Print Page | Close Window

Poor OC / PSU Problem

Printed From: ASRock.com
Category: Technical Support
Forum Name: Intel Motherboards
Forum Description: Question about ASRock Intel Motherboards
URL: https://forum.asrock.com/forum_posts.asp?TID=4068
Printed Date: 22 Jul 2025 at 5:07am
Software Version: Web Wiz Forums 12.04 - http://www.webwizforums.com


Topic: Poor OC / PSU Problem
Posted By: Ky0sHiR0
Subject: Poor OC / PSU Problem
Date Posted: 20 Dec 2016 at 7:24pm
" rel="nofollow - Hello,
In October I bought computer with following specification (I built it)
i5 6600K
CPU Cooler Fortis 3
GTX 1070 Palit Super Jet Stream
2800 MHz DDR4 Corsair Vengeance LPX
PSU Corsair CX450M

The first case - Poor OC
Im not able to get higher OC than 4,4 GHz @ 1,344 V its really really poor.
What is more CPU is delidded and temps are low.
Someone suggest me that my PSU isn't meet requirements for GTX 1070 and is a low tier PSU.
In that case I bought FSP/Forton Hydro-G 750W from Tier One (PSU Tier list 2.0 etc.)
and then... we are coming with second case - PSU Problem
I mounted PSU and everything have gone right but...
Everything is connected to my power strip - computer PSU, monitor, speakers etc.
If I am out of home I turn off power strip and in that case :

1) When You are trying to turn on computer I just turn on power strip
2) Push ON button on my computer
3) PC starter for 2-3 seconds (fans test etc) and then power goes off
4) After 1 sec PC computer will turn ON and boot normally

There was no problem with previous PSU (Corsair CX450M)
I contacted AsRock and get following response:

Dear Patrick,

 

This is not caused by the board.

When you press the start button, you start the PSU, NOT the board, this is how it works:

1.       The start button short circuits 2 pins in the big ATX power connector, sending the PSU a signal to start

2.       The PSU runs a selftest

3.       When the selftest is okay, and the system is stable and within specs, the PSU send a "Power Good" signal to the board

4.       Then the board starts

 

I know for a fact this is PSU related, I had a Enermax PSU, and had this issue, when I replaced it for a XFX PSU, the problem was gone.

I understand that it is not nice when you change the PSU, the system waits 2 seconds before it starts, but this cannot be solved by the board, or BIOS.

 

I have spoken to several manufacturers, but no one can really tell me why this is, but the fact that one PSU, works fine, and another gives this delay, makes this clearly a PSU issue.

Manufacturers do not make the PSU's themselves, there are only a few manufacturers of PSU, all the others buy from these, and just ask them for specific options and specifications, and of course by their own design.


also created topic on reddit:
https://www.reddit.com/r/overclocking/comments/5i9xfh/new_psu_and_weird_selftest/

important message:
I have the same motherboard and the same issue when unplugging the PC for a longer time period. My PSU is the EVGA 650GQ so that shouldn't be the problem.


With all that knowledge I think that problem is MOBO related not PSU related.
Can anyone help me ?

Thank You for reading :)



Replies:
Posted By: Xaltar
Date Posted: 20 Dec 2016 at 7:58pm
What is happening is the system is running additional post tests after "power loss" is detected. This is a standard safety feature with most motherboards today (of all brands). Happening with one psu and not another could be caused by any number of reasons. The original PSU may hold a small charge for longer than the new one thus preventing the "power loss" flag or the older PSU discharges it's capacitors in a more agreeable way that the motherboard does not flag etc. I can reproduce the "issue" on 2 of my own systems by doing as you say, switch off at the wall. This happens on my Zalman 700w PSU as well as my Cooler Master 500w unit but does not occur with my OCZ 1000w unit. It varies from one PSU model to another I guess.

As for the overclocking, I don't see a board listed in your specs so I can't offer any tips/suggestions.


-------------


Posted By: Xaltar
Date Posted: 20 Dec 2016 at 8:15pm
Just as an aside, it is not at all healthy for your system to continually power off and on at the wall. It creates a power surge every time you power off and on that will over time damage your PSU and possibly even other components. 

It is also likely that the more expensive, higher tier PSU cuts power more abruptly to mitigate the power off surge which does not allow the board to enter a proper power off state before power is cut completely. Even higher end PSUs have additional capacitors and protections that again mitigate this effect but without causing the power loss flag.  

The system barely draws a few watts of power when in it's usual powered down state. It is recommend to not switch off at the wall unless you plan to be away for a while (like a month holiday for example). 


-------------


Posted By: Ky0sHiR0
Date Posted: 20 Dec 2016 at 9:25pm
" rel="nofollow - Hello,
MOBO model: AsRock Extreme 4 Z170
I really dont know that PC should be constantly on power. That's really weird for me. I was always turning off power strip. What do You recommend in that situation ? I will check in home all cables. Should I clear CMOS too ?


Posted By: Ky0sHiR0
Date Posted: 20 Dec 2016 at 9:35pm
I can return current PSU. I should decide before 25 December


Posted By: wardog
Date Posted: 20 Dec 2016 at 9:45pm
" rel="nofollow -
Originally posted by Ky0sHiR0 Ky0sHiR0 wrote:

Everything is connected to my power strip - computer PSU, monitor, speakers etc.

That line above grabs my attention.

Let's for a moment not consider it a Power Supply issue but instead one of simply proper power.

What does your term "etc' consist of connected wise?

IMO you have too much connected to the power strip now, w/o knowing what that "etc' includes. If that includes a laser printer, all bets are off the table.

