Why Intel CPU's run at 95°C and why AMD's should, also

[mtcn77]

In a hypothetical universe where the best coolers we had were barely enough, then yes, in that scenario running the silicone as hot as it could endure would allow the highest amount of heat to be transferred. However back in the real world there's plenty of extremely good coolers so we have the luxury of being able to cool to lower temperatures which results in higher clock speeds, slighly less power consumed and longer life span.

Well, it is your hardware. If you like getting 80w cooling from 250w hardware, I won't be the judge of that. I'm not preaching to the choir. I made it clear that this is a patch up job; however it is with the best intention this is something people don't know about it, since it has been addressed by both vendor, journalist and aftermarket oem levels, so it is on emphasis. I cannot be the only one making this up.

 

[infrared]

I mean... can't you see that you don't want to end up using all of the cooling potential. That's the wrong place to have a bottleneck imo. Having cooling headroom is a good thing, not a wasteful thing.

 

[mtcn77]

But for me (sorry @mtcn77) its completely dumb, because it contradicts the idea of silicon cooling and preserve its integrity along with highest clocks and voltage possible.

It is okay, the gist is practically a joke at this point. However if it works, give it a shot.
To make it work, run it manually and make a custom fan curve to run only near 80-90°C. It shouldn't be too difficult to enter the necessary values in the bios. You run it at undervolt and under restricted cpu airflow.

 

[TheoneandonlyMrK]

I'm starting to feel people just like to sound smart without considering, the wheels already invented, many thousands of scientists worked millions of man hour's all to be down played by Dave( hypothetical smart person).

 

[mtcn77]

I mean... can't you see that you don't want to end up using all of the cooling potential. That's the wrong place to have a bottleneck imo. Having cooling headroom is a good thing, not a wasteful thing.

Yeah, but the heat transfer coefficients are skewed in the chip's favor. I won't go in the details at this post; however if it heats up 4 times faster than the transmittance level of your coldplate, a temperature gradient will develop irregardless of your total heat load transfer capacity.
TL;DR: if the heatpipes, or fluid in the loop doesn't get hot - until they are at thermal equalibrium(equal temperature) with your chip; you aren't making the most of your precious little cooler. Ouch.

 

[TheoneandonlyMrK]

Yeah, but the heat transfer coefficients are skewed in the chip's favor. I won't go in the details at this post; however if it heats up 4 times faster than the transmittance level of your coldplate, a temperature gradient will develop irregardless of your total heat load transfer capacity.
TL;DR: if the heatpipes, or fluid in the loop doesn't get hot - until they are at thermal equalibrium(equal temperature) with your chip; you aren't making the most of your precious little cooler. Ouch.

See with talk like this I'm getting a mindset that you just use your pc casually like gaming, my PC has sat at a thermal equilibrium temperature of 68-82 since it was bought, if you use it then it will naturally utilitize it's cooling system and therefore won't be wasting the time money and effort your putting into cooling your (and running) pc.


To have a CPU sat at a high thermal point at all times A LOAD needs to be present, some power has to be burnt.

For what, exactly to run a HSF effectively or just to kill the planet for naught.

 

[mtcn77]

I'm starting to feel people just like to sound smart without considering, the wheels already invented, many thousands of scientists worked millions of man hour's all to be down played by Dave( hypothetical smart person).

There is truth to your objection. There has not been a single successful mainstream exotic cooler in the market. I have interspersed the issues that go along with those as best as I could without naming any names; however I have been a naïve chump, I will concede that.

 

[TheoneandonlyMrK]

There is truth to your objection. There has not been a single successful mainstream exotic cooler in the market. I have interspersed the issues that go along with those as best as I could without naming any names; however I have been a naïve chump, I will concede that.

That's not right either though, a few have brought to market unconventional approaches to cooling see coolzone freezit for one, I nearly bought one , a HSF with a Tec built in and thermostatic control.

As I also said and V many skipped passed.

Both/all afaik chip manufacturers use a temperature offset value, when the temperature says 95 the Tdie would be between 100-115°c as the actual typical limit.

Which makes for a happy you since they're already doing this to a small degree.

 

[mtcn77]

See with talk like this I'm getting a mindset that you just use your pc casually like gaming, my PC has sat at a thermal equilibrium temperature of 68-82 since it was bought, if you use it then it will naturally utilitize it's cooling system and therefore won't be wasting the time money and effort your putting into cooling your (and running) pc.


To have a CPU sat at a high thermal point at all times A LOAD needs to be present, some power has to be burnt.

For what, exactly to run a HSF effectively or just to kill the planet for naught.

This is definitely the kind of health scepticism I appreciate, nothing good has ever come out of blind zealots.
You see, your intro;

See with talk like this I'm getting a mindset that you just use your pc casually like gaming,

'Casually' is the keynote here. Of course, an advanced user does NOT default to precision boost 2 & its bigger brother precision boost overdrive.
This is the merit of this thread, I'm sure you aren't having any difficulty hiding that fact from your intellect with much conscientious scrutiny.
It is the default settings of the general community and how these default settings are presented in bios is the point of debate here.
You cannot have that healthy 82°C operability with precision boost 2, or precision boost overdrive - they max out at 61.8°C is the problem.
It wasn't long until I found out manual clocking is the only setting that drops all temperature controls. That overrode the whole original statement - that this could be approximated by default settings if pttl was set at a high threshold without overriding pb2/pbo. This is a work in progress definitely, but I'm just here to make a head way, fortunately.

