If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
A7N8X-E Memory Question / Blue Screens of Death
download a program called memtest86 and let it run a diagnostic on the
memory..maybe you have a bad stick...and non-matching memory may cause some issues also. |
#2
|
|||
|
|||
The Old Man wrote:
Hi all, I have an Asus A7N8X-E Deluxe based system (my sons computer) and have been receiving intermittent blue screens of death (I also here the hard drive power down) when playing games. Sometimes the error mentions nvdisplay but usually it doesn't but varies which random numbers. Sorry to be vague. It sometimes happens shortly after Windows has loaded and before I get chance to start a game or application. Things I've tried: I have switched my graphics card and the fault remains. I have ran Memory test for hours and it found no faults. I have run 3DMark2003 for successive tests without it happening. I have reinstalled my graphics card drivers and installed Windows XP Pro Service Pack 2. I have replaced the Power Supply. The temperatures seem fine and the case is well ventilated 18C to 25C motherboard temp, CPU runs about 38C-45C according to Asus monitoring software. The system was fine for months and not used much as my son is seldom here. Its driving me nuts as all intermittent faults do. I looked at the memory and noticed although one stick is single sided DDR 400 Crucial, the other stick (they run in dual mode) is double sided with chips on both sides of the DIMM (again a Crucial DDR 400). They are not the same as one was an upgrade from 512MB to 1024MB some time ago. Would mixing single sided and double sided DIMMS cause this error? The system specs: AMD XP2500 Barton Mobile version. Do you overclock this processor ? I suppose you do, as I don't see any reason to buy a more expansive mobile processor for a desktop. What if you run it at stock speed ? Asus A7N8X-E Deluxe 2x512MB Crucial DIMM (1GB total) Aopen FX5900 128MB Graphics Card with latest Nvida Driver (61.77?) Onboard sound 80GB IBM Deskstart HDD EIDE Pioneer DVD-Rom (which now doesn't recognise DVD's only CD's for some reason!) Well, if it does that even if you install it on another computer, the DVD reading parts are dead... WinXP Pro SP2 Many thanks for any advice on how to track this annoying problem. Mys on has given up using his PC because of it. Regards, Graham |
#3
|
|||
|
|||
In article , "The Old
Man" wrote: Hi all, I have an Asus A7N8X-E Deluxe based system (my sons computer) and have been receiving intermittent blue screens of death (I also here the hard drive power down) when playing games. Sometimes the error mentions nvdisplay but usually it doesn't but varies which random numbers. Sorry to be vague. It sometimes happens shortly after Windows has loaded and before I get chance to start a game or application. Things I've tried: I have switched my graphics card and the fault remains. I have ran Memory test for hours and it found no faults. I have run 3DMark2003 for successive tests without it happening. I have reinstalled my graphics card drivers and installed Windows XP Pro Service Pack 2. I have replaced the Power Supply. The temperatures seem fine and the case is well ventilated 18C to 25C motherboard temp, CPU runs about 38C-45C according to Asus monitoring software. The system was fine for months and not used much as my son is seldom here. Its driving me nuts as all intermittent faults do. I looked at the memory and noticed although one stick is single sided DDR 400 Crucial, the other stick (they run in dual mode) is double sided with chips on both sides of the DIMM (again a Crucial DDR 400). They are not the same as one was an upgrade from 512MB to 1024MB some time ago. Would mixing single sided and double sided DIMMS cause this error? The system specs: AMD XP2500 Barton Mobile version. Asus A7N8X-E Deluxe 2x512MB Crucial DIMM (1GB total) Aopen FX5900 128MB Graphics Card with latest Nvida Driver (61.77?) Onboard sound 80GB IBM Deskstart HDD EIDE Pioneer DVD-Rom (which now doesn't recognise DVD's only CD's for some reason!) WinXP Pro SP2 Many thanks for any advice on how to track this annoying problem. Mys on has given up using his PC because of it. Regards, Graham I've been experimenting here the last few days with that board and a mobile processor, and here are my results so far (still working on it): 1) My mobile doesn't like too much voltage. Setting it to 1.75 would give an overclocking failure at startup. Using 1.65 seems to make it happier. 