Zen to be soldered

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

DrMrLordX

Lifer
Apr 27, 2000
21,570
10,762
136
Good. Do you think that if not soldered it would have overclocked a lot less? And Zen will be soldered?

I have no Bristol Ridge or Carrizo hands-on experience, so I will not comment on them.

KA-V1 Kaveri (A10-7700k) w/ stock TIM under IHS: my max OC was 4.5 GHz
KA-V1 Kaveri (A10-7700k) w/ delid/relid + lap job on IHS and CLU under IHS: my max OC was 4.7 GHz
GV-A1 Godavari (A10-7870k) w/ stock solder under IHS: my max OC is 4.7 GHz (haven't really tried for 4.8 but the voltages are getting too high for my tastes)

Note that all of the above overclocks used the same IHS (NH-D14 + really loud fans) and same TIM between IHS and HSF (CLU).

So in the case of Kaveri/Godavari, I estimate that delidding was about the same as having solder under the IHS, and that it was good for +200 MHz OC. Godavari seems to clock better at the same voltage than Kaveri at speeds up to about 4.2 GHz, after which point it's a wash. At least with my samples anyway.
 

beginner99

Diamond Member
Jun 2, 2009
5,208
1,580
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/

Which is about liquid nitrogen cooling. He talks about cracks in the solder after 200-300 thermal cylces where 1 cycle is from -50 to 125 °C.

Maybe someone with physics degree can tell us how much bigger this stresses the interface compared to an average overclocker that cycles from around 20 to 100° C. Also the stress probably has a breaking point and if you don't reach that no damage will be done. So no, this link while interesting isn't relevant at all.

The issue is that a i5-6400 and a i7-6700k are the same CPU. I would assume they are already fully packaged before they get binned and hence either all quad-cores need to be soldered or none of them. Another issue could be the iGPU which often is disabled and hence colder. That might give even more stress on the die if you solder it because everything expands unevenly now. HEDT is CPU only and workload usually is distributed evenly over cores so they have about them same temperature.
 

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,785
136
Which is about liquid nitrogen cooling. He talks about cracks in the solder after 200-300 thermal cylces where 1 cycle is from -50 to 125 °C.

They were also citing other concerns. Like being environmentally damaging. Also being a conflict material.

The volume on PC is immeasurably higher than on big die chips.

Thathe makes me question why they couldn't improve on applying TIM so it runs cooler. People say it's the application that's at fault rather than material. The killing of R&D teams may contribute to that? Lack of soldering won't make people angry if the thermal issues are solved. The company probably has much bigger issue to tackle internally than this, meaning the attention goes there instead.
 

superstition

Platinum Member
Feb 2, 2008
2,219
221
101
They were also citing other concerns. Like being environmentally damaging. Also being a conflict material.
Environmentally damaging? Like the fluorocarbon used to etch silicon? The vast amounts of e-waste from planned obsolescence business plans? Companies use up the resources then switch to an alternative, as is happening with touch screens. That's how our global business operates.

Indium is hardly that pressing of a concern. Besides, the smart thing to advocate is TIM for low-end chips and indium solder for chips like the 6700K.
Maybe someone with physics degree can tell us how much bigger this stresses the interface compared to an average overclocker that cycles from around 20 to 100° C. Also the stress probably has a breaking point and if you don't reach that no damage will be done. So no, this link while interesting isn't relevant at all.
The "truth" about solder material is designed to make Intel's practice seem more palatable. It also makes the delidding industry happy. Save the whales and Intel's profit margin. The stress crack concern is only for liquid nitrogen cooling.

Meanwhile, I don't see anyone delidding clamoring to get manufacturers of liquid metal TIM to reveal how much indium is used in their formulations.