Crosshair III crashing like it's it's job

Python14

Limp Gawd
Joined
Jul 29, 2004
Messages
191
Okay, here's the setup:

Asus Crosshair III
AMD Phenom II x4 955
Sapphire Radeon HD 5870
PC Power and Cooling 750w silencer
Caviar Black 640gb HDD
OCZ Platinum DDR3 1333 4gb in Dual channel

I did the build on Friday. It ran stable while loading up OS and apps (Vista x64) until midday saturday when I got two crashes from "hyper transport sync flood error" while playing WoW. I hadn't started OCing or tweaking settings yet, just used the stock settings that the board chose at install. Since then, it has basically been crashing randomly. I pushed it's voltage up from 1.35 to 1.4 incrementally this afternoon and it ran stable for about an hour, and then crashed. Accompanying these crashes have been application and OS errors ranging from game crashes to various parts of Windows shutting down (For instance, right now I can't connect to any network through the onboard network interface.) The initial HT sync flood error was what initially tipped me off to mobo issue, but at this point I haven't a clue. I've tried updating BIOS (to 1003) and moving back to BIOS that others had running stable (0903 current). I'd get cpuz screens but I can't get it running long enough to do it. Any ideas? Is this some other piece of hardware or what?
 
Have you tried fiddling with the RAM settings? Maybe what the motherboard automatically set them to isn't (for whatever reason) ideal for your particular sticks. Going further on that, try running Memtest86+ to see if there's anything wrong with the RAM to begin with (once set properly).
 
Can't run Memtest. Any given boot has about a 20% chance of successfully making it to windows, most just either hang somewhere between loading the OS and logging on or loop back to the boot up.

Another new symptom.....my power/reset buttons don't work and the CPU is around 48C atm (haven't run anything beyond just trying to boot it for the last hour) Only the Start/reset buttons on the mobo itself work. I honestly think the solution is an RMA, but I want to be sure it's just my mobo and not maybe the processor or something.
 
Memtest doesn't require booting into Windows... unless you don't have another computer to burn a disc with, or you have a 20% chance of booting into anything successfully.

I'd probably try to isolate the problem via process of elimination at this point, unless someone knows off the bat what might be causing your symptoms. If you have access to another desktop, see if you can swap and test individual parts, particularly the RAM (try different slots and each stick individually). Check whether everything is plugged in correctly, watch for shorts, and make sure the CPU and HSF are properly seated/applied. If every part comes out clean leaving the motherboard the only probable culprit, then an RMA is a given.

Did the power/reset buttons on the case work before? How were temperatures before? I've only had a couple of experiences with bad CPUs, but in both cases the computers they were in won't even boot.
 
Last edited:
Looks like it's gonna be a long week >.< I'll give memtest a shot, but the fact the power button doesn't even work leads me to believe there is something more than just memory errors or what have you.

*edit*

The power button issue just randomly appeared. It worked fine an hour or so ago, but now it doesn't. Temps have been a little high, but nothing absurd. CPU = 48C and the MB in the low 30s. I can boot alright right now, but I'm hesitant to stress it as I don't want to push another crash needlessly.
 
Last edited:
Yeah, most likely. Sucks to have this kind of stuff happen to a new build, especially one with an HD 5870. :(

*edit*

It does sound like the motherboard's the likely culprit, since it's unlikely that anything else would discriminate between external power/reset controls and the onboard ones.
 
Last edited:
Ran memtest, failed at the last second but during it, it said there were hardware problems but didn't give details. Then it restarted and went to a screen that let me choose either my OS or do memtest again. If I chose OS, it restarts and comes back. If I pick memtest, it says there is an error and goes back. ie: Loop any way I cut it.

*edit*
I put in a ticket with Asus last night and I'll wait until I hear from them before I do anything else, but it's looking like it's gonna be a RMA.
 
Last edited:
I just did a build with a M4A79XTD EVO and some of your problems are similar to what i just had to deal with. The main problem was that the board would randomly loose power and shut off. Wasnt psu related since i didnt buy a new one. My power/reset buttons didn't work off the bat. well they did work, but very intermittently. Also my ethernet port worked fine at first then just stopped working. tried reinstalling the driver but the driver said it couldnt find it. Then windows 7 stopped booting, it would try to load then a bsod would flash really quick and then a restart. I reinstalled 7 and i could get into windows for a little bit then i ended up with the same problem with a bsod when loading. Ran a memory test, it checked out fine.

Also, before the problems started i ran a test of 3DMark06 and it ran through it fine. scored really good considering my system (PII 955, 4 gigs ram and 8800GT got 13,3K)

i issued a rma with buy.com. shipping it out tomorrow. Considering a diff. mobo to go with. any recommendations? AM3 please
 
Please post your bios settings. Sounds like memory error, either settings or insufficient voltage. I have this board flying with a PII X4 965. It's very solid so I'm sure I can help u.
 
I set everything to auto except voltage, which I had around 1.325 (stock is 1.3). Memory was around 1.65 most of the time. Clock speed was at 3.2 (200 fsb x 16) and HT was reading 2000 on cpuz the last time it didn't crash immediately. Memory timings were auto as well (cpuz had them at 7-7-7-20)

I'll check my other settings when I get home, but everything is currently set to default settings.

