Du verwendest einen veralteten Browser. Es ist möglich, dass diese oder andere Websites nicht korrekt angezeigt werden. Du solltest ein Upgrade durchführen oder einen alternativen Browser verwenden.
Also ganz ehrlich. Das klingt nach Schnapsidee.
Warum von 13700K auf 14900KS?
Batches sagen gefühlt bei KS noch weniger als K/KF aus.
RAM 7800 zu 8000 macht jetzt nicht den Mega Unterschied aus.
Wenn Du diese Anzeige nicht sehen willst, registriere Dich und/oder logge Dich ein.
Also ganz ehrlich. Das klingt nach Schnapsidee.
Warum von 13700K auf 14900KS?
Batches sagen gefühlt bei KS noch weniger als K/KF aus.
RAM 7800 zu 8000 macht jetzt nicht den Mega Unterschied aus.
Also ganz ehrlich. Das klingt nach Schnapsidee.
Warum von 13700K auf 14900KS?
Batches sagen gefühlt bei KS noch weniger als K/KF aus.
RAM 7800 zu 8000 macht jetzt nicht den Mega Unterschied aus.
Das Update will ich machen um den ddr5 zu optimieren. Habe eh den thermal boost so eingestellt all core p-core 5,0 e-core 3,0 GHz haben.
Update ist nur wegen des besseren MC
No need to question if you can have a question.
Just ask.
PTM is bad for direct die.
LM is bad overall
Kryosheet for Crystal, PTM7950 for IHS+a lot of pressure.
GPUs may be ok, but PTM requires a lot of pressure.
so basically Kryosheet > LM > PTM on CPU DD? I've never used the Kryosheet before, anything I should know beforehand other than it's conductive?
what is the advantage over good ol' Conductonaut and its friends?
Das Update will ich machen um den ddr5 zu optimieren. Habe eh den thermal boost so eingestellt all core p-core 5,0 e-core 3,0 GHz haben.
Update ist nur wegen des besseren MC
so basically Kryosheet > LM > PTM on CPU DD? I've never used the Kryosheet before, anything I should know beforehand other than it's conductive?
what is the advantage over good ol' Conductonaut and its friends?
Wenn man sich den Kühler erhalten will, vielleicht. Ansonsten ist LM immer noch deutlich vor Kryosheet bei Direct Die Anwendungen und PTM ist ca. auf dem selben Niveau wie eine vernünftige Wärmeleitpaste. Dass das Zeugs ewig halten würde ist allerdings auch nur ein Gerücht. Ich habs getestet und nach knapp 3-4 Monaten verliert PTM nach und nach minimal an Leistung, da habe ich mit Noctua NT-H2 und der alten CM Master Gel Maker bessere Erfahrungen. Mag sein, dass es über Jahre hinweg dann eventuell besser performen würde, aber so lange Wartungsintervalle habe ich nie und bezweifeln würde ich es auch.
Wenn du ständig mit Temperaturen oberhalb von 85° unterwegs bist, dann würde ich generell von PTM und den meisten gängigen Pasten Abstand nehmen, da die alle zu pump out neigen wenn sie ständig hohen Schwankungen ausgesetzt sind. Ist also eher ne Frage welchen Temperaturen die CPU ausgesetzt ist und was so deine Wartungsintervalle sind.
so basically Kryosheet > LM > PTM on CPU DD? I've never used the Kryosheet before, anything I should know beforehand other than it's conductive?
what is the advantage over good ol' Conductonaut and its friends?
I used to use the Shamino RTTs that Veii recommended, with an adaptation in the RONs, RTT's 48-34-34--34-34-240-0-0-60-40-40/40-40 and DQVrefUp 172-Down -90 Black Apex, White APEX Encore's PCB has different characteristics :) RTTs on DIMM may work due to Groups But Groups & RONs are on CPUs...
Paste looses not because of thermal conductivity, but because of cavity filling ability.
A liquid fluid which makes no perfect contact, is as if not more worthless than any other substrate that fills the delta.
Air is an insulator. A very good one.
Soo because of that, if you can't even get it anywhere near sub 100 microns, liquid metal will always be the worst option.
Outside its material destructive properties, or its gravity flow-away properties.
