Asus P8Z68-V Pro/GEN3 - Will not POST -- please help

F

fredgarven

Weaksauce
Joined
Nov 9, 2006
Messages
67
The new system I bought is as follows:

Antec P280 case
Asus P8Z68-V Pro/GEN3 motherboard
Corsair Vengeance 16GB (4 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 CMZ16GX3M4A1600C9B
EVGA 560Ti 2GB video card
Intel 2500K CPU
Cooler Master Hyper 212 EVO
Corsair HX750 power supply
2 x Crucial M4 CT128M4SSD2 2.5" 128GB SATA III MLC Internal Solid State Drive (SSD)

I assembled everything this morning and went to do a test boot. The system turned on but did not ever display anything on the screen. This board has LED indicators to help troubleshoot problems. The DRAM LED was lit solid. Next to the LED is something called a MemOK! switch. The motherboard manual says to hold the button down until it starts to flash and it will automatically adjust the memory so it will be compatible with the system.

I held the button and let it do its test and it still would not post. I removed all the memory and tried each stick in each memory slot and ran the test each time. I also tried them in pairs. The slots are color coded to indicate where to put them for dual channel so that wasn't the issue.

I even took out all the memory and powered it up. I took out the video card and plugged into the onboard DVI port. I reseated the CPU and cooler. I unlugged everything unnecessary and did a basic build to eliminate every possibility.

In doing all of the above, shouldn't I see something on the screen, even if it is nothing but errors? It should at least pop and say ASUS blah blah blah, press DEL to enter BIOS, right?

What do you guys think? Is it most likely a bad motherboard or bad memory? I'm leaning toward motherboard because I find it hard to believe that all four memory sticks would not function even individually.

I guess another possibility is I could have made an error somewhere. But I'm pretty sure I plugged everything in correctly. I undid everything and put it all back twice with the same result. I've been messing with computers for over 30 years now so this really should be old hat by now. But I could have missed something.

Anyway, please look this over and if you can think of anything, let me know. I will greatly appreciate any help I can get.
 
First two things that come to mind:
XMP profile has to be set in bios
Initiate graphics adapter option has to be set PCie/PCI in bios if using a vidcard
If using iGPU as first initialize option, have to use DSUB port for installing INtel graphics driver/control panel before using DVI mobo port
Cant just switch iGPU or vidcard mode around at will - they are mutually exclusive.
If you have Intel drivers loaded, will not boot in PCIe mode
If you have AMD/NVidia drivers installed, will not boot in iGPU mode
If you have no drivers, must be DSUB first, basic vga for bios
 
My first attempts at booting the system was plugging into the video card's DVI port. I should at least see something on the screen when powered up. I see nothing. I even tried a different monitor. Even if I plug into the onboard DVI port, I should at least see something indicating POST activity. I see nothing. I have never reached the BIOS. There are no drivers installed as there is no OS yet.

Are you saying if I use the onboard video, that I have to use the DSUB first?
 
Is your vidcard out?
DSUB from monitor to DSUB on mobo I/O?
Monitor on - yellow led?
Initiate graphics adapter set at iGPU in bios?
Did a cmos clear?
PC off
PSU rocker to "0"
Unplug PSU
Push in start button for 10 sec until lights go out
jumper on mobo to right 2 pins
Take out batt
Wait 30 min minimum, overnight much more likelly to succeed
Reverse proceedure
Suggest you read this thread very carefully
http://forums.anandtech.com/showthread.php?t=2208558
It seems screwing up iGPU initial boot corrupts the bios because its on die graphics

Edit: Also if mem not down all the way will give red led. These sticks are a bitch to get in
 
Last edited:
I would advise you clr the cmos and ensure you also have installed the CPU correctly and not over torqued the backplated/CPU cooler.

Once you have done that as was noted previously focus on just positng first witht he iGPU SO REMOVE your GTX card and see if the system posts correctly witht the iGPU inplace. As the memory you are using has been qualified internally that should not be an issue and i agree that it would be very unlikely to have 4 dimms fail.

If the system still fails to post with just the iGPU in place ( and CLR CMOS ) then i would say you have a faulty PSU or faulty motherboard.
 
cisco guy said:
Is your vidcard out?
DSUB from monitor to DSUB on mobo I/O?
Monitor on - yellow led?
Initiate graphics adapter set at iGPU in bios?
Did a cmos clear?
PC off
PSU rocker to "0"
Unplug PSU
Push in start button for 10 sec until lights go out
jumper on mobo to right 2 pins
Take out batt
Wait 30 min minimum, overnight much more likelly to succeed
Reverse proceedure
Suggest you read this thread very carefully
http://forums.anandtech.com/showthread.php?t=2208558
It seems screwing up iGPU initial boot corrupts the bios because its on die graphics

Edit: Also if mem not down all the way will give red led. These sticks are a bitch to get in
Click to expand...