Okay, ran home for lunch to play around and it's basically doing a repeat of yesterday. I set everything to stock speeds, reset CMOS, tried both the most recent and two revs back of BIOS and it's really just a matter of how quickly it crashes assuming it makes it all the way to the desktop. 7 out of 10 times, I doesn't completely boot. The other 3 it'll make it to the desktop and crash on anything opening CPUZ, Asusupdate (thank god for EZ flash), or just sitting there. Ram timings are as I listed and even with other sticks it's crashing. I went ahead and started an RMA for it because I really don't feel like playing 'How does my board feel today?" every time I want to use my computer. Should I add my processor to the RMA as well? It doesn't seem like anything was wrong with it, and I don't have another rig to test it on (all my friends are Intel or use older AMD chips). Thanks for the help so far.

I'm back at work now, but when I get home again I'll get as much info for you as I can. If nothing else I can learn something from ya (This was my first build since the Athlon barton core days :p)
 
Last edited:
Here are the settings via cpuz. I just got home from work and it booted up just fine, including using the case power button.

Core: 3210 mhz
multiplier: 16
FSB: 200.6 mhz
HT link: 1003.4 mhz
Core Voltage: 1.375v

Memory: Link to actual ram for comparison as it's obviously not running at the speed it should

Freq 669 mhz
FSB:Ram 3:10
CAS#latency: 9 clocks
RAS# to CAS# Delay: 9 clocks
RAS# Precharge: 9 clocks
Cycle time: 20 clocks
Bank cycle: 33 clocks
Command rate: 1T
DRAM V: 1.5v

*edit*
Just crashed while I was upstairs. Waiting for the validation email from CPUz so I can post that

*edit2*

Here is the updated validated CPUz I basically tweaked numbers in BIOS until they mirrored the numbers I was seeing stock 955s get online. The voltage and a few others are still getting some play though. Any feedback?
 
Last edited:
Ran a good 3 hours solid with no problems....then randomly two crashes in a row while playing WoW. Still can't really figure out how/why it's occuring. RAM settings seem stablized. Maybe more power to the DRAM?
 
Any changes since your last post? If you haven't already, setting everything (like DRAM voltage) to what they should be set to might help. If it isn't the motherboard (which still seems like the culprit), I keep thinking there might be a short or some other power-related problem. I had a system several years back that would blue-screen if the 24-pin cable to the motherboard was just physically touched.
 
have you tried running 1 stick of ram at a time? Then running memtest on it and see if it passes.
 
I've basically narrowed it down to a RAM issue. I manually adjusted all of the settings to be synced up correctly and had a pretty stable run playing L4D for a couple of hours and then a raid in WoW without any crashes. I did get a few fatal errors with WoW that after checking the tech forums, were reported to be the result of a RAM error. I'm going to test out my memory sticks this afternoon one by one (as well as some from my roomates intel rig) to see where the issue lay.

One question though, what's a safe operating range heat wise for stock cooling/voltage? CPU idles in the low 40s/high 30s and upper 40s/50 under load, as does the SB. The NB his been breaking 50 regularly under load though. Unfortunately it's in a slightly cramped quarters (waiting on a new case to arrive, so it's sitting in a cramped mid-tower case atm.
 
CPU should be good up into the 60*'s

If your sb is anything like mine, ASUS probe II goes crazy once mine hits around 45* (can do this at idle easy! without gaming). I used an old "The card Cooler" bracket with a 80mm fan pointed at to keep temps usually around 38*, and really heats into the low 40's once playing a game for awhile (radiant heat from GTX260)

Start with memory and see how that goes.

GL
 
I've basically isolated it to memory. Now it's just a matter of determining whether it's the sticks or something between them and the CPU.


/praying for sticks
 
 
Last edited:
I've pretty much solved the issues. I haven't had any problems for a few days now, though I will probably look into replacing my RAM in the near future, but I'll take those questions to the appropriate forum :p

Thanks for the help everyone though. It seems in my hiatus from computer building, I lost my touch and made a few rookie mistakes that I've now fixed :p
 
Congrats! Enjoy your new rig. :D

We all make rookie mistakes despite the number and frequency of systems built. My last one (however insignificant) was not plugging in the power LED cable onto my motherboard correctly (I got the +/- flipped), resulting in about two months of the power LED on my case not lighting up. Being insignificant, I simply ignored it. It didn't occur to me until I was jump-starting my car after returning from a trip a month and a half ago. :p
 
Well, the troubles are back. I went a couple weeks with an occasional crash, but not really enought to make me get anything other than annoyed. Now it's pretty much a fight to even play games sometimes.

I just did a clean install for Windows 7 and got a few Superfetch, Windows Explorer, and other random app crashes. Then WoW installer crashed. Then 10 minutes into playing WoW, I crashed again. It simply rebooted and that was it. Before installing Windows 7, I did hard drive scans with the WD tool and memory checks with the vista tool. Both came up clean, even after multiple scans. Any other recommendations?

*Edit*

0000000015: 2009-11-03 23:52:59:878 _IDEMDeviceDFP2Settings_0812.GetDFP2ITCFlag failed with status 2 Error Called by: ATI.ACE.CLI.Aspect.DeviceDFP.Graphics.Runtime.RT_DeviceDFP::privateRefresh processID:02816 threadID:( ) domainName:(CCC.exe ) assemblyName:(CLI.Aspect.DeviceDFP.Graphics.Runtime, Version=2.0.3568.36865, Culture=neutral, PublicKeyToken=90ba9c70f846762e) ------------------------------------------------------------------------------------------------------------------------

Got this error.....restarted in safe mode and wiped my GPU card drivers with driversweeper and installed CCC 9.111 and then got this same error again about an hour later playing wow (specifically, while porting to Dalaran from Trial of Champions)

Anything?
 
Last edited:
I'll bet it's the memory. OCZ takes way more voltage than it should for the rated speeds. Try more voltage and/or get some Gskill.
 
Back
Top