| |||
|
Hello. :) I bought this Asus P5Q Pro with the P45 chipset for my new build which I just did this past weekend. Well, first the brand new WD 250 GB SATA HDD I installed Vista x64 on took a dump after only a few hours, but other than that things have gone OK. I do have a memory problem though... I think, but I'm not sure. I have Vista Ultimate x64 installed on this, with an Intel quad core 9550, and I tried out the memory diagnostic tool in Windows. It restarted, ran its test and then reprted I had malfunctioning RAM. I have 4 2 GB sticks of OCZ Reaper in it, 1066, 5-5-5 15, 2.2 v. After several hours of testing one stick at a time, in each DIMM, and various combinations, I think I've determined it's the B1 DIMM slot. Well, I don't ever get the hardware failure message unless I have a stick in that particular slot, and it doesn't matter which stick I use - a stick of it in A1 tests perfectly, but in B1 fails. The kicker though is that even with only one stick in B1 - even though the test reports it failed, my PC will still boot and act just fine. The amount of RAM reported in Task Manager's Performance tab is exactly what it should be, and I don't crash, no BSOD's,no error messages, and it appears to be working fine. I've also tested every stick in another PC, without failing once. And with all 8 gigs installed, it shows 8192 gigs of RAM, still no errors... It just won't pass that damn test, and will ALWAYS pass when I don't have anything in B1. Oh, one other thing - maybe one out of ten times it WILL pass, if I have a stick in B1. If anyone has any suggestions what might be going on here, I sure would appreciate it - I'm at my wit's end. Is the memory diagnostic tool in Vista just broken or something? Too sensitive, as in maybe the slot does have some minor flaw, but not enough to affect performance, just to fail the test? I also might add, Windows consistently awards me a 5.9 "score" in that stupid little rating thingy Vista has, if that makes a difference. Thanks! :) Last edited by ByteMe; September 4, 2008 at 05:04 PM. |
| ||||||
|
Well, 8 GB is alot unless your running something like a Tyan or a Supermicro mobo. And seconds, your ram may be the issue with the board. Third, take it to a PC shop and have run it through the tests. I know alot of store will diagnose issues if your gonna replace hardware
__________________ Antec 900 with Antec Quattro 850W Gigabyte Ga-x48-DS4 q6600 1333 FSB ( 333 x 9 ) stock air Kingston HyperX PC-8000 4 GB @ 1066 MHz Saphire Radeon 3870 X2 OC'ed to 880 MHZ 3X WD 160 GB 16 MB Cache ( RAID 5 ) some generic floppy drive! some LG black DVD writer of some sorts:P a real cheap MS ergonomic KB Microsoft Habu gaming mouse Screen 1: Samsung 2053 Screen 2: Panasonic Viera Plasma 42" |
| ||||||
|
I'd definitely increase the north bridge voltage. 4 DIMMs puts quite a bit of stress on the NB and usually requires a voltage bump. Though it's odd how it still fails with only 1 DIMM. I've never used Vista's memory diagnostic though, always used MemTest86+ though it doesn't have P45 support yet so you can't really rely on those results either. Maybe try updating your BIOS too if you haven't already.
__________________ Xeon Quad X3350 | 3.40GHz (Damn X38 chipset) ASUS Maximus Formula | BIOS RF0410 2GB Corsair Dominator | 850MHz [/font]4-4-4-12-2T 2GB OCZ Platinum Rev2 | 850MHz 4-4-4-12-2T HIS ICEQ4 Turbo HD4850 @ 700/1200 X-Fi XtremeMusic Antec TruePower Quattro 850W Antec 900 [ HEATWARE ] |
| |||
|
Thanks for the replies, guys. :) Yeah, I flashed the board to the latest BIOS, thought about a possible voltage problem but didn't want to try that until I researched it more. I really didn't know it was possible for one DIMM slot to fail, while the board still boots with only one RAM stick in the "bad" DIMM slot only - leads me to believe it's the result that's flawed somehow, not the DIMM slot or the RAM. But I'm gonna do some tests, try a few things and so on. I do 3D work, and I'd hate to have the RAM fail in the middle of a render, so I need to figure this out. Thanks again. :) |
| |||
|
I haven't adjusted anything, voltage-wsie in my BIOS and still have everything set to Automatic. I'm still getting the report that I have a memory error, but only from the Windows Memory Diagnostic Tool, and, once again, only sometimes, not all the time. SiSoft Sandra II doesn't seem to have a problem with it, and reports no errors. Additionally, I loaded a LightWave 3D scene I have that needs around 4 GB of RAM (lots of huge image maps), and it rendered with no errors, crashes, or hangs. I also loaded a 3D mesh into ZBrush and subdivided it to death, using up almost all the listed 8 GB of RAM for it, and had no problems. I would really like to think it's just an error in the diagnostic tool, but I'll still be trying other things meanwhile. I have to be sure. Although, for the life of me I can't see any sort of degradation, errors, or any other problems in actual use of that RAM in that DIMM slot. Again, all 4 of those RAM sticks test just fine in my other Vista machine - 64 bit Ultimate, with the same tool, but give no errors at all. |
| |||||||
|
Use either Memtest86+ - Advanced Memory Diagnostic Tool Memtest86+ or MemTest: RAM reliability tester for Windows for memtest from windows HCI memtest is high accurate. Since that one memory slot is causing the errors on its own, with 4 sticks, etc etc, and only randomly works / no works Everything your stating is pointing to bad memory slot, but on the other hand, 4 memory sticks , when all memory slots work alone, could also not work. This is when an increase to nb voltage can help. |
![]() |
| |
| Thread Tools | |
| Display Modes | |