Jump to content


  • Content Count

  • Joined

  • Last visited

About mikesm

  • Rank
    New Member
  1. Don't knock software raid 5 as a whole. Linux software raid5 is quite reliable. On the other hand, windows software raid 5 blows chunks. If you want Raid 5 on windows, use a real hardware controller, preferably on the PCI-E bus. Thanks, mike
  2. Did you try moving the safe boot jumper for a few secs and then trying to boot? This has worked for me in the past. Thanks, Mike
  3. How do I remove data in the CMOS reloaded banks? There is no command to clear them? I could load optimized defaults, and then write that into every bank, but I still have the same issue after loading optimized defaults and booting from there. Some bad data is here somewhere messing things up. I've never seen a motherboard remember anything other than it's base rom config after being disconnected from power and removing the battery... Can someone from DFI confirm this behavior with the CMOS memory banks? If I reload the BIOS, will that help clear them? I suspect not from my previous experience, but I can't tell for sure. Thanks, Mike
  4. I did this. Initially it wouldn't go into the BIOS, so I enabled the safe boot jumper for a few secs, and then reset it. It came back up, but this didn't fix the issue with the 1394 controller and the ethernet. I did notice a brief message on the first BIOS load about cmos bank 3. I repeated the procedure, and before loading defaults, I asked it to load from bank 3 - lo and behold, it still held the config I had saved a month ago! Do the CMOS banks not get cleared by removing the battery and doing a CMOS clear?? There is no command in the BIOS to clear them, and I am wondering if somehow the BIOS is fetching bad data from saved configs in CMOS. How do I clear things completely and go back to all defaults, including the CMOS banks?? thanks, mike
  5. PSU looks fine voltage wise. I haven't ever seen such an odd failure before. Thanks, Mike
  6. Ok, I took out the fusion card, removed it's drivers, and renabled the 1394 controller and the the yukon ethernet controller. I completely removed the fusion software, drivers, reg entries and all. No matter what I do in terms of dealing with the wizard, neither of these devices works. The 1394 controller finds the driver, but can't start. The Yukon chip driver (already installed) can't be found by windows, and when I force it to load the driver by manually selecting the driver, the device won't start (Yellow !) in device manager . What's the deal here? How come these devices don't work? Am I missing something? Thanks, Mike
  7. I tried this. The Fusion card still won't load drivers. I have the IRDA port, the 1394 and the Yukon controller all switched off in the BIOS. Any other ideas as to what's going wrong here? MSINFO32 doesn't show the Fusion card's "multimedia controller" or "multimedia video controller" as being assigned any IRQ's. Thanks, Mike
  8. I had tried this before without sucess, but will try it again. Thanks, mike
  9. Ok, I'm back with some more data on this issue. I did the CMOS clear, and that didn't help the problem. So I went and reinstalled windows, completely clean configuration, then added the latest Nforce4 chipset drivers. I still have the same issue! No drivers are installed for either the 1394 controller, Yukon ethernet, or Fusion. I do notice on boot up the OHCI 1394 controller, the Yukon ether and the Fusion HDTV card all seem to be assigned IRQ 3 - this info is from the BIOS during the boot sequence. I am curious if there is some problem related to IRQ assignment. Any ideas as if this could cause a driver installation problem? And is there any way to change the IRQ assignements? It appears everything hung off the PCI bus is being assigned the same IRQ slot. BTW, I have tried loading Windows XP x64, that made no difference, except that the 1394 controller seemed to install during the base OS install, but then after the Nforce drivers were installed, it reverted to the same condition - yellow !'s for all PCI devices. I tried installing drivers for 32 bit XP before installeding the Nforce drivers, but that didn't change things either. No devices installed and came up. Any ideas? Thanks, Mike
  10. I did not delete the drivers at that time, but since have done so using the hauppauge hcwclear utiility. I have not tried doing a a CMOS clear. I will try and do that and see if it makes any difference. Why should that be required if I loaded optimized defaults? I'm not sure a repair install will be any less disruptive than a full reinstall - it screws up all the application state on the system. Thanks, Mike
