This Cooler Might Kill Your CPU

[AlexTheGreatish]

EK just released a direct die cooler for LGA 1700 CPUs. Will we finally be able to tame the i9-13900K, and will it survive the day at all?

 

Buy an Intel Core i9 13900K Processor: https://geni.us/tqwraq

Check out the EKWB EK-Quantum Velocity² Direct Die D-RGB 1700 Nickel + Plexi: https://lmg.gg/e4Uvr

Purchases made through some store links may provide some compensation to Linus Media Group.

 

 

 

[Lurick]

Opportunity missed to say

"Will this CPU see tomorrow, will it even see this segue to our sponsor"

[Gokul_P]

It is not the coolers fault it is more like users fault. Yes hitting with a hammer do have a possibility of killing your cpu. 

[tkitch]

Just now, Gokul_P said:

It is not the coolers fault it is more like users fault. Yes hitting with a hammer do have a possibility of killing your cpu. 

little of both.

Needing to delid a CPU to use a cooler is an inherently risky thing.

[RandomLegoBrick]

12:40, good job editors lol

[dogwitch]

2 hours ago, Gokul_P said:

It is not the coolers fault it is more like users fault. Yes hitting with a hammer do have a possibility of killing your cpu. 

true. but cpu had way more glue on ihs. then normal.

[Ubersonic]

Considering the way so many TechTubers have switched from saying MHz to MTs in relating to RAM and adopted similar corrections.  It is somewhat surprising that so many still refer to the weight spreaders on consumer CPUs as heat spreaders, it's a misnomer that's been going on for 25 years now and seems to have no ned in sight.

 

It's actually quite funny as when AMD/Intel first introduced them they got rinsed for calling them heat spreaders when they were actually just thermal insulators designed to spread the weight (dies had been cracking due to sizes getting smaller and a solution was needed), but then over time people just accepted it.

[Nystemy]

54 minutes ago, Ubersonic said:

Considering the way so many TechTubers have switched from saying MHz to MTs in relating to RAM and adopted similar corrections.  It is somewhat surprising that so many still refer to the weight spreaders on consumer CPUs as heat spreaders, it's a misnomer that's been going on for 25 years now and seems to have no ned in sight.

 

It's actually quite funny as when AMD/Intel first introduced them they got rinsed for calling them heat spreaders when they were actually just thermal insulators designed to spread the weight (dies had been cracking due to sizes getting smaller and a solution was needed), but then over time people just accepted it.

The lid has a few reasons why it exists.
Spreading the mounting pressure is indeed one reason. However, not applicable to ZIF sockets like the AM4 one and earlier. _{(LGA sockets however require some serious mounting pressure.)}

Spreading heat is also useful, but I will agree that it really doesn't seem to be a major concern. _{(saturating heat pipes could be a valid concern, but soldered heatpipes on a heatsink base plate is generally going to handle all the necessary heat spreading just fine by itself, so the IHS is technically redundant.)}

Though, protecting the die from damage is also quite the value add as far as Intel/AMD is concerned. Replacing damaged products isn't really going to give one a good image. _{(Since if the products are easily damaged, then users thinks that is bad. If the product doesn't get covered by warranty due to heavy handed end users, then the end users will consider it bad. If the product gets replaced under warranty, then the vendor has to bake in the cost associated with the dead units into the product price, and end users don't like paying more. In the end, adding a lid to protect the most sensitive part is easier, most people happily shrug their shoulders at slightly worse cooling.)}

But it would be nice if Intel/AMD sold a "direct die cooling" version of their CPUs. _{(Likely having a shim around the die to help with mounting pressure, similar to what we can see on a lot of GPU packages. But for a CPU it would be more crucial to system stability due to the socket needing good contact pressure for all its pins, not just the pins beneath the die.)}

[JeffLo]

lots of people having problem with EK's Direct Die and i think i found the flaw
please look into the DD cold plate its machined convex which is good for proc with IHS but for direct die im sure you dont want a convex plate you need
it to be as flat as posible for perfect full contact and also you dont crush the middle part of the die
 

[Needfuldoer]

Are you guys ever going to make up a proper heavy duty extension cord so you don't have to mess with the janky one, or is that mostly-safe solution just ingrained as a running gag at this point?

 

(Saying this as someone who's made more than their share of sketchy "just make it work" extension cords.)

[dogwitch]

15 minutes ago, Needfuldoer said:

Are you guys ever going to make up a proper heavy duty extension cord so you don't have to mess with the janky one, or is that mostly-safe solution just ingrained as a running gag at this point?

 

(Saying this as someone who's made more than their share of sketchy "just make it work" extension cords.)

a fellow just make it work.... nice

[Amaranth]

I actually find the 'pop' when the IHS detaches rather satisfying, although I've also delidded a fair number and might have become desensitized to the innate feelings of 'my god, did I break my CPU' that many people seem get at that stage of the process.  

[Spindel]

Ouch!

 

The picture of the chipped AMD Athlon brought back painful memories. 

 

I've mentioned this in other threads on this forum but I'll say it again:

Back in the days I had a Athlon Thunderbird 1333 MHz and when installing a new cooler on it I dropped it and chipped a corner of the die, instantly killing it. 

It's a painful memory since this was when I was younger, still at school and I had really limited amounts of money to get a new CPU for. 

 

But in general, specially nowadays, I don't understand why CPU manufacturers put IHS:es on CPUs. Sure they protect the CPU from clumsy idiots (like me). But other than that I see no benefit to them. I don't agree with @AlexTheGreatish point of risking to hit heat-pipe coolers to hard, for the heat pipes, since in most heat-pipe air coolers the heat-pipe does not touch the CPU directly anyway but is soldered into a metal plate that acts as a heat spreader anyway.

 

Heck I remember in the Athlon days that the CPU coolers had the horrible spring loaded cooler mounting system where you pivoted the cooler in place and then needed to apply a really uneven pressure to be able to make the spring clamp to catch on. And at least I personally don't know of anyone that chipped the die while doing this (but I'm sure it happened). this compared to todays mounting systems that are way easier to get a fairly even pressure while mounting (since you usually screw in four screws in the corners).    

[FPC Gaming]

Having the same problem as you with the hot core, tried mounting four times and still not great it's got to be something to do with the pressure or Liquid Metal ( to much of it maybe I really can't work it out) idle temp are good around 29c game 88c on the hottest core 4 more are high 70c then all the rest 50 to 66c. Any help would be great

Attached is a screenshot of temps game fyi

Any help would be great.

[AlexTheGreatish]

On 5/3/2023 at 10:21 AM, FPC Gaming said:

~snip~

I honestly have no clue.  I've had a bunch of trouble getting it to seat correctly and don't have a solution besides just remounting it a bunch of times until it works.