TR Forums

When I bought my first phase-change cooler, I had a Phenom II X3 720 that I wanted to test.

I had to build a bracket, and it didn't fit at first. The CPU was running essentially without any heatsink for a few tens of minutes while I installed the OS because I didn't realized the bracket I'd built was sitting on a capacitor I didn't see. After nearly an hour of running the OS booted (Windows 7, IIRC) and I realized the CPU temp was at 100C. No damage at all, the bracket was fixed, and 3 C temps were achieved quickly.

That CPU still lives today after running at > 4 GHz for a few months and around 100 C for a few days when the phase change cooler lost its compressor and ran essentially on radiation alone.

CPUs are sturdy beasts.

Waco wrote:

When I bought my first phase-change cooler, I had a Phenom II X3 720 that I wanted to test.

I had to build a bracket, and it didn't fit at first. The CPU was running essentially without any heatsink for a few tens of minutes while I installed the OS because I didn't realized the bracket I'd built was sitting on a capacitor I didn't see. After nearly an hour of running the OS booted (Windows 7, IIRC) and I realized the CPU temp was at 100C. No damage at all, the bracket was fixed, and 3 C temps were achieved quickly.

That CPU still lives today after running at > 4 GHz for a few months and around 100 C for a few days when the phase change cooler lost its compressor and ran essentially on radiation alone.

CPUs are sturdy beasts.

Or at least they WERE up until around the turn of the past decade. I haven't done any "unplanned thermal torture tests" of CPUs since the Phenom generation. Maybe when I finally retire my secondary rig (FX-8320) or file server (1090T) I should do some planned thermal torture, just for grins...

Well I have to admit I tortured my 6700K with a block that didn't quite fit over a capacitor on the board I didn't see. Hmm, recurring theme here.

It would hit 95+ C under load and I swore it was a bad thermal paste job under the IHS. One very attentive wife's comment later, and I had a CPU that would run 4.5 GHz at 50 C.

So at least Intel is still creating CPUs that don't mind stress.

Sounds like your wife could be a closet PC enthusiast. If she'd had a YouTube channel maybe she would've gotten a Threadripper review kit.

She's written far more reviews for PC hardware than I ever did...so not really closet.

I can't say I'd hate a free ThreadRipper for 6 hours of work or so though.

Waco wrote:

She's written far more reviews for PC hardware than I ever did...so not really closet.

Oh... my bad!

So why doesn't she post here?

Edit: My wife does not share my tech inclinations, but she does appreciate (and brew) good beer, so we have that hobby in common!

just brew it! wrote:

Waco wrote:

CPUs are sturdy beasts.

Or at least they WERE up until around the turn of the past decade.

Same seems to apply to storage media.

I had one of those Athlons with an Alpha PAL6035 heatsink. After the fan failed, the heatsink go hot enough to deform the solder holding its pins in place. The CPU was fine.

Forge wrote:

derFunkenstein wrote:

Heat death was a real thing back in 1999, since you could fry Athlons. Since then, AMD has learned its lesson and the CPU is much more resilient.

Athlon MP 1.2GHz, ran with a failed fan for between 2 and 3 months at 120-140C throughout. Only noticed when smell of carbonizing dust became notable. Fan later replaced, and CPU continued for several years of issue-free useful work.

I flatly deny the THG-inspired stories of heat death! Barring removal of the heatsink, Athlons simply did not cook to death! If removing the heatsink was allowed, there are LOTS of P3s and P4s that would also fry to death, thermistor or no.

Yeah, well, heatsink removal was a real danger. I had a Biostar Socket A board where one of the socket tabs broke off. Definitely cooked it.

just brew it! wrote:

Waco wrote:

She's written far more reviews for PC hardware than I ever did...so not really closet.

Oh... my bad!

So why doesn't she post here?

Edit: My wife does not share my tech inclinations, but she does appreciate (and brew) good beer, so we have that hobby in common!

