New Build - R4 Black Edition main suspect

[Pkekyo]

Greetings, Republicans of the Gamings.

I wrote this post for a different forum initially, but in my googling for solutions to my issues kept bringing me here, I figured this could be a good place to ask for new input.
If anything seems out of place/unrelated to the topic at hands, that part was probably meant for a different forum.

-------------------------------------------------------

Right.

I've been building a new rig as of late (buildlog can be found here - my ranting begins on page 4), and ever since I got it put together earlier this week, I've been running into a metric ****ton of roadblocks, whoopsies and dingdongs. I'm posting a thread here, since I figured the people oogling buildlogs might not be the same people that are wading to their knees in troubleshooting all day.

First of all, the specs:

- Asus Rampage IV Black Edition
- Intel i7 4930K
- Kingston HyperX Beast, 2400MHz 64GB
- EVGA GTX 780Ti Hydro Copper Classified
- Custom liquid cooling loop (dual 480mm radiators)

What initially happened after putting it together and booting up:


- I was completely unable to install Windows (7 Ultimate X64), until I reduced the memory clock speed to 1600MHz
- Then I attempted small incremental overclocks, using both the "CPU Level Up" and "X.M.P." UEFI functions. This in turn led to a seemingly infinate loop of failed attempts to boot, further resulting in a perplexing number of different "Q-codes" at the point of failure. "34", "AA", "60", "66", "6F" and "43" were some of them. The only way I found to resolve this infinite loop, was by using the "MemOK" button.
- Once Windows and drivers were installed, the RAID array (Intel RAID on the motherboard's SATA 6G ports) suddenly, and inexplicably shat the bed and died. The Intel RAID splash-screen during post listed both the Samsung 840 EVO 500GB SSDs as "Not Recognized", and the drives were no longer listed anywhere in the UEFI. Deleted the RAID array, built a new one and reinstalled Windows.
- UEFI is reporting that my RAM sticks are in fact 1333MHz DIMMs overclocked to 2400MHz. No idea if this is an error, or if it's an eccentricity of the Kingston RAM.

Had another go at modifying the clockspeeds of the CPU and RAM - a repeat performance for the asinine loop of failures to boot.
- Got it to boot again, "MemOK" reduced the "Target DRAM Speed" to 1333MHz, whereas the "DIMM Post" section had them listed as running at 1066MHz.

These are all the settings under the "Extreme Tweaking" tab in the UEFI that were not set to [Auto]:

All Core Target CPU Turbo-Mode Speed: 4500MHz
Target DRAM Speed: 1333MHz

AI Overclock Tuner: Manual
CPU Level Up: Disabled
BCLK Frequency: 125.000
CPU Strap: 125MHz
CPU Core Ratio: Sync All Cores
DRAM Frequency: DDR3-1333MHz
CPU VCORE Voltage: 1.104V [Manual Mode]
CPU VCORE Manual Voltage: Auto

VTT CPU Voltage: 1.050V
2nd VTT CPU Voltage: 1.050V
CPU VCCSA Voltage: 0.956V [Manual Mode]
CPU VCSSA Manual Voltage: Auto
DRAM Voltage (CHA, CHB): 1.560V
DRAM Voltage (CHC, CHD): 1.560V
CPU PLL Voltage: 1.800V
PCH 1.1v Voltage: 1.100V

At this point @andyysplash brought to my attention that unlike the retailer where I bought the motherboard, Asus' spec-sheet had every memory frequency above 1600MHz listed with an (O.C.) suffix. This is where I tried finding a way to trick my motherboard into believing that my DIMMs were overclocked 1600MHz sticks. I have as of writing this, not found a way.

-------------------------------------------------------------------------------------------------------------------

I got some more troubleshooting done last night:

- Flashed it with the most recent stable UEFI version for this board (0701 x64)

- It appears that under UEFI --> Tool --> ASUS DRAM SPD Information, DIMM_C1, DIMM_C2, DIMM_D1 and DIMM_D2 are being reported as "Abnormal". Pulled the RAM sticks from the C- and D-DIMM slots, and put them in the A- and B-DIMM slots. No errors reported. After shifting around all the DIMMs in various combinations, only the aforementioned DIMM slots reported the "Abnormal" status, regardless of which of the eight sticks were in them.
- Also,the front-USB3 connectors have stopped working. They do deliver power for charging, but the computer does not react when I plug any USB devices like thumbdrives, eHDDs and card readers into either port. Tested with the front USB3 setup from my other rig, which I know to be functional - same result (or rather, lack of a result).

At this point, I began leaning towards the Motherboard being bricked before I even bought it.