To test, pull the strip and all connected out of the wall. Now, with just the computer and essentials to operate it, plugged directly into the wall outlet(s) sans all that other hardware,, does the system behave the same or better?




Now let's move on to discuss the damaging inrush current that you are unnecessarily subjecting all the items connected to the power strip. An excellent article I post here every time I read of people cutting power at the strip instead of leaving their newer highly efficient computers in a powered down state.

http://sound.whsites.net/articles/inrush.htm

Leaving capacitors charged in a powered down state is far far better on parts/components than justifying ANY energy saved by doing what you currently do, flip the strips switch and cut all power.

Possible short term and definitely long term damage happens each and every time you flip the strips switch to On.





Posted By: wardog
Date Posted: 20 Dec 2016 at 9:52pm
And I should include that ALL computer power supplies are switching mode power supplies.

Important when digesting the article I linked of inrush current above.

It directly apples to all desktop computers.


Posted By: Xaltar
Date Posted: 20 Dec 2016 at 10:00pm
Thanks for posting the article link Wardog, I remembered reading it last time you posted it but couldn't find it to share here. A very interesting read.

-------------


Posted By: Xaltar
Date Posted: 20 Dec 2016 at 10:02pm
Originally posted by Ky0sHiR0 Ky0sHiR0 wrote:

" rel="nofollow - Hello,
MOBO model: AsRock Extreme 4 Z170
I really dont know that PC should be constantly on power. That's really weird for me. I was always turning off power strip. What do You recommend in that situation ? I will check in home all cables. Should I clear CMOS too ?

Basically just shut down the system and walk away, it's that simple Wink

The only time my systems are ever cut from the wall is when there is a thunderstorm or when I am away for more than a day or 2. 


-------------


Posted By: Ky0sHiR0
Date Posted: 20 Dec 2016 at 10:28pm
I am coming home right now. Will check cables, and connection without power strip and post results :) 
Thank You guys


Posted By: Xaltar
Date Posted: 20 Dec 2016 at 10:51pm
Happy to assist Wink

As for the overclocking, your board is a solid one so in all likelihood you just have a poor overclocking CPU. My 6600k is also only properly stable at 4.4ghz, I can get it all the way up to 4.6ghz (like the validation in my sig) but it isn't stable and needs a lot of vcore. Overclocking is a very luck oriented thing, that is why some stores actually sell CPUs that they have tested stable at certain overclocks for a lot more than the retail value of the CPU. 

Google "Skylake overclocking guide" and see if you can find any info you have missed. On my system I find the easiest way to overclock is to set one of the EZ-OC modes (usually the lowest one) then adjust the multi higher until it becomes unstable. Once that happens I bump up the vcore 0.05v and try again until stable or I get to 1.4v at which point I stop and back off a little on both the multi and vcore. I would start with a fixed vcore of 1.2v and say a 42 multi and work my way up from there. Setting one of the EZ-OC modes sets a lot of things automatically for you like short and long power limit, LLC etc so you don't need to think about them. 

Good luck and let us know how it goes. 


-------------


Posted By: parsec
Date Posted: 21 Dec 2016 at 12:06am
" rel="nofollow - This issue reminds me of the Wake On LAN situation, when power to the PSU is removed, and then restored. When Wake On LAN is enabled in the UEFI/BIOS, the board will start for a short time to initialize the networking chip, and then shut off, when power is restored to the board. BUT that does not seem to be what is happening to Ky0sHIR0. You could try setting Wake On LAN to Disabled in the UEFI/BIOS, if you have that option and it is Enabled.

Personally, I have never had any of my ASRock boards start for a few seconds, and then stop, when AC power is applied to the PSU, such as after working on the PC. I have all Seasonic Gold and one Platinum PSUs, and one EVGA G2 PSU, which is made by Seasonic. Honestly, I have never had a PSU that would start for a few seconds when AC power is first applied.

Ky0sHIR0 do you use the power switch on the PSU, or just on the power strip? I mean, do you use the switch on the PSU, set it to Off, and then set the power strip to Off? I don't think you do that. wardog makes a good point about the surge, inrush current spike to a PSU.

As wardog said, I also do not turn off my power strips when I shutdown the PC. The amount of power being used by the PSU is very small, two Watts with a modern PSU. The +5V standby power from the PSU is only capable of providing 15W or less to the PC. My Kill A Watt power strip shows about 5 Watts for three PCs in Shutdown, S5 mode. I use multiple power strips, all with filtering and surge protection, the PCs on one strip, monitors and networks switches on another, and one main distribution strip.


-------------
http://valid.x86.fr/48rujh" rel="nofollow">


Posted By: Ky0sHiR0
Date Posted: 21 Dec 2016 at 3:26am
Hello all,
Thank You for all your posts. That was very fast and precise help.
I clear CMOS + reconnect all cables and now everything works flawless : )

About OC - currently im on Offset + 85mV with 4,3 GHz LLC level 1 and cache ratio 39.
I tried some of 4,5 GHz with 1,39 V ~ 1,408 V but it crashed after 5-10 minutes (1 core).
I think I just lost silicone lottery :)
But... about PSU topic - everything works and im really grateful :)


Posted By: Ky0sHiR0
Date Posted: 21 Dec 2016 at 3:09pm
If You have any suggestions about OC (different options or something) please respond :) 
Thank You :)



Print Page | Close Window

Forum Software by Web Wiz Forums® version 12.04 - http://www.webwizforums.com
Copyright ©2001-2021 Web Wiz Ltd. - https://www.webwiz.net