System Locks on first boot of the day

[pnicholls]

Having a wierd issue with the GA-MA770UT-UD3P Motherboard and
F3-12800CL9D. The first boot of the day or the first boot after I get home from work I get memory errors using both windows memory diagnostic and memtest provided by the Hiren boot disk.

Tests on the MB and CPU turned up nothing.

The computer will freeze or I will get memory errors and a crash. After about the third reboot the system will become stable. I can run windows memory diagnostic and get no errors but memtest still gets errors.

I have done the following after reading this form:

(1) removed and cleaned the contacts.
(2) reseated the RAM modules.
(3) manually set the times.
(4) increased the voltage to the ram to 1.646. Not the motherboard goes no lower than 1.600 volts. I am NOT trying to overclock the ram. I'm setting it to 9-9-9-24 and 1333mhz.

Any suggestions? I have to rma the ram back to newegg soon.
All of the event logs seem to suggest memory issues. All the bugcheck codes also.

Frustrated and perplexed.

[gpb500]

Sounds a little familiar. See my post above. I've never seen this problem prior this build, but maybe others are running into it too...? In my case, I get errors on a cold boot...but if I cycle power off/on I'm good until the next cold boot. Replaced mobo and problem remains.

>> All of the event logs seem to suggest memory issues. <<

Memtest86 will confirm if there is any doubt and you won't need to risk OS corruption.

[pnicholls]

Thank you. I have aged a couple of years trying to solve this problem.

This is not my mission critical computer, it's for gaming only.

Did you attempt to swap out the power supply? It seems to me that the requirement to power cycle to get a sable system suggests something hinky about the power supply. As it happens, I have an extra power supply which I will swap in and see if that solves the problem.

Will report back on how it all works.

[gpb500]

It's not the power supply in my case. I updated my thread this morning, turns out my second pair is stable on a cold boot. Memtest will isolate a range of addresses of memory failure, and in my case, they were concentrated near a single address rather than scattered all over. So...out of 8GB, I was getting 70-ish bit failures...and that certainly seems to point to bad memory.

That said, I'd wait for the tech to get with you just to verify you don't have something else to check first.

[GSKILL TECH]

What is the end of the model number of that memory package? Try testing them individually to see if one stick will produce errors while the other does not.

Thank you
GSKILL SUPPORT

[pnicholls]

The model number is F3-12800CL9D-4GBNG. I purchased one pair from newegg.com and will the RMA'ing them Wednesday.

It is difficult to test them separately. I have the cold boot problem twice a day. Once in the morning and once when I come home from work. I did test them separately using the windows memory diagnostic. In the morning I get LRAND errors. After power cycling, the run fine. I've spent two weekends messing with this and that's enough.

[elysoto]

There may be some other information in this thread.

http://gskill.us/forum/showthread.php?t=2479

[GSKILL TECH]

4GBNQ? Hopefully the new set will be better. Please keep us posted.

Thank you
GSKILL SUPPORT

[pnicholls]

This morning I booted my computer and as soon as it seemed ready to use, I shut it down. I then unplugged the power connector from the power supply. I reconnected it and powered the system on a gain.

And it seems the system is stable.

Now if I could just figure out WHY this works.

[elysoto]

Arg the Round Trip Latency fix didn't work.

The only thing that works is what you stated, powering down the PSU.

[gpb500]

Same here, was a no-go. 170 errors out of the full 8GB. RMA'ing one pair.

[pnicholls]

Ok, so I replaced the ram chips with a set of 1333Mhz with tighter timings (7-7-7-20). Put them into my system when I got home from work. First boot of the day with the new memory (after manually setting the timings) and I have, so far at least, a stable system.

So as of now I would have to say it was the memory. Perhaps a flaw in those specific chips, I don't know.

I won't call it cold boots without issues a few times more.

Note that I shut down to take a short nap (had the flu) and rebooted and got a spontaneous reboot. Checked the event logs and no bugcheck errors or even a minidump file.

Unrelated glitch?

[gpb500]

Ok, so I replaced the ram chips with a set of 1333Mhz with tighter timings (7-7-7-20). Put them into my system when I got home from work. First boot of the day with the new memory (after manually setting the timings) and I have, so far at least, a stable system.

Well, that's good news.

Note that I shut down to take a short nap (had the flu) and rebooted and got a spontaneous reboot. Checked the event logs and no bugcheck errors or even a minidump file.

Unrelated glitch?

I RMA'd two sticks, still running two sticks and stable. Tried "sleep mode" yesterday...a few hours later, hit the space bar, entered password...BSOD. I haven't a clue what caused it. It could be W7, drivers, or memory/cpu. Unrelated glitch? Who knows. Everything else for me has been stable. I've been booting with memtest to find no errors...so at least this single pair seems good. Good luck, bummer about the flu.

[Vorg]

Thermal expansion. I used to work on vcr's, monitors, etc. And we had a can of cold spray and a heat gun for finding this kind of problem. Some parts fail when cold, some when hot. Sounds like you had ram that didn't like it cold.