- Last thing I did last night, was grabbing one of the Storage drives out of my current rig, and put it into the motherboard - the drive is detected in UEFI and Windows' Device Manager, but I cannot get it to show up as allocated space with data on it (which it does have). I might be doing something wrong, but the people I've spoken to says it should be possible to move the drive with data on it between two rigs. My Google-fu has failed me, the only search results I get are about bringing a full Windows installation over to a new computer, whereas I only intend to move over storage drives.

- Furthermore, the computer has frozen upon using the "Shut Down" function from the Windows start menu - it has been in this state for the better part of four hours now.

-------------------------------------------------------------------------------------------------------------------

Also, I've grabbed some screenshots from the UEFI. They are in their default .bmp format, so I will insert links rather than embedding them in the post:

"Extreme Tweaker" - SS1
"Extreme Tweaker" - SS2
"Extreme Tweaker" - SS3
"Extreme Tweaker" - SS4
"CPU Performance Settings" - SS1"GPU Post" - SS1
"DIMM Post- SS1"
"CPU Configuration" - SS1
"CPU Power Management Configuration" - SS1
"USB Configuration" - SS1
"Onboard Devices Configuration" - SS1
"Onboard Devices Configuration" - SS2
"Voltage Monitor" - SS1
"Temperature Monitor" - SS1
"Fan Speed Monitor" - SS1
"Fan Speed Control" - SS1
"Boot" - SS1
"ASUS DRAM SPD Information" - SS1 (I have removed four of the DIMMs, the ones currently not populated slots all report the status "Abnormal" with DIMMs in them)

-------------------------------------------------------------------------------------------------------------------

This is basicly where it sits as of now. I would greatly appreciate any and all help in the matter.

[jeepcoma]

Forgive me if I am misunderstanding the order in which you did things, but it seems like you are trying to do way too much all at once:

OC your CPU to a high speed (4.5 is NOT considered a guaranteed OC for all chips)
OC a large quantity of ram (8 GB density)
OC your RAM to 2400
Install on a RAID

First, I would focus on getting a stable install. This means going back to 100% default BIOS settings and just get things running to confirm stability. In my experience I have found Win 8/8.1 to be very susceptible to corruption from crashes and OC failures. After a good crash or two, all sorts of stuff starts going wacky and you'll fault the hardware, and start playing around with settings and downclocking, but the damage was already done. Running a Linux distro off a USB drive is a great way to mess with settings in the BIOS and not hose your main OS install.

That said, here are a few things:

1. What is the part number for your RAM? Is it sold as a single kit for 64 GB quad channel, or is it perhaps two 32 GB kits you have mixed together?
2. The fact that the BIOS reports RAM as "abnormal" is troubling; this could be due to a bad kit or mix of kits, faulty motherboard, or even just improper seating of the RAM (I've done it). Further testing will need to be done here to verify.
3. Don't worry about the 1333 report, it has to do with what the "official" DDR3 speeds are (I think this is from the JDEC spec for DDR3). Higher speeds like 1866, though common, are technically considered "overclocks". In any case, there's no need to "trick" the BIOS as to what your RAM is - when you are ready you can just enable XMP and it'll do what it's supposed to, or set the DRAM speed manually along with the timings.
4. You mention 125 MHz for strap and BLCK. You for sure want to start at 100. Doing otherwise is a guaranteed headache.

Like I said, I think it'll be helpful if you just go back to the basics, back to BIOS defaults, clean OS install, and verify stability in order to ensure your hardware is working, and then start making changes one at a time, so you can spot where things start to come apart. At this point, there are too many unknowns as to what could be causing your problems.

[Kosmosagamemnon]

jeepcoma wrote:
Forgive me if I am misunderstanding the order in which you did things, but it seems like you are trying to do way too much all at once:

OC your CPU to a high speed (4.5 is NOT considered a guaranteed OC for all chips)
OC a large quantity of ram (8 GB density)
OC your RAM to 2400
Install on a RAID

First, I would focus on getting a stable install. This means going back to 100% default BIOS settings and just get things running to confirm stability. In my experience I have found Win 8/8.1 to be very susceptible to corruption from crashes and OC failures. After a good crash or two, all sorts of stuff starts going wacky and you'll fault the hardware, and start playing around with settings and downclocking, but the damage was already done. Running a Linux distro off a USB drive is a great way to mess with settings in the BIOS and not hose your main OS install.

That said, here are a few things:

1. What is the part number for your RAM? Is it sold as a single kit for 64 GB quad channel, or is it perhaps two 32 GB kits you have mixed together?
2. The fact that the BIOS reports RAM as "abnormal" is troubling; this could be due to a bad kit or mix of kits, faulty motherboard, or even just improper seating of the RAM (I've done it). Further testing will need to be done here to verify.
3. Don't worry about the 1333 report, it has to do with what the "official" DDR3 speeds are (I think this is from the JDEC spec for DDR3). Higher speeds like 1866, though common, are technically considered "overclocks". In any case, there's no need to "trick" the BIOS as to what your RAM is - when you are ready you can just enable XMP and it'll do what it's supposed to, or set the DRAM speed manually along with the timings.
4. You mention 125 MHz for strap and BLCK. You for sure want to start at 100. Doing otherwise is a guaranteed headache.