That's not right either though, a few have brought to market unconventional approaches to cooling see coolzone freezit for one, I nearly bought one , a HSF with a Tec built in and thermostatic control.

You see tec has a higher thermal transmittance, but you cannot limit yourself to pure copper. Heat transfer coefficient is best served with vapor chambers, comparatively diamond is the best 'solid' conductor 3x better than copper.

 

[TheoneandonlyMrK]

This is definitely the kind of health scepticism I appreciate, nothing good has ever come out of blind zealots.
You see, your intro;


'Casually' is the keynote here. Of course, an advanced user does NOT default to precision boost 2 & its bigger brother precision boost overdrive.
This is the merit of this thread, I'm sure you aren't having any difficulty hiding that fact from your intellect with much conscientious scrutiny.
It is the default settings of the general community and how these default settings are presented in bios is the point of debate here.
You cannot have that healthy 82°C operability with precision boost 2, or precision boost overdrive - they max out at 61.8°C is the problem.
It wasn't long until I found out manual clocking is the only setting that drops all temperature controls. That overrode the whole original statement - that this could be approximated by default settings if pttl was set at a high threshold without overriding pb2/pbo. This is a work in progress definitely, but I'm just here to make a head way, fortunately.


You see tec has a higher thermal transmittance, but you cannot limit yourself to pure copper. Heat transfer coefficient is best served with vapor chambers, comparatively diamond is the best 'solid' conductor 3x better than copper.

You can't best physics , you might think you can game the situation somewhat but your setting up a hypothetical nonsense in that your just priming a system to run in a particular way without any goal or point other than how it runs.
That's not going to end up being a viable general use pc on one of the other fronts your then forgetting about.
total cost of ownership.
Efficiency.
Ease of use.
Surrounding room air temp.

You should try living in the same room as two mining rigs, a few crunching rigs, a big gaming pc, in summer.

Heat is Always the enemy, but sometimes it's personal.

 

[mtcn77]

You can't best physics , you might think you can game the situation somewhat but your setting up a hypothetical nonsense in that your just priming a system to run in a particular way without any goal or point other than how it runs.
That's not going to end up being a viable general use pc on one of the other fronts your then forgetting about.
total cost of ownership.
Efficiency.
Ease of use.
Surrounding room air temp.

You should try living in the same room as two mining rigs, a few crunching rigs, a big gaming pc, in summer.

Heat is Always the enemy, but sometimes it's personal.

It is always welcome to see the gears are starting to turn. We are, at last, starting to convince people to speak up about mitigating the progressively incremental heat loads. That is what eventually evokes a lower temperature target down the line of thought. I have always anticipated for better air flow efficiency being contested in debate. The time has come! Let the overclocking diaries begin...

 

[Zach_01]

@mtcn77 there is no point for users to run their CPUs hot just to utilize the all of the cooling capacity of their coolers. Users, including me wants to keep their CPU and GPU and every other chip that matters as cool as possible. Its useless to dissipate and transfer high amounts of heat if I cant keep the CPU as cool as possible. I dont want to transfer high amounts of heat just for the bragging rights, by keeping CPU temp high!
There in no point, no benefit in CPU operation, function and longevity...

You want to keep temp gradient high? Try to further cool the IHS and/or the cooler it self . If ambient temp is a ristrictive point, accept it and move on, or pay for sub-ambient/sub-zero cooling solution, and keep your CPU cool and happy...

We are going circles...
Farewell!

 

[eidairaman1]

@mtcn77 there is no point for users to run their CPUs hot just to utilize the all of the cooling capacity of their coolers. Users, including me wants to keep their CPU and GPU and every other chip that matters as cool as possible. Its useless to dissipate and transfer high amounts of heat if I cant keep the CPU as cool as possible. I dont want to transfer high amounts of heat just for the bragging rights, by keeping CPU temp high!
There in no point, no benefit in CPU operation, function and longevity...

You want to keep temp gradient high? Try to further cool the IHS and/or the cooler it self . If ambient temp is a ristrictive point, accept it and move on, or pay for sub-ambient/sub-zero cooling solution, and keep your CPU cool and happy...

We are going circles...
Farewell!

Heat damages parts, end of story.

 

[mtcn77]

You should try living in the same room as two mining rigs, a few crunching rigs, a big gaming pc, in summer

You are only misconstrued in doubting the chip will run hot because we let off cooling too easily; the contrary is true, imo.
A better heatsink will make your room hotter likewise by extracting more effective heat. We are only emulating that, through a temperature gradient higher than what is preset. Heat just follows through.

Users, including me wants to keep their CPU and GPU and every other chip that matters as cool as possible.

