I recently purchased a DX58SO2 motherboard and I,m having boot issues. the problem is very specific and reproducible . It occurs in two different situations.
The computer is turned off for a period of 2 hours or more. (I'll call that a cold start) when the power button is pressed the motherboard powers on
runs for 2 seconds then powers off for 1 second ,then powers on again with no boot, all fans running and a 00 on the LED. I then press the Reset button
once and the board powers off for one second and powers on again and then boots. the computer seems to run very stable afterward with no other problems.
The computer goes into hibernate state and sits that way for 2 or more hours, I click the mouse and the board powers on but hangs with all fans running and
a 00 on the LED. I then press the Reset button (sometimes up to 3 times in this scenario) and the board powers off for one second and powers on again and then boots. also I would like to note that any application/s that were running at the time the computer went into its hibernate state were returned to the same
states they were in upon windows resuming, with no stability issues noticed.
This doesn't seem to be a simple turn the computer off and it won't boot back up problem. its time sensitive in some way.
My System Spec's:
Motherboard: Intel DX58SO2 (ofcourse) BIO's Updated using Express BIOS Update SOX5820J.86A.0765 and set to default
Processor: Intel i7 960
Ram: 6GB Corsair CMZ12GX3M3A1600C9 Vengeance PC12800 DDR3 RAM 3x 2048mb triple channel (listed as CORE i7 on the box it comes in)
GPU: EVGA Geforce GTX 460 2GB GDDR5 PCI-Express 2.0 (one card only, not running SLI at this time)
Hard Drive: Seagate ST3320418AS Barracuda Hard Drive 320GB 7200RPM SATA 3G 16MB cache (pluged into SATA port0)
DVD: Asus DRW-24B1ST 24X Internal DVD Burner (pluged into SATA port5)
PSU: Corsair TX950W Power Supply
Case: Cooler Master HAF 922
OS: Windows 7 pro 64Bit
Steps I've taken to try and resoulve this issue:
1. Unpluged SATA hard drive and DVD from motherboard and attempted boot.
2. Removed RAM and ran each stick indiviualy
3. From the BIO's I turned down the voltage on my RAM it was running by default at 1.5125v I turned it Down to 1.5v (per tech support suggestion)
this of course is the only change I have now made to the default BIO's settings.
None of these steps had any effect or helped in any way, the problem persists. I'd like to add that this was a painfull touble shooting session as I had to wait about two hours inbetween tests after successfull boots. to make sure I was reproducing the problem if it was going to occur again after a change.
It has been suggested to me by Tech support that I do not have compatible RAM . I was very supprized to hear this since the DX58SO2 supports the RAM type I purchased, but it turns out that my i7 960 CPU does not infact if you look at the specs none of the 15 Intel Processors listed as compatible with the DX58SO2 are spec'ed out for RAM over 1066mhz. I'm told that the Ram is handled by the CPU now and not the motherboard and that the the BIO's is forcing my 1600mhz ram to run at 1067mhz, which it is. it was suggested that this is the reason that my DX58SO2 is experiencing this boot issue and that if I continued to use this 1600Mhz RAM that I would probably start to experience other unpredicable side effects. I'm not particularly interested in over clocking, I chose the DX58SO2 because it seemed to have alot of potential for me to upgrade my system over time.
I'm was'nt realy So I'm biting the bullet and I have orderd 6GB of 1066mhz
Swiss ... My situation is exactly like yours. My shutdown comes before anything else can happen and then it restarts a second or two later. This keeps making me think it is a hardware issue that is somehow related to temperature. I'm currently thinking about the possibility that it has something to do with the PCIe 16 connection. I have my video card in the PCI Express 2.0 x16 connector that is closest to the CPU (top one when I'm looking at the boad in the case). I think this is considered primary and my guess is that we all have our cards here. Reseating the card has not made a difference for me but I'm considering the possibility of using the other connector ... farther away from the CPU. Yes, I know this is a desperation move, but I'm almost there.
I just remembered that this motherboard has Diagnostic/status LEDs that blink, go through color changes, etc on boot. I have the system shut down and cooling right now. When it gets cold I'm going to pull the side panel off the case and watch these leds during the boot process. This might give us something.
I left the computer off for 2.5 hours so it would cool down. I decided to try it first with the monitor turned off. With the monitor off, I pressed the power button on the case as I watched the bank of LEDs. The video LED started to blink first and all by itself. It blinked for a long time and then went solid. The other LEDs then started to blink one at a time and finally all of the bottom row of LEDs were on solid. Windows was up. I turned on the monitor and waited for it to scan and come online. It worked without any problems and Windows was fully up and running.
I now have the system down again. I will wait anouther 2.5 hours and start the monitor first this time, before I power up the board. I'll let you know what happens.
I got the '00' thing again from a cold start. The bios release notes didn't mention this problem so I didn't move to the newest bios. I've tried a 9800GTX+ as well as a GTX570 in different slots, without luck.
I gave up on this board and have since put together a core2 based system. It works 100% and helped me reuse some leftover parts.
