P5K-E problem
in [Asus]
|
Prev: Dead A7N8X-X
Next: P5B Sound recording inop
From: OzarkMtbr on 14 Apr 2008 00:53 I just built my first system from scratch and am having trouble with the ASUS P5K-E. First my system specs are as follows: Core2Duo e8400 3.0GHz (45nm CPU) ASUS P5K-E (not a WiFi board) 2x Corsair DDR2 1Gb RAM modules 36Gb WD Raptor boot drive 500Gb WD Caviar data drive XFX Geforce 9600GT PCI-E I loaded Windows XP Pro and it worked fine with a few lockups the first few days. I put it in standby earlier without knowing that DVI displays will not wakeup. After restarting I kept getting a "Bad BIOS checksum" error. My BIOS version is 0906. I read somewhere that the RAM chips must both be in the yellow or both in the black slots. I had one in yellow and one in black. I moved one and it seems to be running stable now. Why would this fix the problem?
From: Paul on 14 Apr 2008 01:37 OzarkMtbr wrote: > I just built my first system from scratch and am having trouble with > the ASUS P5K-E. > > First my system specs are as follows: > > Core2Duo e8400 3.0GHz (45nm CPU) > ASUS P5K-E (not a WiFi board) > 2x Corsair DDR2 1Gb RAM modules > 36Gb WD Raptor boot drive > 500Gb WD Caviar data drive > XFX Geforce 9600GT PCI-E > > I loaded Windows XP Pro and it worked fine with a few lockups the > first few days. I put it in standby earlier without knowing that DVI > displays will not wakeup. After restarting I kept getting a "Bad BIOS > checksum" error. My BIOS version is 0906. I read somewhere that the > RAM chips must both be in the yellow or both in the black slots. I > had one in yellow and one in black. I moved one and it seems to be > running stable now. > > Why would this fix the problem? There is single channel and dual channel mode, for boards with four slots. ---A1--A2 -----------B1--B2 A1 and A2 are on one channel. B1 and B2 are on the second channel. A1+B1 matched, would be dual channel. A1+A2 would be a single channel case, with two loads on the same channel. Colors are used (at least sometimes), to make placing the sticks in a dual channel configuration, easier to figure out. Using two slots of the same color, means one slot is on one channel, and the other slot is on the second channel. Sticking both of the sticks on the same channel, in theory, is more loading. But really, with DDR2, there is no excuse for a difference in behavior. All possible combinations should be perfectly stable. I'd check with CPUZ, that in both cases, reasonable settings are being used for clock and timings. And check Vdimm in the BIOS - use more than 1.8V, as the memory may be specified for that. The spec sheet or web page might mention using up to 2.1V or so. It is possible, with a different BIOS loaded, that the memory setup issues will be handled better. You'd want to check the forums for the board, and see if there is a BIOS dependency. Bad BIOS checksum, *should* be well before a lot of other components on the motherboard, are enumerated and analysed. In the past, an example of a reason for it, was overclocking was causing the clock feeding the EEPROM, to be pushed past its acceptable speed. If a person followed advice, to reflash the board while it was in that state, then brickage was the result. So don't jump to conclusions based on the message on the screen, that you should immediately flash the BIOS. What you'd want to try first, is clearing the CMOS with the CLR RTC jumper. Generally, the manual will tell you to unplug and power off, before following the rest of the clear CMOS instructions. It is possible, if this is somehow related to clocking, that it could be fixed by clearing CMOS. A second possibility, is the BIOS is writing to itself. The BIOS has code to do that, because it uses that code to update the DMI/ESCD segments. Since the BIOS container has switched to SPI chips, I don't know whether exactly the same recipes are used, as in the past, or not. The poster here, mentions that bad BIOS checksum, happens after a BSOD at the OS level. I would not stick the "driver CD" in the drive, unless all other avenues have been explored first. http://vip.asus.com/forum/view.aspx?id=20080228210508359 There is a flashback here, to the old "legacy USB" BIOS setting interacting with memtest. I thought that caused a freeze though. http://vip.asus.com/forum/view.aspx?id=20080205043920812 More P5K-E posts are here. http://vip.asus.com/forum/topic.aspx?board_id=1&model=P5K-E&SLanguage=en-us If you want another memory test utility, there is this Microsoft one. http://oca.microsoft.com/en/windiag.asp Paul
From: Jungle Man on 14 Apr 2008 10:21 On Sun, 13 Apr 2008 21:53:58 -0700 (PDT), OzarkMtbr <ozarkmtbr(a)gmail.com> wrote: >I just built my first system from scratch and am having trouble with >the ASUS P5K-E. > >First my system specs are as follows: > >Core2Duo e8400 3.0GHz (45nm CPU) >ASUS P5K-E (not a WiFi board) >2x Corsair DDR2 1Gb RAM modules >36Gb WD Raptor boot drive >500Gb WD Caviar data drive >XFX Geforce 9600GT PCI-E > >I loaded Windows XP Pro and it worked fine with a few lockups the >first few days. I put it in standby earlier without knowing that DVI >displays will not wakeup. After restarting I kept getting a "Bad BIOS >checksum" error. My BIOS version is 0906. I read somewhere that the >RAM chips must both be in the yellow or both in the black slots. I >had one in yellow and one in black. I moved one and it seems to be >running stable now. > >Why would this fix the problem? With the many Intel chipsets, inluding the P35 chipset on the P5K series mortherboards, the memory modules must be installed in pairs and in the correct slots in order to work properly. There is much more technically detailed info on this if you search for it on the Web.
