GA-P67A-UD3-B3 power on issue

[Gulliver_69]

Hello, and sorry for my poor english.
I have a motherboard GA-P67A-UD3-B3 which does not start until I switch off the power and take off the CMOS battery for about 30 seconds. After this procedure, the computer starts with the default settings and I have no problem until I power off the system. At that point, if I want to power on the computer again, I must to repeat the same procedure. I read on the forums that this defect is quite common on GigaByte's P67 motherboards: I tried to update the BIOS with the new version F8 (I've updated also the recovery BIOS, via the command ALT-F12, because I've read that sometimes this solve some stability issue) but nothing has changed. The motherboard is under warranty, but before I send it, I want to know if there is a faster solution.
N.B.:
I am a computer technician: before write here, I made ​​a test with a new power supply but without results. I've tried also to start with only one DIMM at a time, but nothing happens. The DIMMs are 2x4GB Corsair Vengeance (CMZ16GX3M4A1600C9B). I would think a hardware defect, but I've read that many people have the same problem. I've also followed the procedure on this topic: http://forum.giga-byte.co.uk/index.php/topic,8392.0.html

Thanks to everyone for the help. :-)

[Lsdmeasap]

So you have both BIOSes, main and backup flashed to the latest BIOS?   Are you sure it flashed the backup properly?   Just checking, it does show on a black screen that it's updating the backup BIOS while it does it, so if you never saw that when you used ALT+F12 then it didn't flash the backup.

Have you tried using other memory, just to be sure it's not that memory causing issues?  Have you tested that memory with Memtest86+ one stick at a time lately?  If not, please do, and run at least 5-10 full passes per stick (One at a time) to be sure it's OK.  Here is the latest version of Memtest86+
http://www.memtest.org/#downiso

Don't clear the CMOS with the board powered on as I mentioned in that thread, I only said yes to him because he directly asked about it.  The normal method to clear CMOS is to turn off the PSU and then short the CMOS pins.   And extended method would be to unplug the PSU or flip it off in the rear, then press and hold the case power on button for one minute, then remove the CMOS battery, and then short the CMOS with a jumper and let it set overnight or 8+ hours.   Then put the battery back, turn on the PSU, and remove the jumper, then try to start the motherboard.

[Gulliver_69]

Thank you for your answer, Lsdmeasap.
Yes, I've flashed the backup properly and yesterday I proceeded with the procedure you described to clear the CMOS. This morning I've powered on the PC but, after a shut down, I've got the same issue. Then I've tried to go back to original BIOS, version F4, but this time I decided to flash it under DOS, with FlashSPI. Then, Ive loaded the optimized settings and I've updated the backup BIOS with ALT + F12. The problem occurred again, I do not know what other attempts I could do...
Unfortunately I have no other sticks of ram available for testing but the fact that the system, once started, works without problems for several hours leads me to think we can rule out a compatibility problem between the motherboard and memory modules. Instead, it seems that the motherboard is unable to provide the needed boost to start the system: like a car, which does not operate the starter...
I do not know if it can be a hardware defect or can be solved by changing some of the supply voltage within BIOS. I have not overclocked anything, also because the processor is a non-"k" and I am not familiar with motherboards based on P67 chipset. But I wonder if there might be a voltage equivalent of the Core2Duo Vdroop?!?

[Lsdmeasap]

Yes, there is Vdrop and Vdroop, but those should not affect you at stock speeds.

You may need to RMA the board, as the latest few BIOSes should have fixed these issues.  Older BIOS like F4 will not fix them for sure!

[Gulliver_69]

Hello Lsdmeasap,
thanks for the help you gave me. Today, accidentally, I found the solution attempting the tests most unlikely, dictated by desperation. The computer was connected via an extension cord, and I decided to replace it: I discovered that the defect was probably due to the fact that the first extension cord had no ground wire. In fact, since I changed it, I powered the PC several times and it always started at the first turn.
I've also reflashed the latest BIOS F8, and for now all is well.
Thanks again, and good evening. 

[Lsdmeasap]

Very odd to hear, but it's great that you found that out by accident!!

Nice to hear all is running smoothly now, post back anytime if you need more help with anything!

[Gulliver_69]

I apologize for not replying earlier, but I had little time to devote to my computer.
Finally, I discovered that the problem of power on the computer was due to having enabled the ErP: every time I turned off the computer, this does not even restart by pressing the power button. The only way I had to reset the CMOS by removing the power supply and backup battery.

Now I've upgraded my MB to the UEFI BIOS u1a, and the UrP option works fine.
Problem solved!!! :-)

[SalkinA]

I actually experienced the same issue this week. I have a Gigabyte GA-P67A-3UDP-B3 and read that "ErP Support determines whether to let the system consume less than 1W of power in S5 (shutdown) state" which sounded good so I enabled it. Didn't think much of it until I shut down my PC and the next day it was completely dead. Not even then I thought much about it since clearing CMOS and at several times also removing the battery usually, welll, clears the CMOS?  I've never experienced that you actually have to save the settings again to avoid a now disabled ErP setting still causing issues. It wasn't until I searched google for "Gigabyte GA-P67A-3UDP-B3 battery" that I found this topic which basically saved me from buying a new graphics card and/or a new motherboard.

This was nby far the hardest issue I've had to resolve with a PC, regular hardware failures are way more consistent and logical in their "failure". Having a BIOS setting that you can enable and have causing issues even if you do clear CMOS AND remove the battery is just weird.

So basically this bug was NOT solved even in F9. This together with the fact that this motherboard was supposed to "support" Ivy Bridge, which it does, kind of, at least as long as you don't want to overclock (you can't raise the multiplier above 38-39), will probably push me towards buying another brand the next time I want a new PC or someone else needs advice or help putting together a new PC.

It's a shame since other than these 2 major issues, they were for me at least, it's worked flawlessly.

I don't expect it to be fixed at this point in time, but it makes me wonder, how many with these P67 boards have had this issue and haven't been able to locate the issue but thought buying new stuff was the only solution? I've done my fair share of problem solving around PC hardware before so I'm not completely inexperienced, and even I was on the verge to simply replace my motherbard. Which would have solved the issue and left me thinking something hardware related was the cause (and not a BIOS setting as it really was).