2) In dual channel mode, I find problems at 200MHz clock speed. I get some stability at 180MHz (which gives DDR360/FSB360), and tried 12x180. 3DMark2001 = 16660. Rather than test for complete Prime95 stability, I moved on. 3) I moved the two DIMMs to slot 1 and slot 2, for single channel mode. This allowed me to get to 200MHz. I tried 12x200 and it will run Prime95 for about 30 minutes before it gets an error. 3DMark2001 = 17746. At this point, I don't know if this is still a memory problem, or if dropping the multiplier will help. (My only reason for the 12x multiplier, was to get it to fail faster, if it was going to.) I find memtest86 tends to pass all these configurations, and doesn't seem to be as sensitive a test as Prime95 (mersenne.org). The power consumption under load is about 110W, with the worst offender the +5V @ 15.6A, so I doubt it will be the power supply at fault. Over on nforcershq.com, there were reports that earlier BIOS may have been better for overclocking, but at this point I'm not really interested in trying to find and flash a bunch of old BIOS. I don't know if I could find an old BIOS like 1004 if I had to anyway. The board shipped with 1011 in it. To do stability testing, I don't immediately boot into Windows, as this could corrupt the disk or the registry. What I cooked up as a test method, is I downloaded the ISO for Knoppix3.6 (700MB) and made a test CD. Knoppix is a read-only Linux system, that sets up a ramdisk for storage when it runs. I use a floppy to carry test code, such as Super_PI for Linux and Prime95 for Linux. There are a few problems with memory management, so you have to be a little careful interpreting test results. I find I can run several copies of Prime95, by setting limits on memory consumption to say 32MB for each one, and then surf while the tests are running. Running "vmstat 5" in a console window allows monitoring free memory in the system. This is how I've been tuning the system. Once a config is stable enough not to crash for an hour or so, I can then connect up the hard drive, boot into Windows, and try 3DMark for a benchmark. So, things to try are, drop CPU clock and stay in dual channel mode, bump up multiplier. Or, move the two sticks to single channel mode, and leave the clock at 200MHz. As I'm still not finished, that is the best advice I can give so far. I haven't tried volt modding the Vchipset, as one post on nforcershq claims that for a system at 200MHz, you can only squeeze an extra 10MHz out with more than 1.6V on the chipset core voltage. The Vchipset mod used to help rev 1.04 A7N8X boards get above 180-185MHz, until the claim was a BIOS fix allowed higher speed operation. I am using CAS3 memory, and I vaguely remember someone suggesting that CAS2 memory might make the board happier. Maybe someone else who has tried both CAS2 and CAS3 can comment on that. HTH, Paul |
#4
|
|||
|
|||
The Old Man wrote:
Hi all, I have an Asus A7N8X-E Deluxe based system (my sons computer) and have been receiving intermittent blue screens of death (I also here the hard drive power down) when playing games. Sometimes the error mentions nvdisplay but usually it doesn't but varies which random numbers. Sorry to be vague. It sometimes happens shortly after Windows has loaded and before I get chance to start a game or application. Things I've tried: I have switched my graphics card and the fault remains. I have ran Memory test for hours and it found no faults. I have run 3DMark2003 for successive tests without it happening. I have reinstalled my graphics card drivers and installed Windows XP Pro Service Pack 2. I have replaced the Power Supply. The temperatures seem fine and the case is well ventilated 18C to 25C motherboard temp, CPU runs about 38C-45C according to Asus monitoring software. The system was fine for months and not used much as my son is seldom here. Its driving me nuts as all intermittent faults do. I looked at the memory and noticed although one stick is single sided DDR 400 