When I was researching the mess related to strange and poorly documented CPUID bits in the Athlon MP and XP processors, I had to swap a lot of CPUs. This was uneventful in most cases… but only most.
When I removed the heatsink from one unlucky Athlon XP, the processor looked like this:
And this is what the heatsink looked like, with a piece of the CPU glued to it:
There are two interesting things about this. One is that the CPU worked just fine until a good chunk of it came off. The other is that removing the heatsink did not need any particularly excessive force, although some heatsinks do have a tendency to get stuck on.
Based on the shape of the “extracted” piece of silicon, I suspect there was a relatively long and straight micro-crack in the silicon which did not noticeably impact the operation. But when force was applied, the micro-crack gave way, and then a whole big chunk of silicon came off.
Notice how the right-hand side of the gouge in the CPU is very straight, whereas the left-hand side shows typical fracture marks.
It is notable that around 2000, both Intel and AMD used flip-chip PGA packaging in an effort to provide better cooling (something both companies struggled with as the TDP of the processors quickly shot past 50W and approached 70-80W).
Both companies, especially Intel, gave up on this type of packaging relatively quickly. Intel only used it for some PIII models and switched to lidded CPUs for the P4 line, as well as the PIII-S processors. AMD used flip-chip packaging for PGA Athlons but not for Opterons.
The exposed silicon was a little bit too fragile and required assemblers to install heatsinks very very carefully—if installing a heatsink applied pressure that was too uneven, the silicon would crack. Many vintage flip-chip CPUs have chipped corners, although that usually does not affect operation.
Processors with lidded packaging still provided very good cooling, but compared to CPUs with exposed silicon they proved much sturdier and far less susceptible to mechanical damage. Especially Intel’s pin-less LGA processors are quite sturdy and not at all prone to mechanical damage, although the weak spot simply moved to the motherboard socket instead.
Overlay it with its die picture, to see where that straight-line crack falls on it…
>Intel only used it for some PIII models and switched to lidded CPUs for the P4 line, as well as the PIII-S processors.
My experience from that time was Intel somehow made their CPU cores much more durable. Not only were dies harder to crack, they also had greater chance of working with chipped corners. AMD ones on the other hand were very brittle thus inventions like special shims just for Athlons/Durons while no one peddled such stuff for Intel.
Afaik cant really do this to greater degree to already processed die, so it must have been special sauce in Si ingots.
Did you put it back and tried it out? I know there’s 0% chance of that working. It just can’t, an uncountable amount of transistors and connective fabric will have become dust.
You should do it anyway.
@rasz_pl
Yeh, Coppermine CPUs were a lot more durable. Also their die was slimmer than AMD’s.
I’m not confident that it couldn’t damage the board, and I don’t have a lot of Socket A gear. So I’d rather not…
I’ve never had that happen, but after reading stories of others who managed to pull a socket 478 P4 out of its socket, bending quite a few pins when it stuck to the heatsink, my standard procedure for removing CPU heatsinks is to do it while the system is warm, and to gently rotate the heatsink back and forth to break the bond with the thermal paste instead of pulling straight up.
Silicon is brittle but not exactly weak; I would suspect as you mention that the bulk of the damage had already occurred before you pulled the heatsink off, possibly thermal stress from repeated uneven heating/cooling. Looking at some die photos, there is a noticeable feature line around 1/4 of the way across from one edge, roughly where the straight side of your crack is.
Defective Athlons bothered me too.
It looks like you kept very old thermal paste on the CPU heatsink. That might have been causing the damage, if it has some hard spots or is sticky.
It could have been a factor, although with all the CPU swapping I did I somewhat doubt it.
The straight line of the crack does line up pretty well with a line visible on the Palomino die shots, so that was almost certainly a factor.
Hi Michal Necasek
Did you ever work for CaseWorks in the late 1980’s??
Asking for a friend.
Definitely not, no. I wasn’t really even in high school.
I worked with a guy named Michal Necasek at CaseWorks from 1986 to 1989.
I thought the last name to be unique as I had never heard of another Necasek.
I was hoping….. Although he was about 10 years older than me so he might not be around anymore as I am 61.
Fascinating. Necasek is a Czech surname and it is indeed quite rare. Even in Czechia there’s maybe only a dozen or two, with another handful around the world (some spell their name as Nechasek). Most likely your coworker was a very distant relative of mine.
Intel actually started out using soldered IHS packages, but then for the Pentium 4 and others switched to using a cheaper bonding thermal glue instead of solder (you can tell the difference of packages whether or not they have that rubber sealant, soldered IHSes do not have it). I believe today they finally went back to soldered IHS? Not that… anyone would ever buy an Intel CPU today :p
One guy I knew got one of those AMDs with the exposed dies and set it up with water cooling. Well…. the water wasn’t cycled in the loop and didn’t make it to the die in time and the CPU fried itself since there was no thermal mass for the excess heat. Needless to say he never built water cooling again. hahaha.