the only real reason they should break).
On the same note, did anybody ever hear about the research into heatsyncs which AMD was doing which let a K6-2 run without a fan? That could be useful if applied here.
It's been a long time.
Fortunately, AMD honored the warranty on the chip because I was using it in an "approved" configuration, and replaced it with a 1.2Ghz (!) chip for free. I'm not sure if Intel would do something like that (especially given their financial shape). But it has definitely made me a more loyal AMD customer, despite their debatable thermal issues.
-all dead homiez
in toms hardware they dident have any saftey systems on the MB turned on that shuts down the pc if the cpu gets too hot so the cpu went up in smoke.. the AMD one had it set up correctly where the cpu gets too hot the system turns off. I have 3 AMD athalon systems all of them are diffrent bords and have the saftey feature.
Daddy would you like some sausage?