Crucial, the other stick (they run in dual mode) is double sided with chips on both sides of the DIMM (again a Crucial DDR 400). They are not the same as one was an upgrade from 512MB to 1024MB some time ago. Would mixing single sided and double sided DIMMS cause this error? The system specs: AMD XP2500 Barton Mobile version. Asus A7N8X-E Deluxe 2x512MB Crucial DIMM (1GB total) Aopen FX5900 128MB Graphics Card with latest Nvida Driver (61.77?) Onboard sound 80GB IBM Deskstart HDD EIDE Pioneer DVD-Rom (which now doesn't recognise DVD's only CD's for some reason!) WinXP Pro SP2 Many thanks for any advice on how to track this annoying problem. Mys on has given up using his PC because of it. Regards, Graham Oh my G**. I am having the *exact some problem with my brothers PC. Getting random BSOD's, last one mentioning nv4disp.dll?? I think. Also getting "Page_Fault_in_non-paged area" BSODs. Changed pagefile size location and drive settings. Have rebuilt his PC and have been fighting this POS for weeks trying initially to even get a clean Windows install. Replaced PSU to get Windows to install. Now is BSODing frequently and have swapped tested and replaced almost everything at different times. System is Asus A7N8X-E Deluxe AMD1700+ changed to AMD2400+ New Maxtor SATA 80G HDD, old 15G IDE Maxtor for file storage. Replaced Thermaltake 360W PSU with old one for testing Uses No-name-ish 256Mb PC2700 RAM. Memtested for hours without errors Geforce3 Ti-200 64meg video card Running Win2K SP4 I hear your pain and sympathize. Let's nail this issue as I am sick or trying to rebuild his PC. :-\ J -- Jethro[AGHL] aka Phat_Pinger Reply Email: jeff (at) tibben (dot) ca |
#5
|
|||
|
|||
|
#6
|
|||
|
|||
"dino" wrote in message . .. http://www.nforcershq.com/forum/viewtopic.php?t=53194 here is some more info Hi all, thanks for the replies. Dino, I have tried Memtest86 and ran it for several hours with no problems. I can run 3DMark without problems too. I also read the thread you linked to and am wondering if it might be the Nforce drivers. On the site they say it might be the Nvida sound driver. (3DMark might be working becasue there is no sound.) I just checked and I'm using Nvidia Driver v4.27 with the audio drivers too. I just checked and a new version is out, v5.10 which I'm downloading now. When I look at the remove options under Nvida Driver in the Control Panel, there are a few other possibilities: NVIDIA IDE Driver - it does actually warn about this when installing come to think about it. NVIDIA Memory Control Driver NVIDIA nforce PCI System Management Driver Chong Woo: The system specs: AMD XP2500 Barton Mobile version. Do you overclock this processor ? I suppose you do, as I don't see any reason to buy a more expansive mobile processor for a desktop. What if you run it at stock speed ? Hi, I originally ran this Mobile CPU as it was guranteed overclockable out of the box with standard heatsink and fan and to be fair its been fine at 2300Mhz from www.overclockers.co.uk . I have dropped it to normal though and the problem persists. Pioneer DVD-Rom (which now doesn't recognise DVD's only CD's for some reason!) Well, if it does that even if you install it on another computer, the DVD reading parts are dead... Yes its def dead but not bothered about that at the moment, something I can replace at a later time. Paul, Thanks for the extremely detailed advice. I will try with one stick at a time to see what happens. Perhaps one stick is bad although Memtest86 has found nothing wrong. On the subject of the CPU, its currently running at 200x11.5=2300Mhz at 1.70v I'm using Asus BIOS v1010. Jethro, I'm sorry to hear of your problems too. Your errors are duplicates of mine, completely at random times but usually after starting a game or doing a windows task. Perhaps we can nail this! Regards, Graham. Regards, Graham. |
#7
|
|||
|
|||
In article , "The Old