Are there plans to fix this. I'd like an Intel representitive to tell us what is going on.
RAM passed 5.5 hours of MemTest86 V.4.20, without any errors at the present setting of: CAS 184.108.40.206 @ 1600MHz.
Double checked polarity from case wiring to motherboard, especially Power and Reset connectors; pulled DVD/CD burner off-line and rebooted.
While still taking seconds to get past codes b8 and b9, I did eventually end up at the login screen.
Once again, I decided to test front panel Power Button for induced S3-Sleep. Appears to enter S3 as it should; but this time Wake with front panel Power Button did not hang the system at the usual "00" but at code: bF (6F) with fans blazing. I left it like that to see if anything would happen for about ten minutes. Annoyed, I hit the front panel Reset Button, with the now momentary pause at codes: b8 and b9, finally offering me the option of: Booting the computer normally or otherwise because it did not shut down properly.
I'm beginning to run out of ideas myself!
Even though I did pull both the power cable and SATA-II cable from the optical drive, I now see in the Event Viewer: Event 7026, Service Control Manager – (The following boot-start or system-start driver(s) failed to load: cdrom) I guess I'll need to disable the optical in the BIOS also.
I should probably run another disk check, given all of these failed reboots!
I was being facetious about the green grass, but this is nowhere near being funny! Another day of one step forward and two steps back! I feel your pain!!!
I wonder if Intel will take my 'expensive' premium board back in exchange for a 'Green Grass' board also?
Regarding the HP LP2475w monitor that we both have, look in your Nvidia control panel, you'll notice that according to Nvidia this HP monitor is running at 60Hz, but if you go to the Start Menu, type into the search bar: dxdiag.exe From the DirectX dialog panel that comes up, go to the monitor tab and you'll see that it shows this monitor running at 59Hz. What gives?
I mentioned in a post earlier, somewhere here or elsewhere, during installation of W7, the process didn't recognize the displayport connection. I had to shut down and swap over to DVI before I could continue with the installation. Sometime later, after the Nvidia drivers were installed, the displayport connection worked as it should. I've checked the Windows Hardware Compatibility List against this monitor and it is indeed listed as compatible for 64-BIT Windows 7.
My monitor's firmware is: GIG 072. The OSD shows the monitor to be running 1920x1200 @ 60Hz.
Maybe later, I'll dump the Nvidia drivers and see how Microsoft picks up the monitor, followed by a check into weather the 59Hz/60Hz have evened out between the different reporting sources.
We'll if and when I return mine it will be to the company I bought it from. they have a 30day no questions asked as long as its in working condition guarantee. I know you were being facetious, my scene of humor is heading that way too over this past couple days. but the grass is always greener is true. I've been looking for a new Board, I mentioned EVGA but I'm by no means settled on them. I already looked for a month and a half before I bought this Intel. EVGA was my second choice. But I'm so up in the air about it because their don't seem to be any good settled in LGA 1366 Motherboards. hey all have there problems and basically your taking your chances. I only paid $250.00 for this board. But I'm not made of Money.
I did think of another thing to try though. just when I think I'm all done another idea pops into my head.
I had reinstalled the Intel "Extreme Tuning Utility" and I decided to try and run the Autotune feature. I set it to only run with in system specifications for safety.
it changed my XMP profile back down to 1066MHz and ended up adjusting it to 1072MHz and set my CPU to 3.48GHz for a system performance improvment of 0.5% . I know but,.... I just wanted to see if it would do something for the Boot Issue. Unplugging the power, reset and LED leads had no effect.
Toxicwolf, I had been watching my LED's with case open since first day of testing. the thing is, none of those LED's do anything unless the board posts. its not doing anything but powering on, powering off and powering on again, no LED's no windows no anything, unless I press the reset button.
I did do some render testing durring this past few days, I was only seeing small increases in render speeds between 1066MHz and 1600MHz. I've been using a small scene as a benchmark. I was rendering that scene at 58 sec at 1066MHz, same scene at 1600MHz rendered in 57sec. not realy much difference.
So far I haven't been able to crash it today. I'll let it sit overnight and give a try again in the morning. I was looking for the flash that I thought might be coming from the Watch Dog Fire/Back to BIOS command coming on. I noticed that they had problems with it in a past bios update.
Sorry I can't say the same. I've done 3 cold boots today, with varying results. Twice the all familiar hang at "00" and the third time at "bF" or as Swiss thinks, more likely "6F".
Tomorrow I may try your boot up with the monitor powered down trick. Not a real solution but could certainly be a good indicator of a problem within the PCI-E slot. I thought of another thing we might be able to try: Disable Power Boost in the BIOS, which is tied to EIST and the C-States. I think, disabling Power Boost, automatically disables EIST and C-States...see what happens there.
Looks like no more S3-Snooze States for this computer, until Intel Techs give us a solution, this computer will need to run wide open all day, and powered down at night.
Regarding the 59/60Hz thing.