I read through that whole forum post. I was able to get to the BIOS by clearing the CMOS and only using 1 memory chip in the B2 memory slot. I am able to also run a second memory chip in the B1 slot. If I put any memory chips in either or both A1 and A2, the system will not POST and the DRAM LED stays lit. Pressing the MemOK! button does nothing to resolve this.

When I got into the BIOS with just the B1 and B2 banks populated, I set memory to be XMP 1600 9-9-9-24-2N-1.50V.

I went to DRAM timing control and left everything in there default. Are there any other settings I may be missing? Or do you guys think the A1 and A2 memory banks are bad?
 
Before you RMA the mobo, you have to check two things
You have to first take the mobo out of the case
You have to take off the EVO HSF, take out the CPU, and with a 3" (minimum) round magnifying glass you have to look (in good light) at the pins in the socket and look for pins that are slightly bent in one direction.

Then you have to carefully reinsert the CPU, reinstall the EVO and put the 4 nuts on backside on JUST BARELY SNUG! Even with the fan vibrations, the HS is NOT going to loosen. Then, leaving the mobo still out of case, on your desk, try things again. Now, if the red RAM LED stays lit with A slots, its time to RMA the mobo.

If possible its always best to insert RAM before installing mobo in case, so as not to bend the mobo downwards excessively when its on the brass standoffs support. My case/case window allows me a support area to place a buildup of black electrical tape on the sheet metal under the central RAM slot position. Never use 1/4" long brass standoffs from old case hardware - only 5/16"
 
cisco guy said:
Before you RMA the mobo, you have to check two things
You have to first take the mobo out of the case
You have to take off the EVO HSF, take out the CPU, and with a 3" (minimum) round magnifying glass you have to look (in good light) at the pins in the socket and look for pins that are slightly bent in one direction.

Then you have to carefully reinsert the CPU, reinstall the EVO and put the 4 nuts on backside on JUST BARELY SNUG! Even with the fan vibrations, the HS is NOT going to loosen. Then, leaving the mobo still out of case, on your desk, try things again. Now, if the red RAM LED stays lit with A slots, its time to RMA the mobo.

If possible its always best to insert RAM before installing mobo in case, so as not to bend the mobo downwards excessively when its on the brass standoffs support. My case/case window allows me a support area to place a buildup of black electrical tape on the sheet metal under the central RAM slot position. Never use 1/4" long brass standoffs from old case hardware - only 5/16"
Click to expand...

I have removed the HSF EVO from the CPU. I have inspected the CPU. There are no pins. It is not like the older processors with pins that can bend. I have not overtightened the back plate or the HSF EVO. I would never do that. No matter which memory chips I use, I can populate B1 and B2. If any memory touches A1 or A2, it is a no go. I have gone through the BIOS extensively along with reading everything I can find online about this motherboard. There are no settings left. If it can read the chips in the B bank, it should be able to read the chips in the A bank.

The new board will be here on Monday. It will be no trouble at all to change it out. So I'll know for sure then.

BTW - I have tried every single memory chip in every single configuration in bank B and they all work. I am confident it is not memory. At this point I have nothing to point at but a defective memory bank A. However, in my experience, if one memory bank was defective, the board should not work at all. That is why I am thinking that I may be missing something.

Also, I have installed Win 7 and all the drivers and updates. So I know it is not the CPU causing the issue. In fact, I am writing this post on the machine in question.

I do appreciate the help so far. I would have not thought to put only one chip in memory slot B2 and clear the CMOS if not for your help.
 
Last edited:
erm, thnx for the info on the new CPU's, but the pins are not on the CPU - they are IN the socket
and there are 1155 pins, if one is borked it may only bork that one related feature
Tho that bent pin should trigger the CPU red LED, but sometimes it doesnt depending on which one.
 
Oh brother. Well, I did not look at the socket. I only inspected the CPU and saw no pins. At this point, I am just going to wait for the new board to arrive and see what happens.

Again though, I cannot imagine that if a pin was bent on the socket or the CPU that it would work at all. It seems that in that case, it would just not function on any level.
 
you posted while I was editing
look again
But its 95% now that you have a bad ram slots - possibly cold or hairline cracked solder joints. When you push in the ram you will see the mobo go down considerably. The further down it bends, the more chance there is of a prob. Before you put in RAM you should look at case sheet metal for arcing "poof" marks.
 
Thanks for the help cisco guy. It turned out to be bad RAM slots. I cannot figure out how the board would even work in that case, but somehow the A bank was no good but B was OK.