Brewing beer win! The wife goes by Bluepanda but she doesn't post much.

derFunkenstein wrote:

Forge wrote:

derFunkenstein wrote:

Heat death was a real thing back in 1999, since you could fry Athlons. Since then, AMD has learned its lesson and the CPU is much more resilient.

Athlon MP 1.2GHz, ran with a failed fan for between 2 and 3 months at 120-140C throughout. Only noticed when smell of carbonizing dust became notable. Fan later replaced, and CPU continued for several years of issue-free useful work.

I flatly deny the THG-inspired stories of heat death! Barring removal of the heatsink, Athlons simply did not cook to death! If removing the heatsink was allowed, there are LOTS of P3s and P4s that would also fry to death, thermistor or no.

Yeah, well, heatsink removal was a real danger. I had a Biostar Socket A board where one of the socket tabs broke off. Definitely cooked it.

That is precisely why I stayed on socket A for so long, and I only ever bought boards with the mounting holes. I took a power drill to my NF7 boards literally on the day I got them, since Abit didn't drill out the pads for some bizarro reason. Now through-board mounting is de rigeur!

Weird, I don't remember seeing any mounting holes on Socket A until the nForce 2 chipset. The machine I'm talking about had a KT133 chipset. Fortunately I "only" lost a Duron 600. Unfortunately, that's the Duron 600 I took to 1GHz.

just brew it! wrote:

Going a little further back, I also had a K6-III+ that ran with a failed fan for a number of days/weeks/months (really have no idea now long it was), without showing any signs of trouble. I only noticed the fan was completely seized up when I opened the side of the case to install additional RAM. IIRC it was even running the distributed.net client at the time, so 100% CPU load 24x7.

I once ran a Celeron-533A without HSF installed for 20 minutes before it overheated. Worked fine for many years afterwards.

derFunkenstein wrote:

Weird, I don't remember seeing any mounting holes on Socket A until the nForce 2 chipset. The machine I'm talking about had a KT133 chipset. Fortunately I "only" lost a Duron 600. Unfortunately, that's the Duron 600 I took to 1GHz.

I definitely had a KT133 or KT133A with mounting holes, because that's how I killed my golden Dresden Athlon. Athlon 600, pretested to 1.2 or some other twaddle, and I MASHED THAT DIE TO DEATH while screwing down my waterblock.

Good times.

Well that's cool. I just wasn't aware it was an option that early in its life.

Yep, good times, me and my KT7-Raid killing CPUs and burning things.

http://images.anandtech.com/reviews/mot ... a/abit.jpg

At some point I decided to stick to HSFs that used all 6 of the tabs. Thermalright was good (but a little pricey), and Taisol made a relatively inexpensive one with the 6-point clip.

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

Captain Ned wrote:

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

God...reading that made me cringe. I only ever gauged a single board, but it's a memory I won't forget.

Captain Ned wrote:

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

Oh yes. At least once, the screwdriver tip slid out of that damn recess and skittered a scratch across a dozen motherboard traces. It still worked, but damn if I didn't hate that stupid setup.

Captain Ned wrote:

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

Yup. *ALWAYS* use the big flathead screwdriver. NEVER EVER the one on your Swiss Army knife that doesn't lock down.

I was a somewhat foolhardy young padawan.

Captain Ned wrote:

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

The trick was to take some of the bend out of the mounting strap (they were always too tight) and use a pair of thin jawed needle nose pliers to attach the clips.

Captain Ned wrote:

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

I must have been lucky+careful since I never slipped a screwdriver and scratched the mobo. Though I worked it really slowly, tight as those things were.

I worked in a local PC shop in 1999 and 2000. When Socket A came about, we actually took the HSFs apart and tried to relieve a bit of the pressure from the clip. And it worked - temps were fine and it was a little easier to install the heatsink. Right or wrong, I dunno (probably wrong) but it made for better builds.

Captain Ned wrote:

Those of us who lived through Socket A will never forget. Those who didn't will never understand just how hard you had to press down on that screwdriver to get the clip to lock and the concomitant feeling of imminent disaster.

Dual Socket A was even more fun. Especially considering the board cost.