lolfail9001

Golden Member
Sep 9, 2016
1,056
353
96
Considering that the very article contains demonstration of opposite being true (namely AM4 cpu in Bristol Ridge not being soldered), it is kind of hilarious.
 

superstition

Platinum Member
Feb 2, 2008
2,219
221
101
I think it's highly unlikely that AMD will go with polymer TIM for Summit Ridge, don't you?

The worst thing they could do is challenge Intel's superior process node by impeding thermal transfer with cruddy TIM. Intel can afford that luxury for segmentation but AMD can't — at least not for Summit Ridge.

No one cares about Bristol, do they?
 

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
Considering that the very article contains demonstration of opposite being true (namely AM4 cpu in Bristol Ridge not being soldered), it is kind of hilarious.

Just like back in the 775 era, when we had the high end C2Ds that were soldered, and the Celerons and Pentiums that used TIM all sharing the same socket. Try delidding an E2xxx CPU, and an E6xxx (avoid Allendale parts, that is, Conroe-2M/L2 stepping, these aren't soldered) or E8xxx CPU. E2xxx uses TIM, Conroe-4M parts in E6xxx line are soldered, E8xxx parts are soldered. E7xxx cut down Wolfdales are TIM'd.

It's not hilarious nor a demonstration of the opposite. Would you waste money on soldering low wattage parts like BR? Zen in its Summit Ridge form isn't low wattage (95w TDP), and I find it not surprising that it's soldered. That's what should be done. That's gonna help over 4GHz.

Intel on the other hand cheaped out from Ivy onwards on high wattage CPUs using TIM for unlocked parts, leaving solder just for HEDT. A brain dead move IMO that hinders the very reason of existence for unlocked parts: overclocking and the need for a quick and effortless path for the heat to go to the heatsink.. solder.
 
Last edited:

bjt2

Senior member
Sep 11, 2016
784
180
86
Even if Bristol Ridge is not soldered, it overclcoks at 4.9GHz on air (the 65W part), with <1.4V. Not bad...
 

lolfail9001

Golden Member
Sep 9, 2016
1,056
353
96
Just like back in the 775 era, when we had the high end C2Ds that were soldered, and the Celerons and Pentiums that used TIM all sharing the same socket. Try delidding an E2xxx CPU, and an E6xxx (avoid Allendale parts, that is, Conroe-2M/L2 stepping, these aren't soldered) or E8xxx CPU.
You know what, i may just verify that, i do have Wolfdale Core 2 Duo and Core 2 Pentium, after all.
Even if Bristol Ridge is not soldered, it overclcoks at 4.9GHz on air (the 65W part), with <1.4V. Not bad...
Actually, it requires 1.431V and we have no clue if it is stable at all.
Because Bristol Ridge is suddenly Zen?
Because AM4 cpu suddenly means: "CPU that won't be available for 2 more months at least"?
AMD mention that in their presentation that it will be solder for Ryzen
Well, it basically convinces me that Summit Ridge is larger than 200mm^2. With all the joke a certain folks like to make about Intel's process not being dense at all, comparing it to GPUs for pete's sake, it sure is ironic that Zen looks to be even less dense.
 

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
You know what, i may just verify that, i do have Wolfdale Core 2 Duo and Core 2 Pentium, after all.

The Pentium will have TIM, the C2D will be soldered. Clarifying my post, E2xxx will be TIM'd, E6xxx apart from L2 stepping (Conroe-2M) parts will be soldered, E8xxx will be soldered. No need to do your own CPUs, google around.


Here's a quick list of what used to be soldered and didn't back in the day. Lots of pictures in this thread.

Delidded E8400, soldered
Delidded E2140, TIM


LOL, Celeron D P4 was soldered, those space heaters had to remain cool somehow.

Skt775 had it all.
 

bjt2

Senior member
Sep 11, 2016
784
180
86
Actually, it requires 1.431V and we have no clue if it is stable at all.

You are right, my bad. But i found a cpuz validation... Should this mean that is stable somehow? Is it mandatory to perform the CPUZ bench to validate a screen?
 

EXCellR8

Diamond Member
Sep 1, 2010
3,979
839
136
they should just offer delidded versions at a reduced price and no warranty... I like to live dangerously.
 
  • Like
Reactions: Doom2pro

BeepBeep2

Member
Dec 14, 2016
86
44
61
Considering that the very article contains demonstration of opposite being true (namely AM4 cpu in Bristol Ridge not being soldered), it is kind of hilarious.
1. This is old news
2. Regardless of what you think you see with your own eyes, that pictured CPU is soldered

Also @ OP: This also isn't bad news "for the liquid nitrogen people", if the chip has a coldbug like a lot of Intel's recent chips it doesn't matter quite as much and soldered IHS performs great at full cold.
 
Last edited:

BeepBeep2

Member
Dec 14, 2016
86
44
61
Indium based solder, the news stories and images originate from overclocker Namegt's facebook page. He never claimed what APU/CPU it was, only that even though it looked fine originally to his naked eye, he mentioned that sometimes you can "break the CPU with high heat".



uc


uc
 

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
I can tell there is a lot of confusion about soldered IHS and such... So I'll just leave this here.

http://overclocking.guide/the-truth-about-cpu-soldering/

Nice writeup.