Like I said, I think it'll be helpful if you just go back to the basics, back to BIOS defaults, clean OS install, and verify stability in order to ensure your hardware is working, and then start making changes one at a time, so you can spot where things start to come apart. At this point, there are too many unknowns as to what could be causing your problems.

This is excellent advice.

In addition to the above, I'd like to mention that the RIVBE is a very busy motherboard with additional SATA and USB controllers, Wi-Fi, Supreme FX audio etc. While it has been my experience that resource sharing between devices can be... hmmm.. picky at times you have not gone into much detail about what on-board devices are enabled.

As with the above poster I'd suggest simply loading the Optimised defaults in the BIOS and working from there. I would also strongly reccomend that you get MEMTest86 and actually test your RAM when you have them in slots that identify the RAM as normal. You can deploy MEMTest86 as a bootable image to a USB Key/stick/drive. Then leave MEMTest86 to run overnight so several passes can be made to verify the results, bnad memory will show up as red blocks on the main screen. It is only when you have verified that your components function at normal operational parameters that you should begin your overclock work. Good luck! 🙂

[Pkekyo]

I wouldn't read too much into the quoted BIOS settings - those were the settings that the "EZ Mode" performance setting enabled - ALL frequencies have been running at default ever since, except for the RAM that is running at 1600MHz. (See the linked screenshots near the end of my initial post).

As for the RAM:
I am currently writing this on a computer that is using the four RAM sticks the R4BE first reported as "Abnormal", all four of them running at 2400MHz without any issue.
I also exclude improper seating as an issue, seeing as I can't possibly keep improperly seating RAM in those four slots, with the amount of times I have jumbled around the RAM sticks to see if it would change the "Abnormal" reporting (it didn't) - There may full well be a problem with seating, but if that is the case, it could only be due to an error in the manufacturing process of the motherboard when the RAM slots were added.

The DIMMs are indeed from separate kits - but I find it unlikely to be the case, considering that mixing the kits in another computer does not cause any issues. (The memory kit isn't on the QVL for either motherboard.)
I also tried putting the RAM from this computer into the new rig, and it did not yield results.
I'll have to get back to you with the parts number.

--------------------------------------

The following has happened since I made the original post last night:

- Last night I created a bootable MemTest86 flash drive, and ran it.

PassMark MemTest86 V5.1.0 Free Intel Core i7-4930K @ 3.40GHz

Clk/Temp: 3410 MHz / 0C | Pass 81% ###########################
L1 Cache: 64K 63247 MB/s | Test 96% ##################################
L2 Cache: 256K44659 MB/s | Test 9 (Modulo 20, ones & zeros)
L3 Cache: 12288K 25333 MB/s | Testing: 0x100000000 - 0x8200000000
Memory: 33007M 15444 MB/s | Pattern: 0xCD02248B
RAM Info: PC3-19200 DDR3 XMP 1200MHz / 11-13-13-32 / Kingston KHX2400C11D3
------------------------------------------------------------------------------------------------------------------
CPU: 0123456789AB | CPUs Found: 12
State: -WWWWWWWWWWW | CPUs Started: 12 CPUs Active: 1
Time: 1:32:35 AdrsMode:64Bit Pass: 1 of 4 Errors: 0

^ The above is what was on the screen when the MemTest froze. It sat there for about an hour, at that point the only way I could get the computer to shut down was by using the rocker-switch on the back of the power supply.
After the MemTest went kablooey, I now get the Q-Code "00", as well as the red "CPU" LED by the 24-pin socket just sitting there...glowing...taunting me...

- After the computer being powered down for roughly 15 hours or so, I had another go at powering it up.
It did cycle trough various Q-codes, but froze at code 5A for ten seconds, before it restarted itself and then came up with Q-code 00, where it now is frozen. The board does no longer POST - powering it on does nothing but spin up the fans, ignite the red CPU LED behind the 24-pin connector, and display the Q-code "00".

This is a list of the Q-codes I remember the board freezing at from the first boot and onward:

34 - CPU post-memory initialization

AA - "Reserved for ASL

60 - DXE Core is started

66 - CPU DXE initialization is started

6F - System Agent DXE initialization

00 - Not used

There are probably more, I just didn't think to start writing them down at first.

------------------------------------------------------------

As for onboard devices, the following I know to be active:
- WiFi
- RAID controller
- Audio
- USB controller