German Company to Sell Binned Core i7 8700K With 99.9% Silver Heatspreader

[EarthDog]

There's like 5 links and the direct article doesn't say anything. No time to search.

The way to test it is to delid, apply new paste under iHS, relid. Rinse and repeat with other IHS, compare.

Silver doubles the amount of heat it can transfer over aluminum. Again, not sure what that really translates to, but...
http://hyperphysics.phy-astr.gsu.edu/hbase/Tables/thrcn.html

 

[Tír na nÓg]

Just below the video. 3°c for the best result. As I said above no IHS can do better than direct contact.

Edit: bud, I won't make you go through the 5 links! on the main page

 

[EarthDog]

Nobody would (or did) deny the fact that naked is better. I'm asking what is the difference between the default IHS and the silver one...

You said:

Well, silver IHS is more a marketing gimmick than anything else IMO.

Me:

Is it? What is the thermal rating of the stock aluminum IHS? Its notably less. Exactly how much and what that translates to temperature wise, who knows.

This has nothing to do with naked die temps. Also, considering that mountning isn't exactly a science on naked processors, you can't really take that info anywhere. It needs to be tested as I said to get a true reading. I see what you are trying to get at, but, I dont feel it can be tested that way.

 

[Tír na nÓg]

Nobody would (or did) deny the fact that naked is better.

I'm asking what is the difference between the default IHS and the silver one...

You said:
Me:


This has nothing to do with naked die temps.

OK, let's slpit hair

Delta between naked die and (made of copper, not aluminium... Check when lapping) IHS is 3°C.

So delta between Silver and copper IHS is between 3° and 0°C, as it can't be more than naked, and can't be less than copper, as silver has better conductivity. So, that's €450 for (realisticaly by extrapolating) 1 to 2c better temp.

Edit: do they say the IHS is made of aluminim? If so, on top of being crooks, they are liars!

Edit 2: sure it is not exact science, but I think those guys are trustworthy (love the word hihihi). Never noticed any biaised or uninformed article on their site.

 

[EarthDog]

No, i said it... its copper with nickel plating or something. My fault.

Ive seen ihs to naked drop temps more than 3c....

 

[Tír na nÓg]

No, i said it... its copper with nickel plating or something. My fault.

Ive seen ihs to naked drop temps more than 3c....

No worries mate, it happens to the best of us

Sorry, it is the only reliable source I've found.

IHS to naked die, after having replaced the thermal paste with iiquid metal? From what I've read in various places, naked die doesn't bring much of an improvement over ihs+liquid metal.

Could not find the sources anymùore, as it is information I thoroughly checked before de-lidding my first 3770k 4 years ago, and did not archive it...

 

[EarthDog]

Yeah, not saying they are not trustwprthy... just that weve seen better temp differences than that. Im wondering if he used a naked mount kit on the board. Anyhooo....

 

[Tír na nÓg]

Yeah, not saying they are not trustwprthy... just that weve seen better temp differences than that. Im wondering if he used a naked mount kit on the board. Anyhooo....

Well, would need to cjeck the video. Instead, I am going to play some Roboooooooo RECALL!

 

[Kenrou]

Wouldn’t temps be a lot more then 3c when using more then a 4c/8t, I mean exponentially higher with core count and voltage ? Wouldn't that make it worthwhile ?

 

[Nebulous]

I still think it's unrealistic. I'll bet most, if not all, chips can get to 5.0Ghz with a robust board and really good watercooling. Lets say I pay the xtra $$ for their so-called "binned" chip that is supposedly guaranteed for 5.2Ghz. I have a high end board, a very capable psu and a hefty watercooling setup. I put it all together and when I hit the pwr switch I'm looking to get that 5.2Ghz that I paid for. Several reboots later the best I can get is 5.1Ghz and that's it. You honestly think Imma be happy?

Then there's the RMA dance. More money to send the "defective" chip back, wait again for another "guaranteed binned chip". Who knows maybe the 2nd chip is the magical one, or the 3rd. To me that's not worth the xtra price of admission just to get a free bag of popcorn. But hey that's just me.

 

[Tír na nÓg]

Wouldn’t temps be a lot more then 3c when using more then a 4c/8t, I mean exponentially higher with core count and voltage ? Wouldn't that make it worthwhile ?

Copper 386 vs Silver 407 (W/m K).

The difference will be marginal at best (Aluminium is 204 W/m K).

We are talking a couple of degrees at best, marketing gimmick.

 

[Alaric]

Gold?

 

[Mr.Scott]

 

[Alaric]

BWAHAHAHAHAHAHA! And that is why we're friends.

 

[RGE]

I thought this was going to be a silicon lottery business model clone....then I read the silver IHS....and the price. Maybe if the silver IHS was visible and it went with ones color scheme, but IHS is under cooler, so dont really get the point. Id rather have the copper ihs.

The gradient across the copper IHS is less than 2C at 150W, measured several different ways. If copper IHS has gradient of 2C, silver IHS would have 1.9C gradient, not going to be able to see the 0.1C difference between copper ihs and silver one. Will still get 10-20+C drop in temps from replacing intels thickly applied 3.5 w/mk polymer paste with thinner bondline of 40-80 w/mk liquid metal, but silver IHS ???

On xtreme back trying to do testing for realtemp prior to tj being readable (im rge there) had drilled holes/slits in IHS with calibrated thermocouples, including delidded ones. Even with thermocouple 80-90% deep in IHS, could never measure more than 1.5C between embedded and surface. Also bare die with liquid metal versus IHS with LM for tim1/tim2, there is 3-4C difference in temps, and the majority of resistance per intel white papers (and common sense) is across tim interfaces, which leaves 2C or less for IHS gradient.

 

[Tír na nÓg]

^THanks a lot for the in depth explanation!

 

[Alaric]

And there will be some people who will want it because it is better. Doesn't have to be a lot, just has to be the best. After all, second place is the first loser.

 

[Nebulous]

And there will be some people who will want it because it is better. Doesn't have to be a lot, just has to be the best. After all, second place is the first loser.