From: OzarkMtbr on 15 Apr 2008 09:25 On Apr 14, 12:37 am, Paul <nos...(a)needed.com> wrote: > OzarkMtbr wrote: > > I just built my first system from scratch and am having trouble with > > the ASUS P5K-E. > > > First my system specs are as follows: > > > Core2Duo e8400 3.0GHz (45nm CPU) > > ASUS P5K-E (not a WiFi board) > > 2x Corsair DDR2 1Gb RAM modules > > 36Gb WD Raptor boot drive > > 500Gb WD Caviar data drive > > XFX Geforce 9600GT PCI-E > > > I loaded Windows XP Pro and it worked fine with a few lockups the > > first few days. I put it in standby earlier without knowing that DVI > > displays will not wakeup. After restarting I kept getting a "Bad BIOS > > checksum" error. My BIOS version is 0906. I read somewhere that the > > RAM chips must both be in the yellow or both in the black slots. I > > had one in yellow and one in black. I moved one and it seems to be > > running stable now. > > > Why would this fix the problem? > > There is single channel and dual channel mode, for boards > with four slots. > > ---A1--A2 > > -----------B1--B2 > > A1 and A2 are on one channel. B1 and B2 are on the second channel. > > A1+B1 matched, would be dual channel. A1+A2 would be a single channel > case, with two loads on the same channel. > > Colors are used (at least sometimes), to make placing the sticks > in a dual channel configuration, easier to figure out. Using two > slots of the same color, means one slot is on one channel, and > the other slot is on the second channel. > > Sticking both of the sticks on the same channel, in theory, is more > loading. > > But really, with DDR2, there is no excuse for a difference in > behavior. All possible combinations should be perfectly stable. > > I'd check with CPUZ, that in both cases, reasonable settings are > being used for clock and timings. And check Vdimm in the BIOS - use > more than 1.8V, as the memory may be specified for that. The spec sheet > or web page might mention using up to 2.1V or so. > > It is possible, with a different BIOS loaded, that the memory setup > issues will be handled better. You'd want to check the forums for > the board, and see if there is a BIOS dependency. > > Bad BIOS checksum, *should* be well before a lot of other components > on the motherboard, are enumerated and analysed. In the past, an > example of a reason for it, was overclocking was causing the clock > feeding the EEPROM, to be pushed past its acceptable speed. If a > person followed advice, to reflash the board while it was in that > state, then brickage was the result. So don't jump to conclusions > based on the message on the screen, that you should immediately > flash the BIOS. What you'd want to try first, is clearing the CMOS > with the CLR RTC jumper. Generally, the manual will tell you to unplug and > power off, before following the rest of the clear CMOS instructions. It > is possible, if this is somehow related to clocking, that it could be > fixed by clearing CMOS. > > A second possibility, is the BIOS is writing to itself. The BIOS has > code to do that, because it uses that code to update the DMI/ESCD segments. > Since the BIOS container has switched to SPI chips, I don't know > whether exactly the same recipes are used, as in the past, or not. > > The poster here, mentions that bad BIOS checksum, happens after > a BSOD at the OS level. I would not stick the "driver CD" in the > drive, unless all other avenues have been explored first. > > http://vip.asus.com/forum/view.aspx?id=20080228210508359 > > There is a flashback here, to the old "legacy USB" BIOS setting > interacting with memtest. I thought that caused a freeze though. > > http://vip.asus.com/forum/view.aspx?id=20080205043920812 > > More P5K-E posts are here. > > http://vip.asus.com/forum/topic.aspx?board_id=1&model=P5K-E&SLanguage... > > If you want another memory test utility, there is this Microsoft one. > > http://oca.microsoft.com/en/windiag.asp > > Paul Thanks for the information and links. Since I moved the RAM to the same color slots I have no longer had the boot error. The system is still experiencing some flaky stability. If I leave it setting for awhile (hour or so) when I get back it does not wake up. I have all the power modes off so it will not standby or hibernate. Usually the display tries to wake up with the press of the enter button, but then mouse and keyboard go dead. The reboot button does not help. I have to hold the power off button until it shuts down. Then it will restart and all is working again. I have not had much time to work with it since I made my first post.
From: Paul on 19 Apr 2008 03:21 OzarkMtbr wrote: > > > Thanks for the information and links. Since I moved the RAM to the > same color slots I have no longer had the boot error. The system is > still experiencing some flaky stability. If I leave it setting for > awhile (hour or so) when I get back it does not wake up. I have all > the power modes off so it will not standby or hibernate. Usually the > display tries to wake up with the press of the enter button, but then > mouse and keyboard go dead. The reboot button does not help. I have > to hold the power off button until it shuts down. Then it will > restart and all is working again. I have not had much time to work > with it since I made my first post. > The poster in this thread has symptoms that are partially similar to yours. You can try adjusting Vdimm, to see if the memory stability can be improved. Or RMA the board under warranty. Or try a BIOS update. http://vip.asus.com/forum/view.aspx?id=20071224145548546&board_id=1&model=P5K-E&page=1&SLanguage=en-us Paul
|
Pages: 1 Prev: Dead A7N8X-X Next: P5B Sound recording inop |