[mdlug] Seeking advice or at least commiseration - possibly solved

[Jeff Hanson]

Exactly why I suggested it.  I've had several persistent lock-ups due to
corrupt CMOS settings.

On Wed, Jul 29, 2015 at 8:24 AM, Ingles, Raymond <
Raymond.Ingles at dynatrace.com> wrote:

> > From: Kirk A Ellis, ThatCyberSecurityGuy
> > Subject: Re: [mdlug] mdlug Digest, Vol 107, Issue 9
>
> > Are there any beeps to indicate the problem?
>
> No, there weren't. Just spinning fans and two lights lit up on the
> USB-based media reader in the case.
>
> > Needless to say if you are getting nothing on the monitor or beeps it is
> either your power supply or motherboard.
>
> It may actually have been something else. Last night, I opened the case,
> pulled the card, and tried to boot with just the onboard Intel graphics. It
> did slightly better, for a moment - a BIOS screen, with a corrupted line.
> Then it froze. After that, I couldn't get the thing to put out any video
> signal.
>
> Naturally, I was depressed about this, but just on the off-chance, I
> pulled the CMOS battery and used the jumper to reset the BIOS. And whaddaya
> know, after that it booted! Even into Linux, no real issues. (Sad to only
> have one monitor, though.)
>
> I reset the BIOS the way I wanted, and then powered down and put the new
> video card back in. It still booted. I installed an overclocking utility,
> and used it to *underclock* the card slightly, as well as setting the card
> fans to run more aggressively.
>
> The computer passed several hours of graphical stress tests without
> crashing at that point, as well as some gaming (gotta do what you gotta
> do). That's not enough to conclusively state the problem is resolved, but
> it's a good sign. The CR2032 battery still had 3V open-circuit, but the
> real test is under load. I'll see if I can scare up a 100-Ohm resistor
> tonight.
>
> So, in the end, it could have been:
>
> 1. Depleted battery corrupting BIOS settings.
> 2. Otherwise-corrupted BIOS settings.
> 3. Too-aggressive clock settings on the card.
> 4. Not-aggressive-enough fan settings on the card. (Without the fan
> adjustment, the card could hit 78 Celsius. With the new settings, I haven't
> seen above 61 Celsius.)
> 5. Some combination of the above.
>
> So, I'll give it time to see if the problem re-occurs. If not, I might
> play with the settings a bit and see if I can allow a higher clock/less
> noisy fan settings.
>
> Tonight, I'm going to play with the Linux Nvidia drivers and explore
> clocking and fan controls.
> _______________________________________________
> mdlug mailing list
> mdlug at mdlug.org
> http://mdlug.org/mailman/listinfo/mdlug
>