Still,

Conclusion
Whenever I read sentences like “What a ripoff – Intel doesn’t even solder a 300 USD CPU” or “Why does intel save 2 USD on soldering” I’m thinking



Stop hating on Intel. Intel has some of the best engineers in the world when it comes to metallurgy. They know exactly what they are doing and the reason for conventional thermal paste in recent desktop CPUs is not as simple as it seems.

Micro cracks in solder preforms can damage the CPU permanently after a certain amount of thermal cycles and time. Conventional thermal paste doesn’t perform as good as the solder preform but it should have a longer durability – especially for small size DIE CPUs.

Thinking about the ecology it makes sense to use conventional thermal paste. Gold and indium are rare and expensive materials. Mining of these materials is complex and in addition it’s polluting.

After soldering one of my 6700K CPUs I can tell it’s a pretty complex process. I’m still working on it and trying to make it available for extreme overclockers. However, I doubt that Intel will come back with soldered “small DIE CPUs”. Skylake works great even with normal thermal paste so I see no reason why Intel should/would change anything here.

Why then does Intel solder HEDT of Ivy-Broadwell-E generation processors? Why HEDT and not mainstream 115x anymore? Margins? I can't think of other reasons. HEDT uses bigger dies as mentioned in there, maybe mechanical stress is mitigated on these due to the increased area vs the relatively tiny mainstream dies.

Oh well. We'll keep delidding + CLU to push the chips to their limits at decent temperatures.

edit: we're always talking about K parts on 115x, locked parts could use ketchup as TIM for all I care, they have no need for high performance stuff.
 
Last edited:

Doom2pro

Senior member
Apr 2, 2016
587
619
106
Nice writeup.

Still,



Why then does Intel solder HEDT of Ivy-Broadwell-E generation processors? Why HEDT and not mainstream 115x anymore? Margins? I can't think of other reasons. HEDT uses bigger dies as mentioned in there, maybe mechanical stress is mitigated on these due to the increased area vs the relatively tiny mainstream dies.

Oh well. We'll keep delidding + CLU to push the chips to their limits at decent temperatures.

There are positives and negatives to soldered IHS and Thermal Compound.... As the link I provided shows, even soldered IHS can have voids over time... Nothing is perfect and if you are trying to market a low to mid range CPU, f-it... use Thermal Paste.
 

DrMrLordX

Lifer
Apr 27, 2000
21,570
10,763
136
Considering that the very article contains demonstration of opposite being true (namely AM4 cpu in Bristol Ridge not being soldered), it is kind of hilarious.

Um, what?

Even if Bristol Ridge is not soldered, it overclcoks at 4.9GHz on air (the 65W part), with <1.4V. Not bad...

But it is soldered . . . so was Carrizo. So was Godavari.

2. Regardless of what you think you see with your own eyes, that pictured CPU is soldered

ding ding ding we have winnar
 

superstition

Platinum Member
Feb 2, 2008
2,219
221
101
I can tell there is a lot of confusion about soldered IHS and such... So I'll just leave this here.

http://overclocking.guide/the-truth-about-cpu-soldering/
The Truth About CPU Soldering Strictly In Terms of Liquid Nitrogen Cooling

The bigger point the article does contain:
article said:
The thermal conductivity is not as high as copper but higher than any other thermal interface material 81.8 W/(m*K). Common thermal compounds have a conductivity of about 5-10 W/(m*K).
Note that Arctic Silver 5 was rated at less than 1 by a government test. Except for liquid metal TIM that 5-10 seems optimistic.
Wiki said:
The company claims that AS-5's thermal conductivity is 8.7 W/(m·K). However, a study led by the US National Renewable Energy Laboratory found that it was only 0.94 W/(m·K).

As for all the worry about Indium consumption, I wonder how much is being used in the liquid metal TIM people are using when delidding.

People who are happy to subject their CPUs to conditions like these:
article said:
A thermal cycle is performed by going from -55 °C to 125 °C while each temperature is hold for 15 minutes. Micro cracks occur after about 200 to 300 thermal cycles.
should probably be comfortable delidding a soldered CPU.

I do agree that low-end CPUs (e.g. Anniversary Pentium and i3) probably are just fine with polymer TIM. Saves on Indium. CPUs like a 6700K aren't.
 
Last edited:

Abwx

Lifer
Apr 2, 2011
10,842
3,295
136
Good. Do you think that if not soldered it would have overclocked a lot less? And Zen will be soldered?

Dunno but Godavari temp at equal power is noticeably less than Kaveri s wich is not soldered.

BVjcLtr.png
 

bjt2

Senior member
Sep 11, 2016
784
180
86
Dunno but Godavari temp at equal power is noticeably less than Kaveri s wich is not soldered.

BVjcLtr.png

Cool... :D (pun intended)

Lower temp mean lower leakage and higher transconductance, hence yes, also higher OC... But from temperature the higher OC margin is almost impossible to derive...