Jump to content


  • Content Count

  • Joined

  • Last visited

About flashburn

  • Rank
    New Member
  1. Ok folks, recently, when I run a short hd tach bench on my Raid 0 boot array, I get what you see in the first image (hdt1, attached below), which can’t be right. When I run a long bench, things get somewhat better, as seen in the second image. Regardless, this still isn’t quite right. Does anyone have any ideas as to what might cause this (edit: i.e. these wild fluctuations) ? Hardware details are in my signature. Both Seagates are hooked up to the nvidia controller (ports 1 and 2) FYI. I’ve tried everything that came immediately to mind, such as defragmenting (seems obvious, but…) reinstalling drivers, reinstalling windows, benching after a clean boot, reseating the connections, etc. The original results (from when I installed the drives and from most of the intervening time) look much more like the third image--as one would expect. I’m stumped here, but I feel as though I must be missing something pretty obvious. Any ideas are greatly appreciated.
  2. Ok, so it's the ram. I've got a stick flaking out at 247 (probably earlier with a more extensive test) I'm a twit and thuda and chris get the nail-on-head award. I'm testing all my spare dimms right now: memtest is my new best friend. Moreover, now that I've gotten systematic, it looks like I may be able to up my CPU clock by a not insignificant margin. Silly, silly me... Thanks again to everyone.
  3. I'm agreeing with you :nod: By "point taken," I guess I meant "yeah, it's time to do this systematically" I've memtested, but not for overlong. I just had a bad feeling that I might have missed something silly that somehow broke the clock lock. I guess I'll check back after more extensive testing Thanks, everyone, for the replies.
  4. I know they exist, but don't know what they are. If I have to make changes like this without reformatting (which I try not to do), I use partition magic.
  5. I'm going to try the higher voltage as soon as a can afford to have the registry fried again...Though I've got restoring it down to a science, I get nervous that I might seriously break something and have to reinstall, which I don't have time to do for a couple of days I thought I’d gather info first, though, before I sit down for another war with the bios settings :nod: Edit: Point taken, thunda. I've been being lazy...however the fact that it's specific HD corruption stikes me as odd.
  6. Yeah, but at least if you can isolate the problem, you'll know how to fix it when a fix exists. Besides, you're really not missing much. Win64 seems snappier, but I suspect it's mostly gui tweaks to make it seem that way. most benchmarks have shown it to be roughly as fast with current software, id est, nothing dramatic.
  7. Hey folks. Everything’s been going well with the new system for the most part, but lately I’ve wanted to see if I could push my OC a bit (hitting this was a non issue, getting it prime stable has been a bit more complicated…), but anything above fsb240 (or htt or whatever we’re supposed to call it now) ends up shredding the registry. Usually I get about 1-3 reboots, with windows complaining at startup that it had to restore the registry, etc. Thereafter, I have to restore the registry by hand. This sounds suspiciously like the pci lock not kicking in, but the boot drive is always on SATA 3 or 4 (which I’m not sure applies to the NF4, but I figured it couldn’t hurt) and pcie is at 100. I’m about 95% sure I’ve had this happen both with and without sata spread spectrum set to on in the bios (which I’m also not sure applies with this drive). Moreover, I’ve reinstalled, reformatted and even zero filled, so this isn’t a data problem. The only thought I have left is to use the standard windows ide drivers instead of the nforce ones, which doesn’t seem like the best solution here. I feel as though I must be missing something really obvious, so any ideas would be appreciated. One other thought: could having the LDT or chipset voltage set too low cause this? I’ve got them at the minimum, since my policy is always to use only as much juice as I have to, and because I’m not really sure what the ‘safe’ (or at least reasonable) range of voltage is for these.
  8. I've read that even if you have the swap file on a drive with larger clusters, NT still reads it in 4k chunks, thus negating the whole deal. I don't know how much truth there is to this, though.
  9. I didn't need to install drivers during install b/c I'm not doing raid (yet..). You might try installing on a single drive and see if that fixes it. But, yes, I did in fact use the 6.66 drivers once I had everything booting, and it's still working w/o a problem.
  10. My advice (for what it's worth): don't bother. I've found that of the reasons to do this, more are bad than are good. And the chances of messing something up are nontrivial.
  11. If you were overclocked during your install, you might want to start over.
  12. Whay are you needing to change the cluster size?
  13. If you're overclocked, try running it at stock or at least kicking it back a notch. Win64 seems (for whatever reason) to not work at as high an overclock. I'm not sure if this is OS dependant or if this has something to do with the 64bit areas of the chip, but I've read about it and personally confirmed it. I'm stable enough in Win32 that it took me a good hour of scratching my head before this occurred to me. Edit: Overclocked while messing with win64, that is. You're obviously overclocked.
  14. Also, don't forget to... Disable unused IDE channels and any other devices you’re not using; clean out / defrag / rebuild the registry; disable any unused programs that are starting with windows; defrag…
  15. My sata drive does this too, the strange thing being that it didn't before I swapped the bulk of the components (mobo, processor, CPU). Same hard drive, same hotfixes, same version of windows...Oh, I bet it has something to do with this board having a sata II controller (my old one only had SATA1), which is itself odd, because I thought SATA1 supposedly had that as well. At any rate, this (http://www.microsoft.com/windowsxp/using/s.../kellyxp03.mspx) tells you how to get rid of it.
  • Create New...