Man" wrote: "dino" wrote in message . .. http://www.nforcershq.com/forum/viewtopic.php?t=53194 here is some more info Hi all, thanks for the replies. Dino, I have tried Memtest86 and ran it for several hours with no problems. I can run 3DMark without problems too. I also read the thread you linked to and am wondering if it might be the Nforce drivers. On the site they say it might be the Nvida sound driver. (3DMark might be working becasue there is no sound.) I just checked and I'm using Nvidia Driver v4.27 with the audio drivers too. I just checked and a new version is out, v5.10 which I'm downloading now. When I look at the remove options under Nvida Driver in the Control Panel, there are a few other possibilities: NVIDIA IDE Driver - it does actually warn about this when installing come to think about it. NVIDIA Memory Control Driver NVIDIA nforce PCI System Management Driver Chong Woo: The system specs: AMD XP2500 Barton Mobile version. Do you overclock this processor ? I suppose you do, as I don't see any reason to buy a more expansive mobile processor for a desktop. What if you run it at stock speed ? Hi, I originally ran this Mobile CPU as it was guranteed overclockable out of the box with standard heatsink and fan and to be fair its been fine at 2300Mhz from www.overclockers.co.uk . I have dropped it to normal though and the problem persists. Pioneer DVD-Rom (which now doesn't recognise DVD's only CD's for some reason!) Well, if it does that even if you install it on another computer, the DVD reading parts are dead... Yes its def dead but not bothered about that at the moment, something I can replace at a later time. Paul, Thanks for the extremely detailed advice. I will try with one stick at a time to see what happens. Perhaps one stick is bad although Memtest86 has found nothing wrong. On the subject of the CPU, its currently running at 200x11.5=2300Mhz at 1.70v I'm using Asus BIOS v1010. Jethro, I'm sorry to hear of your problems too. Your errors are duplicates of mine, completely at random times but usually after starting a game or doing a windows task. Perhaps we can nail this! Regards, Graham. Regards, Graham. An update on my testing. I am back to 12x180MHz in dual channel mode. I seemed to be able to pass memtest86, but was still having trouble with prime95. What I discovered by running Knoppix (read-only Linux) is that, if I run three copies of "mprime" executable, with the memory allocation limited to 32MB, the same instance of mprime fails each time (the first one launched). That implies a particular part of the ram has a reproducible fault. It doesn't imply the ram is bad, as I can pass memtest. (For the record, my Knoppix boot command is "knoppix26 screen=1280x1024 noapic". Running the 2.6 kernel fixed all the memory management problems I was seeing, allowing me to surf and test at the same time without a problem.) I was using the nominal timing for my ram, some cheap Azenram with LEI chips on it, rated by SPD for 3-3-3-8. Last night I discovered if I modified the timing to 3-4-4-11, that I am Prime clean. I ram three copies of Prime95 overnight and no errors were found. I suggest you set the memory timing up manually and relax the numbers a bit. (I was a bit worried that perhaps the BIOS wouldn't like those settings, but it POSTed fine after the change to 3-4-4-11.) I can run the nominal timing in single channel mode, but again, the bandwidth as reported by memtest86 is significantly reduced. It seems some people, even using fast 2-2-2 ram, have to back off at least one of the parameter to get stability. One guy ended up at CAS2.5 (2.5-2-2) with his fast ram. I've also been reading about CPC. It turns out that CPC stands for "command per clock", and is none other than the 1T versus 2T command rate option seen on some other motherboard families. It seems nforce2 BIOSes hide this option, and it is possible that some of the overclocking BIOS that have existed, were running 2T, with a big penalty on memory bandwidth. "CPC off" BIOS means 2T command rate timing. "CPC on" BIOS means 1T command rate timing. Basically, 2T timing is supposed to be used on a Northbridge, whenever the designers think a particular RAM configuration will load the address bus to such an extent that the address bus has to be set up for a cycle, before a second cycle delivers the strobe, to clock the address into the memory chips. On some Intel boards, this seems to be necessary when two double sided DIMMs are sitting on a channel. It is hard to tell what rules are being used with the Nforce2 chipsets, but the basic premise is that if a BIOS uses 2T command rate timing, a larger overclock results, but the lower memory bandwidth that results, defeats the whole purpose of the higher FSB. Anyway, executive summary is, try slower timing, like my transition from 3-3-3-8 nominal to 3-4-4-11 manual (the CAS stays at 3). Report back what you find, using stability testing like prime95 "torture test" from mersenne.org. They have a Windows version you can run if you want. Paul |