Else PTM + lapping, if you stay on the dampened IHS.
DirectDie options don't have the required tolerances for fluids to make sense.
But supposedly ICE DirectDie is a better option.
Any case you'll decide.
I strongly recommend against LM.
Outside of destructive properties, its the half hearted approach that lets you reach the bare die.
In which you make the problem worse and worse.
The crystal isnt flat either.
PTM comes close to LM.
Kryo same, but PTM is not suited for this low pressure usecase.
PTM is useful for flat surfaces, or lazy applications with high pressure.
A good contact looks like a mirror imprint, where you can clearly read the etched font of the IHS on your cooler ~ yet see no "paste" residue.
Beitrag automatisch zusammengeführt:
EDIT:
You can see that contact was not as perfect
It buffered at least >60 microns, as a good contact has to be completely flat on both sides
One could see that 2nd sample was more convex in the middle, although i did work on the cooler.
Soo while it could still get a font stamp, it was ... "ok" to say.
Fluid would maybe work, but its not flat. There is no ringing happening on such stock design. Needs much finer lapping.
Like this is the minimum for a fluid based paste, tho PTM is more fine-granular in their composition.
For direct die, no absolutely not usable. Maybe for bigger crystals like GPUs where margins are smaller.
Still requires lapping
Go with something that can bent, till you are trusting your skillset and have measure paper, to actually get it flat.
Then maybe we can consider fluid based thermal material.
As long as gravity beats you to it, material is not even close to flat enough and likely mounting system neither.
I can not recommend LM. Its way out of logic to even consider that, given the difficulties to actually get it right.
// ignoring its destructive properties to crystal and cooler
I would be lying if i tell you "its gonna be ok". Its not gonna be. You can't eyeball required flatness for fluid like TIM to make any sense.
Especially when you frankenstein mount a non direct-die cooler. @tibcsi0407 ? recently went direct die too
Maybe has a word to say about mounting pressure and this cooler in specific
I would still go with a dampening pad, and don't do half-hearted work on direct-die.
As you introduce cavities on solder removing - which is kind of the problem you are fighting here from the very beginning.
Aka you make it worse~~
I can not recommend LM, its not easy and too high chance of crystal crack.
Thanks, I'll see what I can to with the means I have available. Trying to get ahold of the Fuji paper and decide from there if it's workable for me, however that's a really good hint.
Moving away from direct die cooling, what would you replace the stock Intel TIM with? I can for sure lap the IHS but I am not so sure about the die. Already wrecked a CPU with die lapping before and if anything, the cooler will for sure work on a lapped IHS if it sucks on direct die for whatever reason.
That's great... I'm not happy right now.
The CPU I bought is uncapped. Prepared for DD. I was going to install it next week.
I have a "million" questions now.
In short, my history, as I have been using CPUs with satisfaction for years.
Buying a CPU, I like it, so I'll keep it. A friend of mine does it professionally, I send it to him, he removes the cap non-destructively with an infrared soldering iron, polishes it, replaces the factory soldering with liquid metal, and the cap is reapplied.
That's how it is now. Delided CPUs from several years still work today. We opened one after a year, nothing bad, no drying out.
I always use/lubricate the cpu with normal paste between the block and ihs. 12-15°C cooler at a normal load. For example, Cinebench.
I never read about what happens when normal paste is applied to the chip instead of LM, and then the DD block is added.
Is this version horribly bad?
At the moment, the heat conductor is still paste, since paste is applied to the delided cpu, and the waterblock is applied to it.
The block sits directly on the chip, paste does not destroy or damage anything.
Maybe it's better to de/relide, put the cap/ihs back on and use it as before?
I read today that Intel allows delid. No loss of warranty. True, only for ks models.
So even the manufacturer admits that the factory heat conduction is bad, horrible, practically uncoolable.
I know it's fine if it's 105°C 24/7, but I personally don't like it. It bothers me. That's why I use it at 5400MHz at low voltage.
Well, that's not a good start.
No idea, what better is now.
I'm waiting now for DD block from Tibcsi. Temporary is good him old Supercool dd block to test. It is not the best block i know.
Na ja, we will see...
Digg: Thx, maybe
My first DD, i ask before i do wrong something.
www.hardwareluxx.de
