Trident Z F4-3200C14D-32GTZ on a ASUS Z170-Deluxe: Problems and Questions

[emissary42]

Just a quick question: Wouldn't it imply a fault either in the CPU, RAM or mobo, if increasing the switching frequency actually fixed this problem?

It would hint at the MB, but even that depends on how you look at it. If raising the switching frequency would in fact solve your issue, an argument could be made for a voltage regulation design-, implementation- or component weakness on the MB side. Although to actually confirm that, you would at least have to do a comparison test with your MB and another CPU and memory - or test your CPU and memory on another MB.

And should it even be necessary at stock settings at all?

No, it usually is not necessary. However please also consider that priming for days straight is not something lots of people do.

[BinaryKhaos]

Hi...

It would hint at the MB, but even that depends on how you look at it. If raising the switching frequency would in fact solve your issue, an argument could be made for a voltage regulation design-, implementation- or component weakness on the MB side. Although to actually confirm that, you would at least have to do a comparison test with your MB and another CPU and memory - or test your CPU and memory on another MB.

Ok. I decided to deal with this whole situation very methodically. I have just ordered the same RAM as a replacement. With any luck, it was really just a temperamental kit I got and the new one works fine... and I can finally put this whole mess to rest.

And if this does not fix the problem, I will swap CPU and board together which should then hopefully and finally fix this. If not, then I am out of ideas...

No, it usually is not necessary. However please also consider that priming for days straight is not something lots of people do.

Point well taken. I usually don't do that as well but right now it is the only thing I can do to actually confirm or disprove stability. It is neither convenient nor something I like or usually do.

Thanks,
Matthias

[BinaryKhaos]

Hello @all,

here the conclusion of this whole messy situation: It was Skylake errata SKL082 that Prime95 happily triggered both with FMA3 and AVX. It was supposed to be fixed with MC 0x74 but apparently Intel had to work some more on it as I was still able to run into this. Once I upped to MC 0x9e manually (there is currently no ASUS firmware with 0x9e), I was finally able to stability test the system for 75h straight with Prime95 on stock settings.

I have now upped to DDR4-3000 and started my last Prime95 test run with that. I don't expect any trouble and it has been running fine for almost 38h now. So once 72h are reached, I call it a day.

Six or seven weeks of this madness, I should send the power bill to Santa Clara, CA I guess. I am glad that I have, apparently, finally found the culprit and could forgo swapping CPU and board after all -- even though I really would be curious to know how many years I shoved off the life of my CPU after the torture I put him through... and vice-versa, actually. ;-)

Just for those curious, you can find my thread on the mersenneforum here.

Thanks for all the input and help!

Have a nice weekend,
Matthias