Unknown Temps huge spike

[WCPredator]

SOLVED: VRM Spread Spectrum had been disabled for some reason in ALL my OC profiles.. now those mystery temps are ice cold (<50c lol). So the 129c was real !!!!!!!

I've been using the same profiles for a few years now and have always dismissed some egg cooking temps as errors. Today I was using optimized BIOS defualts to check something and I noticed those "error" temps were in the normal range. Strange thing is that my normal profile settings are just a Vcore undervolt with leds off..
Optimized default (Concerning Temperatures are #3,#4 and $6)

Undervolt

[WCPredator]

MoKiChU wrote:
Hi WCPredator,

Can you try to disable the ASUS MultiCore Enhancement option in the BIOS then look at the temperatures ? This option puts the 4790k to a hard test ...

Um sure? Just disabling the option pushes the 4790k or stress testing without it?

[MoKiChU]

WCPredator wrote:
Um sure? Just disabling the option pushes the 4790k or stress testing without it?

Yes, can you stress testing without ASUS MultiCore Enhancement option

[Nate152]

To see the current bios version bring up "System Information"

Then:

[WCPredator]

Ah thanks man, but i did it the stupid guy way where i just started the install program for the driver i used and looked before actual install started lol, it was the original! New one makes no difference anyway, however, I have a feeling that increase is from anti-surge being enabled by me whereas it's disabled in defaults for some reason. Check in a bit, fedex just dropped off the card - which covers most of that plastic part, the chipset i guess...

Edit** that's the bios version not chipset 😛

[Nate152]

My bad

To see the chipset driver

Open Device Manager
Click system devices
For me - right click Intel(R) SMBUS - A2A3 and click properties
Click the driver tab
Driver version is the chipset driver

[WCPredator]

ah cool thanks. Btw I was wrong, enabling anti-surge didn't replicate the high temperatures.. all still under 80c. I guess when it doesn't reduce the voltage ever it somethings get really hot, I'll try to see tmrw after a few hours gaming if those go back towards or above 100

[Nate152]

yeah pay attention to:

core 0
core 1
core 2
core 3

These are the temps you want to watch in HWinfo.

Your cpu is throttling the voltage at idle so temps should be good.

[WCPredator]

That's only with "optimal defaults" loaded, with undervolt and even more with raised voltage overclock something seems to be tryin to make eggs in there lol. I'm thinking I can try voltage offsets leaving adaptive on, but I shouldn't have to do that - especially with an overclocking board and a processor that can do 4.8 with safe voltages (1.32v? i think, i usually go 1.24v 4.7 cuz there's minimal difference)

[Nate152]

HWinfo doesn't show anything running abnormally warm, maybe it's the 7970.

[WCPredator]

There was no GPU before and the numbers were the same. I just put an r9 390 in. I don't know what I was thinking with adaptive, there's nothing like that in the BIOS lol, been a long time since I was OCing in there. I can change the multiplier to 4.7, but then it's pulling 1.36v... sigh, I'm thinking changing the freq. and vcore and leaving the rest on auto might be the problem, perhaps if I take one off auto I must take all of auto, only I don't know what else I'm supposed to mess with in this myriad of options.

EDIT: I'm not sure if this is funny or just plain mean: At 4.7 auto voltage (1.36v) those previous high temperatures are very cool, even cooler than optimized defaults! If only the CPU weren't practically on fire - as in raging past 80c with the nh-d15 screaming to keep it low 80s.. (stress test temps within 2 seconds lol) edit2: 77c after 2 minutes of BF4 with the cooler louder than i've EVER heard it, louder than the leaf blower gfx card wow..