#8
|
|||
|
|||
"The Old Man" wrote in message ... "dino" wrote in message . .. http://www.nforcershq.com/forum/viewtopic.php?t=53194 here is some more info Hi all, thanks for the replies. Dino, I have tried Memtest86 and ran it for several hours with no problems. I can run 3DMark without problems too. I also read the thread you linked to and am wondering if it might be the Nforce drivers. On the site they say it might be the Nvida sound driver. (3DMark might be working becasue there is no sound.) I just checked and I'm using Nvidia Driver v4.27 with the audio drivers too. I just checked and a new version is out, v5.10 which I'm downloading now. When I look at the remove options under Nvida Driver in the Control Panel, there are a few other possibilities: NVIDIA IDE Driver - it does actually warn about this when installing come to think about it. NVIDIA Memory Control Driver NVIDIA nforce PCI System Management Driver Chong Woo: The system specs: AMD XP2500 Barton Mobile version. Do you overclock this processor ? I suppose you do, as I don't see any reason to buy a more expansive mobile processor for a desktop. What if you run it at stock speed ? Hi, I originally ran this Mobile CPU as it was guranteed overclockable out of the box with standard heatsink and fan and to be fair its been fine at 2300Mhz from www.overclockers.co.uk . I have dropped it to normal though and the problem persists. Pioneer DVD-Rom (which now doesn't recognise DVD's only CD's for some reason!) Well, if it does that even if you install it on another computer, the DVD reading parts are dead... Yes its def dead but not bothered about that at the moment, something I can replace at a later time. Paul, Thanks for the extremely detailed advice. I will try with one stick at a time to see what happens. Perhaps one stick is bad although Memtest86 has found nothing wrong. On the subject of the CPU, its currently running at 200x11.5=2300Mhz at 1.70v I'm using Asus BIOS v1010. Jethro, I'm sorry to hear of your problems too. Your errors are duplicates of mine, completely at random times but usually after starting a game or doing a windows task. Perhaps we can nail this! Regards, Graham. Further to my last post, I've upgraded the nForce drivers to v5.10 and so far touch wood it hasn't crashed. Graham. |
#9
|
|||
|
|||
The Old Man wrote:
snip Further to my last post, I've upgraded the nForce drivers to v5.10 and so far touch wood it hasn't crashed. Graham. Good to hear, Graham. I'm downloading the newest 5.10's also. The previous drivers I was using were from the Asus site. There were fairly old as i forgot to go to the source (nVidia) and get the latest drivers. *crossing fingers* -- Jethro[AGHL] aka Phat_Pinger Reply Email: jeff (at) tibben (dot) ca |
#10
|
|||
|
|||
glad to hear..I have stopped using the onboard sound due to my Z5300's not
being SPDIF speakers...but yes I have read issues of the SW IDE drivers I believe they are called. All my crashes are my own making (OC'ing too high..wrong timings ect.)...but I know when I format next am going to leave the SW alone. A7N8X-Deluxe rev2.0 Uber BIOS 1007 Barton 2500 Unlocked @ 11x200 OCZ PC3200EL 3x256 (2-3-3-6) etc...etc... keep us posted..am curious |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
I still don't completly understand FSB.... | legion | Homebuilt PC's | 7 | October 28th 04 03:20 AM |
nf7 dual channel memory question? | Christo | Abit Motherboards | 0 | September 14th 04 12:48 PM |
A7N8X Deluxe - more memory = slower? or Corsair XMS = slower? | Erik Harris | Asus Motherboards | 2 | August 8th 04 06:56 PM |
A7N8X Deluxe -- False Memory Failure Reports? | alan buckley | Asus Motherboards | 12 | December 23rd 03 11:51 AM |
A7N8X DELUXE memory issue (Ben you there?) | Bitsbucket | Asus Motherboards | 4 | November 23rd 03 03:42 PM |