Many LCD monitors are designed to work at a TV-compatible 59.94Hz refresh rate, versus the normal 60.0Hz of computer displays. I'm assuming this is more common with 16x9 versus 16x10 screens. When this occurs, Windows 7 adds two refresh rate options: 59Hz so that media programs know the system is TV compatible, and 60Hz because many games will fail if they can't find at least a 60Hz refresh rate. Functionally these both map to the exact same setting internally, 59.94Hz, and the system will show 59Hz even if you select 60Hz.
The Nvidia control panel is likely reporting the REAL refresh rate, versus the system-reported "descriptive" refresh rate, and simply rounds 59.94 to 60.
Thanks Sunfox, That's helps.
I did a stone cold start this morning with the monitor on and it came up without a problem. There are only two things I've done that I haven't really talked about here before. One: I ran a chkdsk /r over the C drive and I finished closing up the system and moved it from the build table to the floor under the desk where it will stay.
This is temperature related, intermittent and sounds like a hardware problem. I just can't figure out what causes it to shut off all the power and then restart the system with no beep error codes. I'm using a Lian PC-B25F case, so we aren't using the same cases. I've run a total of 14 hours of heavy rendering over my Kingston 2133 ram without any problems. I'm not giving up. I will keep watching and learning.
I'm strating to feel like the "Good Old Days" where people fixed their televisions by walking up to them and giving them a good, solid slap.
(Possibly Not Answered)
I too did a stone-cold startup this morning, since we have the same monitors, including your trick of powering the monitor down from its main power switch, until after POST.
I switched power back on at the power supply, hit the front panel ON button. Boom! Instantly landed and stalled at code "00" with fans blazing for just a split second, automatically reset itself, and again, snagged at "00" with fans blazing, until I hit the front panel reset button then it went on through POST to login screen.
Yes, Sunfox, Thanks for posting that explanation for the 69Hz-60Hz thing!
Well, I've got to run up the road, take care of some other things. Meanwhile, I'll leave the rig completely powered down, try another cold-boot when I get back.
There is one other change I made that I forgot to mention. I went into the bios and changed the boot section. I disabled booting to everything except the hard drives and the optical drives. I set the hard drives above the optical drives and put the C drive at the top of the list. I doubt this makes any difference, but who knows.
I've been thinking about the ram issue. How many new computers out there do we think have 1033 ram. I went to the Dell site and did their build your own thing with a gaming computer. The processors available were the i7 960 and the i7 990x. Their ram choice was 1333 … 1600 … or 1868. I understand the intel tech people wanting us to look at ram issues instead of the processor, bios or motherboard, but I really doubt that all those computers out there being built with 960 and 990 processor using 1600 ram are failing.
Here I sit again, over 2 hours off and back from a cold boot, following exact same sequence as above. This time it did the brief, split-second hang at "00", reset itself and went on through POST, without need of my intervening with the front panel reset button.
I don't think our RAM is at issue here. Before I swapped my 6GB Dominator (TR3XG1600C8D) out for the 24GB Vengeance (CMZ24GX3M6A1600C9) I was experiencing the same radom, unpredictable re-boot hangs at "00" with the Dominators running at (Default) 1033MHz, as well as the Vengeance modules running at 1033MHz.
I've gone through setting the boot order, several times, even unplugging all unnecessary drives except the system drive, without any effect upon the symptoms.
I just wonder when how much longer before an Intel Tech chimes in here, to let us know what they intend to do about this, or what they think might be happening?
They aren't Failing, there are a lot of Motherboard and LGA 1366 processor combos out there that aren't having these problems. I've been doing alot of looking around and seen alot of different posts for different issues on different LGA 1366 Combo's. the most common thing I notice is that you have to over clock the CPU to take advantage of the XMP profiles, this seems to be different for different motherboards. all of the Manufactures list specific RAM that they have been tested with. some of them can run RAM up to 1333MHz before over clocking and some 1600MHz before needing to be over clocked. there were no other boards Beside the DX58SO2 that are reporting enough boot issues to get my attention. EVGA had a few but they ended up being bent pins according to the posts. however EVGA also seems to have a problem getting RAM to report properly. you see alot of posts for the EVGA boards only reporting 3.99GB of 6GB of RAM, also blamed on bent pins. seems to be a quality control issue to me. I'm feeling that way about this board too I'm thinking there is something physically wrong with the boards. it's not a bent pin though I went over mine with a fine tooth comb and couldn't find a single one. I'm RMAing but still haven't figured out what new board to get.
Running the extreme tunning utility didn't help had the same problem I tried 3 different make and manufacture of RAM didn't help BIOS update didnt help. I must have tried atleast 50 different BIOS configurations, Unplugged HD,DVD and USB devices. unplugged All front case buttons and leads Including front USB. I'm sure I've left a few things that I tried out.
Note: I was able to get good results with the Vengence RAM and the settings I listed before, but if I changed a single thing that went away. I've built several machines in my life time, ths is the most work I have ever put into any of them. My conclusion, The board is flawed.