  11. Hi. I am dealing with a very odd problem with the PCI bus in my main PC, which has a DFI NF4 Ultra-D in it. The issue was flagged this week when I attempted to install a DVICO Fusion HDTV5 RT Lite QAM tuner card in my main PC. The system upon reboot registered two new devices, "Multimedia Controller" and "Multimedia Video Controller", and triggered the add hardware wizard to install drivers for the devices (they are not included in the normal windows XP SP2 driver set). No matter what I did, Windows would not recognize any of the driver software as appropriate for the hardware. Forcing the wizard to use the appropriate driver didn't work either. It just showed up with a yellow balloon with an ! in it. Working through support forums and through the vendor, it became clear this was an odd problem people hadn't seen before, making folks believe it was a hardware problem with the card, and to RMA it. Everyone blames the hardware when they see something odd. But before doing that, I installed it in a different machine, and in that system it installed fine and the drivers were loaded as it was supposed to work (also a AMD CPU with XP/SP2, but not DFI). So I tried going back to the main PC, still no go. At this point I noticed the system had two other PCI devices that were not in use, the VIA OHCI 1394 controller, and the Marvell Yukon GB Ethernet controller. I had always had these devices disabled since I didn't use them, but just to verify the PCI interconnect was working, I enabled both, triggering the hardware install wizard. The Via 1394 controller's drivers are supposed to be in XP/SP2, but again, no appropriate match was found. The Marvell controller needed a special driver download, but it like the 1394 controller would not recognize a match. When I selected the proper driver manuall for both devices and overrode XP's objections that these were not a match for the hardware, both devices loaded the drivers OK, but both also failed to start. It appears that nothing on this PCI bus is being recognized by the OS properly. Note this bus used to work at one point, because I had a old Hauppauge WinTV controller in the system about a year ago, but I had removed it some month ago. Thinking it might be an obscure BIOS config problem, I reset the BIOS settings to default on the system, and rebooted. No effect. I loaded the old 704beta BIOS for the system that it used to run, cleared settings and tried it again. Also no go. I have also uninstalled the PCI system bridge device and reinstalled it, as well as uninstalling the Nvida forceware drivers for the MB and reinstalled all the devices, as well as loaded the latest forceware drivers. Still no effect. PCI-E devices work fine of course. The next step would be to reformat and reinstall windows, but before doing that I wanted to solicit any possible fixes from folks here on the forum. I had reverted the system to it's config of 1 week ago (before I started any work with the Fusion card) (I have full backups available), but this did not seem to fix the problem. At this point, I'm out of ideas. Anyone here seen this before? Thanks, Mike
  12. I don't havea problem with fluctuations, but I get a universally high CPU temp after resuming from S3 suspend. Does the new BIOS fix this? Thanks, Mike
  13. I just got mine. Do the plastic pins fit exactly into the holes for the original cooler? The reason I ask is that I am thinking about not removing the mobo from my case, and just cutting the existing cooler's mounting pins off. This should allow me to install the new cooler witout needing access to the bottom of the board, but once I cut the pins to the existing fan I won't be able to reinstall it. Thanks, Mike
  14. When I submitted a trouble request to DFI on this, they acknowledged the problem and said it would take a BIOS fix to correct. So I think it's on their offciial bug list. I would be interested in seeing DFI confirm this on this list though. If you see this problem and havn't submitted a ticket with them, you should do so as it will increase their visibility of the scope of the problem. Thanks, Mike PS I don't recall this problem with the old 3/10 release of the BIOS, but it's certainly there with the 6/23 official release.
  15. I don't understand why this is so hard to solve. We need new BIOS's to support New CPU types occaisionally anyway. DFI, if you are listening, please fix this! Thanks, Mike
  • Create New...