The new board is here and all is well.
 
I'm having this exact same issue with the non-Pro version of the Asus P8Z68 Gen3 mobo. Boots like a champ as long as you don't plug any memory into slots A1 or A2.

I bought this motherboard from Microcenter and I'm hesitant return it to them for an exchange because this is the 2nd out of 3 motherboards I've bought from them over the past four years that have had problems. I may take my chances RMA'ing through ASUS instead.

If ya'll had to choose, which return method would you go with? I know I'll have to wait longer if I go through ASUS but I'm at a point where i don't really feel like dealing with Microcenter.
 
Remember any board PRO and over you can request an advance replacement within the first year of the 3 year warranty this can help to cut down on down time.

Also general feedback all users should be very careful when installing heatsinks that use a mounting back plate many times installation can be aggressive and the heatsink assembly can be over screwed and can cause damaged to the internal threading of the screw hole causing issues with memory initialization/detection. All in all go for finger tight and watch out to not scratch the board.
 
Juan_Jose said:
Remember any board PRO and over you can request an advance replacement within the first year of the 3 year warranty this can help to cut down on down time.

Also general feedback all users should be very careful when installing heatsinks that use a mounting back plate many times installation can be aggressive and the heatsink assembly can be over screwed and can cause damaged to the internal threading of the screw hole causing issues with memory initialization/detection. All in all go for finger tight and watch out to not scratch the board.
Click to expand...

Is it possible that I've simply shorted some contacts on my motherboard (thereby causing the RAM slots not to function) with the installation of my CPU cooler? I'm using a Corsair H100 for this particular setup which must be tightened to the back-plate by hand (not by screwdriver). If so, then I might try installing the stock CPU cooler and retesting the RAM slots. I haven't done a thorough inspection yet but I don't think I've scratched the motherboard anywhere.
 
Good thinking there.
It could be you bent a pin in the CPU Ziff socket - that would do it. So you should pull it and look very carefully at the pins with a good magnifying glass.
Sorry to say it does seem like bad slots if the sticks working in either B1 B2 dont work in A1 A2 singly
However, I would also do it out of case with stock HSF, on desk, to eliminate any case mounted issues.
Due to trickle voltage, never remove or install Ram unless you first shut off PC, turn PSU rocker to off (0) and then push in "ON" button on case for count of 10 to discharge capacitors.
I guess you have tried ea ram stick in A1 slot by itself.
Do you remember if the ram went in really hard? I dont know how much these multilayered boards can take. I know the X79 is 7 layer. I think the Z68 is 5 or 6.
I guess you could spray some electric contact cleaner into A1 A2, but dont put in anything until it has completely dried (I use spray air can afterwards)

Just fwiw, Amazon has become my best option for mobo purchases (but only "stocked and sold by") Amazingly simple returns.
 
Let us know how it turns out, I have almost the exact setup you do and am planning on installing everything tomorrow night. I'm using 16GB Corsair 4x4, Hope its not going to be stubborn. Can I update the bios before I install anything? I saw the new bios is supposed to provide better stability for Pro Gen3.
 
So I worked on my pc a little bit last night. I removed the motherboard from the case and inspected the back of it. There were minor scuffs from the backplate but no major scratches and nothing that crossed any of the soldier points from the CPU to the RAM. I also re-seated the CPU a couple times but didn't inspect the pins in the socket, I will check into that tonight.

When I re-installed the motherboard I left the backplate off all together and attempted a boot up with a single (confirmed working) RAM stick in A1 and the H100 heatsink hand pressed against the CPU. I got the same result, the system would not boot and went into an automatic power-up/power-down cycle until I pulled the plug from the PSU. Once again moving that same RAM stick to B1 or B2 allowed the system to boot up without issue.

I also tried each stick individually in each RAM slot (A1 through B2) and confirmed that every RAM module is working as they would all allow the system to boot when plugged into B1 or B2, but none of them would allow the system to boot when plugged into A1 or A2 or any combination which included A1 or A2). In all the testing I never really had to apply an unusual amount of force to install the RAM in any slot.

Tonight I'll check out the CPU socket as well as my motherboard tray to ensure I'm not accidentally shorting anything with a misplaced screw/mobo standoff. If I don't see anything out of the ordinary I will probably return the board to Microcenter. Thank you guys for all the suggestions.
 
Last edited:
Just as a follow up, I inspected the socket pins and they all appeared to be lined up perfectly; didn't notice any missing or bent. I did have one extra motherboard standoff than I needed (from a previous installation) but removing it made no difference. I ended up replacing the board at Microcenter with a P8Z68-V Pro Gen3 and I have had no problems at all thus far. :D
 
You must log in or register to reply here.
Back
Top