GPU issue with CPU Overclocked?

[Kratos]

Hi, first of all I'm new on this forum, so good morning everyone

 

So, I'm currently overclocking my system with very decent results, but it is my first time doing this, so, yeah, murphys law.

I tried to have my i7 4790K to 4.8 GHz and it was indeed running very good with Only-CPU softwares, the cpu and cpu single core on cinebench where fine, reching maximum core temps of 76 degrees, the average maximum of all 4 cores was about 73. The thing is, when I run the "Open GL" which is intensive in both CPU and GPU, nearly at the end I get a yellow screen, cinebench closes and nvidia's controll software resets. I've no idea to why it does this and I haven't tried games yet with this profile, I just don't feel safe to do so. 

 

Here some usefull information:

 

-I've got my graphics card overclocked to +150 MHz core and +400 memory clocks speeds both on MSI afterburner. I tried and lower down to +200 Mhz the memory clock speed and it still caused yellow screen. Temps on the GPU while gaming (with other config) rise no more than 65 degrees. TDP is at 112% and voltaje is untouched due to the fact thar MSI afterburner doesn't let me change it, which I think it could be the problem. With my other overclocked profile everything worked fine, I even tried a couple of games for 3 hours and everything OK..

 

-I use my system for gaming + video encoding/editing/rendering purposes, so what you guys think? Here's my system overclocked which works fine and my specs if it is of any help:

 

-LGA 1150

 

-CPU: i7 4790K @ 4.7 GHz 1.180v (1.180v is stable for MY personal use, not prime95, my personal use also is CPU intensive, so for 2 weeks it's been stable making use of 6-8 hours/day)

-MOBO: Gigabyte Z97X-Gaming 3

-CPU COOLER: Noctua NH-D14

-RAM: 4x8 (32GB) Kingston DDR3 HyperX Fury Blue 1600 MHz CL10 @ 1866 MHz 10-11-11-32 1.520V (This gets hot sometimes touching it but it's Ok I think)

-GRAPHICS: Gigabyte GTX 970 G1 Gaming @ +150 MHz core clock, 112% TDP, +400 MHz Memory Clock

-CASE: Corsair 760T 

-STORAGE: ©: Samsung SSD 840 EVO 500 GB, (D): Toshiba 3TB, (E): Segate 1TB

-PSU: Tacens Radix IV 800W 80 Plus Silver

-MONITOR: Just a Samsung 22" FullHD Smart TV, when gaming I use an LG 3D 32" Smart TV FullHD 400 Hz, everything is through HDMI

 

Thanks a lot to at least take the time to read this long post

[Jeburs]

I had a similar issue. I had a old psu, replaced it and found I had a more stable overclock. Use hwinfo to watch the voltage on your 12v rail to see if it drops at all during the stress test

[Jeburs]

Also a word of warning, you should avoid touching your ram if you arnt grounded, can fry it if you have any static on you

[Kratos]

Thanks, I usually touch a metal object before touching RAM in order to remove any static, but I'll try not to touch it at all. I'll use the software you say and keep an eye on the voltaje, seems like the best thing to do.

By the way, what is a safe 12V? On the motherboard APP under gaming with my current profile I get :

12.30v

5.16v

3.36v

 

Those are Ok I think because so far it hasn't caused any problems.

So far from what you told me and what I think (correct me if I'm wrong) the problem could be one of the following:

-PSU Volts

-GPU Volt (TDP is under 70% during gaming)

 

It could be like you said the PSU, but it is recent and it's a 80 Plus silver 800W, shouldn't it be ready for this kinds of things? It was way expensive (60€ which is about 70-75$)

It would be a kick on the balls (sorry) to change PSU again, (I changed before), as I also replaced my cpu cooler Bequiet! Darkrock 3 for the Noctua NH-D14