Yes, we are forfeiting temperature transient load up interval voluntarily, however it is with the overriding condition of reaching the steady state faster. Best use practices versus the default, it is an argument of essentially what overclocking has come to in the ryzen 7nm generation, whether we like it or not.
I don't think you are damaging parts if you have enough ventilation in the case to let the cpu heatsink idle for a bit longer.

 

[JackCarver]

Interesting approach is what CaseKing make, together with der8auer and their King Mod Team. It's a shop in Berlin and they delid Intel or AMD CPUs, select the best and make a new IHS from silver. You get a guarantee that it runs at a given max clock speed, but you have to pay a lot more for this service. But silver has best heat coefficient

 

[EarthDog]

Interesting approach is what CaseKing make, together with der8auer and their King Mod Tea, It's a shop in Berlin and they delid Intel or AMD CPUs, select the best and make a new IHS from silver. You get a guarantee that it runs at a given max clock speed, but you have to pay a lot more for this service. But silver has best heat coefficient

That isn't anything new... It was done with an all copper IHS too. Funny part is that alone, it doesn't yield much over all copper or the nickel-plated IHS already on there.

 

[mtcn77]

Heat damages parts, end of story.

That is a generalization. We are only in it for usability, not misusing parts. We can do much better than stock, both cooling-wise and performance-wise. It is only rational to test it out. You are, after all, pacing at your own leisure. If you want to hit the nail on the head and facts are I'm talking up overclocking to non-overclockers, it suits me.

 

[EarthDog]

Is there a language barrier here? Some of these posts are so difficult to read.

 

[JackCarver]

Funny part is that alone, it doesn't yield much over all copper or the nickel-plated IHS already on there.

Yes but you simply get the best here, so if you have the money and are willing to pay, this service is really good and der8auer is one of the best overclockers out there. They also make conductonaut liquid metal on the die which is also better than Intels solder...

 

[mtcn77]

Is there a language barrier here? Some of these posts are so difficult to read.

Both language and physics training. I cannot break down big subjects other than make up stuff for you guys' inquiry.

 

[EarthDog]

Both language and physics training. I cannot break down big subjects other than make up stuff for you guys' inquiry.

Gotcha, so ESL. It isnt the physics, lol.

Yes but you simply get the best here, so if you have the money and are willing to pay, this service is really good and der8auer is one of the best overclockers out there. They also make conductonaut liquid metal on the die which is also better than Intels solder...

Again, enjoy that 100 mhz difference if you're lucky! This isnt for the average enthusiast, even...

 

[infrared]

I'm talking up overclocking to non-overclockers

No lol, almost all of us in this thread are hardware enthusiasts and very much into overclocking. Because of that we all understand the benefits of running cooler, which is why we are all confused as to why you think higher temps are better..

 

[mtcn77]

No lol, almost all of us in this thread are hardware enthusiasts and very much into overclocking. Because of that we all understand the benefits of running cooler, which is why we are all confused as to why you think higher temps are better..

Well, first of all heat transmittance and second of all 7nm. I really understand it, point taken, that you are a very light-handed person - I don't think this goes against your perspective. Obviously, you don't do this with a closed aio waterloop.
Waterloops aren't for the general userbase. It sounds like I'm going off topic not reserving myself to argumentation, but I like the exercise.

Again, enjoy that 100 mhz difference if you're lucky! This isnt for the average enthusiast, even...

You might be surprised about the effect of your scepticism on my responses. Nothing fans my flames like a good pinching penny challenge.

 

[Zach_01]

@mtcn77 what is your current CPU?

 

[EarthDog]

You might be surprised about the effect of your scepticism on my responses. Nothing fans my flames like a good pinching penny challenge.

ok...

Here is a custom copper ihs - https://www.gamersnexus.net/guides/3238-custom-copper-ihs-tested-on-intel-i7-8700k-cpu-rockit-cool

Don’t expect great things out of the Rockit Cool IHS, but it does decently. Certainly well enough to be worth buying for someone who already has a thermal objective in mind: Maybe you want to reduce fan speed by 100-300RPM (thereby reducing noise), or maybe you’ve got some numerical OCD about thermal numbers, or you just think it’d be fun to play with computer parts for an hour. These are all good reasons to make the purchase. Pushing for higher overclocks or exceptionally lower temperatures would not be reasons to buy the IHS; if that’s what you’re after, you’d be better off buying a higher-end cooling system. Any bit helps when it comes to thermals, but other components can help more.

We liked the product for its affordability, machining quality, and value as a time-passer. Again, this isn’t something that’ll get you more frequency or lower voltage (meaningfully), but it is something that’s kind of fun to work with.

Der8auer Replaces i7 8700K Stock IHS With Silver IHS for Caseking Ultra Edition – Gamers Navy

A veteran overclocker from Germany known as Roman "Der8auer" Hartung is onto something special with the Intel's latest 8th generation i7 processors. Der8auer i7 8700K Silver IHS Utra Edition: Der8auer is a name that comes across to those interested in overclocking scene many times. And just...

www.pgrepublic.com