Delidded Core i7 7700K runs 26 Degrees C cooler
Click here to post a comment for Delidded Core i7 7700K runs 26 Degrees C cooler on our message forum
coth
My 2500K was running with passive Corsair H70 for a few years. It was 45-50C in idle and 60C in load.
LevelSteam
I really hope Zen puts AMD back in direct competition with Intel again. This is just getting silly.
Enticles
I know it is pure speculation, but i would like to believe that Intel has been at least tinkering with their TIM compounds since their TIM and IHS combi has been questioned by the community with Ivy Bridge. To go from a ~5c reduction in temps to a ~26c reduction with Kaby is just plain crazy.
If these results weren't fabricated, they should at least delid another Kaby CPU and see if the results with the second CPU are in line with the first. Others have suggested the amount of TIM being the issue, if that is the case it will be the same across the board - considering the automated manufacturing processes they use.
stereoman
Every chip's different 26 degree drop after delid is massive tho, I saw an average of 13 degree difference after delidding my 4790k which was enough to run 4.8ghz 24/7 so it's definitely worth the trouble but it's a shame to have to resort to this and the fact that it voids the warranty as well.
you can see my results here http://forums.guru3d.com/showthread.php?t=409984
PrMinisterGR
Thanks for the clarification man.
As for the subject itself: CPUs are literally the cheapest high performance hardware made. They are literally just the unit and it's connections to each socket. Intel cheapening out on SOLDER of all things, is inexcusable.
It reminds me of the conversation I had with eclap in the Zen thread. The Intel that made Core 2 and Sandy is not the same Intel of the last 3-5 years.
GeniusPr0
Broadwell e is awesome. I load at 30c with H105 fans running around 1300rpm.
These cpus are best used in a mobile space I guess since they come without IHS.
Neo Cyrus
Do you think it'd be worth the hassle of de-liding in my case? My 4770K absolutely maxes out at 4.5GHz, nothing I do can push it further, and I actually underclocked it down to 4.3GHz, since whatever my settings for 4.5GHz were, it was hitting as high as the 80s, 90C in stress tests.
The thing is, even without temperature problems, it takes a massive amount of voltage compared to any review sample. Turns out Intel sent cherry picked 4770Ks to all reviewers, the real retail samples almost all ended up like mine based on everything I've seen. I've seen a lot of people report not being able to get past 4.2GHz, meanwhile every review I saw hit 4.8GHz+ with far lower voltage.
At 4.3GHz with whatever my crazy voltage setting is (haven't optimized it, it's overkill right now) my temps are tame, I don't think it ever hits any higher than 60C in any real-world stress scenario.
waltc3
If merely delidding the cpu would effect this kind of a massive temp change to the better--it wouldn't make sense for Intel (or AMD) to sell such a cpu in the customary lidded form factor. Interesting though...have to be other factors involved, I'd bet.
cyclone3d
The way IHSes came into being is because people were damaging the CPUs too easily.
Intel and AMD both used to sell desktop CPUs that had no IHS.
Intel did during their Pentium III line.
AMD did it on their Athlon/Duron and AthlonXP lines.
All laptop CPUs still have no IHS because the IHS is just another layer of TIM and metal that the heat has to go through before it gets to the heatsink base.
Not having an IHS makes for much better temps unless the IHS is soldered to the CPU core... and even then, lapping the IHS to make it flat helps a bit.
Gaidax
To be honest, I do not understand how this even should be a thing, they spend billions on optimizing this **** only to have some dude pop the top and use some better, smarter placed paste there slashing temps?
I mean really? It does not matter if it's 20 degrees, 10 degrees or even 5 degrees, it simply should not be the thing!
Amx85
Intel TIM sometimes is trash...
Remenber Haswell before "Refresh" had higher temps than Vishera lol
I hope Zen be great overclocker but i will use stock clocks for 24/7
HeavyHemi
I'd have to see more than one example and more rigorous testing. The only thing that kind of stood out was the TIM seemed to be a pretty thick layer between the die and lid.
Yes they spend all that money optimizing the product for the most profit. For consumer use, the vast majority would not care or know about this. It's a niche concern. However the niche does make a lot of noise, which is good and sometimes leads to a better product.
Elder III
That's a pretty drastic difference.... I'd like to see if this can be replicated with other units to see if that particular cpu was a dud or if something's not being reported accurately here. I don't suppose Intel would let HH delid their review sample sometime early next year? 🤓
icedman
^would be funny if we found out the review sample was soldered, I could totally see that lol.
pegasus1
Not really comparing apples with apples by using different coolers.
However Intel did go through a phase of concave heatspreaders so the temps of a stock chip can be improved on.
TheDeeGee
So what did Intel use? Toothpaste?
pegasus1
Google Toothpaste for TIM, its works, just dries out, but is minty fresh.
AnkaMusik
Out of the AnandTech article
EDIT:
Out of curiosity I googled 7700k delid and saw a number of websites reporting on my findings of the 30 degree temp drop, but were unsure on cooling method in both tests so thought I should clarify: The before and after tests were 100% identical (with exception of delid) to isolate the delid, both tests used:
The same cooler at the same setting: Kraken X62 - Fan @ 50% and pump at @ 65% (by the way, this cooler is only about 3 degrees cooler than the H110i, but has better config available)
Exactly the same BIOS settings - which was a x50 multiplier and voltage offset with LLC level 2 resulting in a stable 1.344v under prime 95 load
Memory stock (C15) and Uncore stock.
Prime 95 v27.9 run for 15 minutes in both tests.
Denial
People keep saying it's the TIM Intel is using, it's not.
https://forums.anandtech.com/threads/delidded-my-i7-3770k-loaded-temperatures-drop-by-20%C2%B0c-at-4-7ghz.2261855/page-23#post-34053183
The IHS adhesive they use on the edges is too thick and creates a gap. Pulling the IHS off, removing the adhesive on the IHS and reseating it with the same TIM is just as effective as replacing the TIM.
They need to find a different way to mount he IHS onto the CPU without the gap. They stopped using solder because they found that multiple thermal cycles on smaller die chips were causing voids/micro cracks in the solder - dropping thermal connectivity and in some cases completely severing the connection of the chip to the solder.
user1
I have a hard time believing that intel would overlook something as simple as Ihs spacing after 3 generations of cpus, something else must be afoot , even kaveri apus that use tim aren't that bad.
