cancel
Showing results for 
Search instead for 
Did you mean: 

Are we using all OC failure diagnosis techniques that we should be?

HalloweenWeed
Level 12
I just had an OC fail that got me thinking about OC failure diagnosis techniques. I had an OC profile that was stable forever in LinX and playing games, but just very occasionally gave me a BSOD, about every several days, usually 101. And also sometimes failed coming out of sleep mode. Already raised Vcore up 0.02V more (to +0.1V), with no difference, and that was hitting my maximum allowable core temps. So today I had another failure coming out of sleep, and this time I had the time to look into it properly; so I rolled the computer out of it's "slot," and read the debug LEDs. They were AF and DRAM_LED. The AF didn't tell me much, but the DRAM_LED did.

So I thought about my DRAM settings, they were Mfgr's spec: What could be off? Well the only thing that comes to mind is DRAM V. The mobo sensors and settings could easily be 0.05V off, as they are not nearly as accurate as a good Fluke DMM. I tried raising the DRAM V by 0.05V and viola, no nore glitches - so far. If it is stable long-term I might even try lowering my Vcore back to +0.08.


After this I started thinking, how many other people are looking at their LEDs when they have a problem like this with their OCs? They can be a great help. Did you even know, do you even remember, that the RIVE has 9 green "Voltminder LED II" LEDs, next to the "start" button, designed to help troubleshoot your OC failures? For some people, this is a reminder. For others, it may be enlightenment. I know that some of you out there have been using this all along. But I have not heard anybody speak of these (except the debug two-digit LEDs) when discussing thier OC issues, it's as if they don't even know. I'm even noticing that the ppl with troubleshooting issues aren't looking at the Q LEDs, I have to ask them, and then they check.

So this is a reminder for all of you who are OCing: Look at the other LEDs on the board. FYI.
i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
(EVGA site: ) And I have a second (wife's) computer, Eve.

Overclocking is useless to me if it is not rock stable.
6,819 Views
3 REPLIES 3

HiVizMan
Level 40
HW I like the way you think.


There are a number of simple tools that we can use as you have highlighted. The high level processing that is needed, is how to interpret the single LED meaning. That comes from experience and prior knowledge about how things work at a lower level of abstraction than most folks have or really need to have.
To help us help you - please provide as much information about your system and the problem as possible.

Zka17
Level 16
HalloweenWeed wrote:
So this is a reminder for all of you who are OCing: Look at the other LEDs on the board. FYI.


He-he, I know what you're talking about... 🙂 It's easy for me as I'm on a test bench, so all those LEDs are almost literally under my nose... - but those who have their mobo built in a regular case... and the case is ditched under the desk...

sectionate
Level 12
HalloweenWeed wrote:
I just had an OC fail that got me thinking about OC failure diagnosis techniques. I had an OC profile that was stable forever in LinX and playing games, but just very occasionally gave me a BSOD, about every several days, usually 101. And also sometimes failed coming out of sleep mode. Already raised Vcore up 0.02V more (to +0.1V), with no difference, and that was hitting my maximum allowable core temps. So today I had another failure coming out of sleep, and this time I had the time to look into it properly; so I rolled the computer out of it's "slot," and read the debug LEDs. They were AF and DRAM_LED. The AF didn't tell me much, but the DRAM_LED did.

So I thought about my DRAM settings, they were Mfgr's spec: What could be off? Well the only thing that comes to mind is DRAM V. The mobo sensors and settings could easily be 0.05V off, as they are not nearly as accurate as a good Fluke DMM. I tried raising the DRAM V by 0.05V and viola, no nore glitches - so far. If it is stable long-term I might even try lowering my Vcore back to +0.08.


After this I started thinking, how many other people are looking at their LEDs when they have a problem like this with their OCs? They can be a great help. Did you even know, do you even remember, that the RIVE has 9 green "Voltminder LED II" LEDs, next to the "start" button, designed to help troubleshoot your OC failures? For some people, this is a reminder. For others, it may be enlightenment. I know that some of you out there have been using this all along. But I have not heard anybody speak of these (except the debug two-digit LEDs) when discussing thier OC issues, it's as if they don't even know. I'm even noticing that the ppl with troubleshooting issues aren't looking at the Q LEDs, I have to ask them, and then they check.

So this is a reminder for all of you who are OCing: Look at the other LEDs on the board. FYI.


When my screen starts artifacting I will look at my leds. I use the leds if I fail at booting up to know what the problem is. ram/hd
Crosshair IV 1090T 4093.7 - Crosshair V FX - 8350 Batch # 1243PGT 4.422MHz - 4816.08 MHz 1.5v - GSKILL F3-14900CL9D-8GBSR(x2)1087.6 11-12-10 1.5volts - ViewSonic VX2439wm - Acer GD235HZ - SyncMaster P2350 - PowerColor Red Devil RX 480 8GBD5-3DH/OC - Current Mode: 5760 x 1080 - (retired) ASUS GTX660DC2O2GD5 - Nvidia 3D Vision Wired - HAF 932 - Corsair HX850 - Corsair H100i - latest quakelive.com - Day thirteen 60fps - as always I'm powered by Futurelooks.com