Well I guess that's proof that even the Athlon XP's thermal protection is nearly worthless compared to Intel's of the time.

AXP added an on-die sensor but this must mean that the CPU doesn't pay attention to it. Perhaps only the BIOS monitors it and it reacts too slow for this situation.

Something similar happened to me a while ago. The heatsink was not properly fitted on the motherboard...although it was clamped, it fell off. Less than 5s after that the capacitors on the mobo around the cpu socket blew and the cpu went up in smoke. That was a XP-2800+ 😀
AFAIK XP series did not have on-die sensor (or perhaps it was disabled or non-functional?), instead the motherboard had a sensor in the middle of the cpu socket so it would touch the cpu from below. Of course it was useless in cases like this 😜

Palomino added a sensor. It was the Thunderbird and older chips that didn't have them.

http://techreport.com/articles.x/2975/1

I think that AMD didn't have any sort of thermal protection though. So the sensor was there for reading but there's no mechanism to react to overheating. I know that Socket A boards almost always have BIOS shutdown options though, but they are surely too slow for this situation.

Hmmm double post 😒 please delete

Yeah this was pretty well documented at that time, I knew a couple of people who didnt install their heatsink properly and smoked their CPU. Crushing your CPU core when fitting a heatsink was also more common than it should have been.

This video was doing the rounds at the time http://m.tomshardware.com/reviews/hot-spot,365-6.html

The P4 was definitely safer and easier to work on but the netburst architecture was a total dog IMHO 😒 I owned a an athlon xp 1800+ and a 2800+ while my friend went the p4 route, and I was never overly impressed by intels offering at the time.

Perhaps RG100 could do a scorchmark test with different Athlon-XPs from 1700+ up to 3200+. A graph of survival time vs. performance rating would make my day.

swaaye wrote:

Palomino added a sensor.

Didn't work for me 😜

I fried one with heatsink installed properly, but forgot to apply any TIM 😵

It's there to read temps, not save the CPU apparently.

retro games 100 is officially our hardware destroyer! any junk you want to get rid of, send it to him!

swaaye wrote:

It's there to read temps, not save the CPU apparently.

Lol, whats the point of a sensor reading the temps if the CPU burns up so quickly that the display doesn't even have time to display anything? 😜 😜 😜

I guess it would be useful if your fan was geting gunked up or failed? the heatsink alone may be enough to keep it alive for longer than a few seconds i'd imagine.

When the idea of a temperature sensor was conceived i doubt they ever considered it would have to deal with people as skilled as RG100!!

To protect the CPU without heatsink they would need many temperature sensors spread around the die as it quickly heats up locally.

I have found a 2 minute youtube video here, containing relevant information to this discussion.

retro games 100 wrote:

What is the life expectancy of an XP-M without a heatsink?

retro games 100 wrote:

I have found a 2 minute youtube video here, containing relevant information to this discussion.

Yup, saw this before. Quite a feat that the Coppermine also survives.
All in all, we'll just have to accept that Athlon (XP) simply can't that this and contemporary Intel's can.

Edit:Not sure if it's related, but the Coppermine does produce a LOT less heat and the P4 (along with dynamic underclocking) has an integrated heatspreader. Could have something to do with it.

Anyone wanna sacri...I mean test a horribly underclocked Socket A CPU? 😁

I wouldn't do that. Even if the CPU ist on death row, the mainboard may go down as well.
A heating of the cpu will increase power drain - caps may blow.

You know in retrospect for Slot A Athlon vs. Coppermine, I think the enthusiast world was way too thrilled by Athlon. You got a much hotter CPU with really shitty motherboards and performance that was essentially the same or intangibly better.

Tbird was lucky that Willamette sucked hard and was hot too. 😀 Although once Willamette got to 2 GHz I think Athlon had been matched overall. And Northwood is definitely a match for Athlon XP.

But Athlon was a giant improvement over K6 junk, aside from the still awful motherboards. I suppose most of the excitement came from that aspect. Intel actually had competition in every application.

retro games 100 wrote:

I have found a 2 minute youtube video here, containing relevant information to this discussion.

This is the same video I linked to above 🤣

Tbird was lucky that Willamette sucked hard and was hot too. Although once Willamette got to 2 GHz I think Athlon had been matched overall. And Northwood is definitely a match for Athlon XP.

Personally I disagree. It wasn't really until the 800mhz FSB northwood c came out that the p4 really matched the athlon XP. The p4 might have won in benchmarks but the athlon. XP always seemed perceptibly faster to me when actually sitting in front of the PC.

they "won" benchmarks because of that intel compiler heist

the early northwoods weren't very impressive... I had a dell precision 360 with one of the last/latest northwoods made... I think it ran 3ghz, the speeds still weren't that impressive although it was faster than my overclocked xp 2100+

I think the prescotts were more evenly matched to the athlon xp even though they were massively inefficient... super high clock speeds, ran hot, but did a whole lot of no good!