Jump to content

LanceDiamond

Members
  • Content Count

    91
  • Joined

  • Last visited

Everything posted by LanceDiamond

  1. 3 works very nicely - have a look at my post in the RAID thread for how 3 x Hitach 250gb in RAID0 work. They're VERY fast. I've got assorted external drives for backup - and if you do RAID0, you really DO want to backup your important data...
  2. I was having odd problems with my BenQ 1640 drive and ASPI stuff - got it fixed. First, the best place I've found for anything optical drive related is club.cdfreaks.com - they have forums for every brand of drive and LOTS of people posting and helping. My problem was I'd inadvertantly installed the NF4 IDE PATA IDE drivers. I went into Device Manager/IDE controllers section. Right click on anything that is NOT SATA related (you want to keep the NVidia SATA stuff) and select Update Driver. Choose "No" (don't search Windows Update, "...specific location..." (Advanced), "Don't Search". Uncheck "Show Compatible" and go find Microsoft/Primary IDE Channel. This will get you the stock Windows XP IDE driver. Restart, it'll redetect some stuff and you may be fixed. I'd be careful doing this if I had anything but optical drives on the IDE - if your boot drive is on IDE and you go messing with it, you could wind up not booting. But if only optical drives, worst case put it back if it doesn't work. This fixed my weird problems with my BenQ & ASPI issues.
  3. To add to what someone said about LCD cleaning, I was using non-amonia window cleaner on my LCDs too. Eventually I noticed they'd been permanently marked doing this and I was sorta annoyed. Then a friend who has two 21.3 Samsungs told me to use just mild soap & water. Put a small amount of regular liquid dish soap in a small glass and fill it with water. Dip a soft towel in the mix, get it damp and gently clean with that. To my amazement, it took what I thought were permanent marks off my 20 inch Dell! My 24 inch has never been cleaned with anything but soap and water. I agree, you need a nice desk now! Edit: ok nevermind about the desk then lol - shoulda read your last comment there duh...Ikea is l33t
  4. I'm very interested in your results with your new RAID card too - you going to post HDTach & ATTO info in that Off Topic thread? That'd be a great place for it, so it'd be easy to compare to all those other configs... Even if it's overkill, it's fun
  5. As he asked - how do you know 275 on that ram with those fairly agressive timings is working? i used memtest in BIOS and thought everything was fine - then I used this tester and it quickly found problems memtest in BIOS hadn't: http://hcidesign.com/memtest/
  6. How do you know this exactly? If it's from reading this review, that was those particular sticks in the review that got to 295 @ 2.8v. I have the same stuff and I can't get to 295 at all - I can get to 290 @ 2.9v and run memtest from BIOS, but Windows memtest still found problems...
  7. Anything more than what is on the thread on the Creative site on this lately? I have an X-Fi Music and it's the only thing I haven't put into my new rig yet - and I am thinking I may return the X-Fi tomorrow and just use onboard sound for now. I did read through the thread on Creative's site and posted there asking about the down side to the firmware fix. It sounds very much like Creative and DFI are doing some finger pointing on this. Sounds a LOT like a bad driver if the system reboots repeatedly with automatic reboot enabled. If it's a driver thing, that sounds like a Creative problem but maybe they're having a hard time figuring out what is wrong or something... In any case, I am hesitant to risk de-stablizing my nice new speedy AMD for a sound card!
  8. The XP120 is working great for me - no fit issues here. It does block the end PCI-E slot with heat pipes - but I don't have any idea what I'd ever use that slot for. It doesn't interfere with either of the slots for an SLI setup. I managed to get memory out of the yellow slot and into the orange slot - the one right next to the CPU which is WAY up under the XP120 - without removing the HS. Just a little patience and dexterity and it'll go. The hard part was getting it pushed down enough to snap the clip - there is about 1/4 inch of the DIMM peeking out on the back end of the module. What I did was take a slot cover from the back, turn it upside down and then GENTLY push down the memory. I have not had any experience with anything other than the XP120, but I'm very happy with it. Idle temp is 5-10f above room temp (31c right now) and load temp is about 45c at 2.6ghz @ 1.5v.
  9. Ahh ok hehe if we're up to 120mm fans now I have to note the one I accidentally wound up with! When I placed my Newegg order, I noticed this in my cart: http://www.newegg.com/Product/Product.asp?...N82E16811999122 I figured what the hell, I'm spending $2k and I can't for the life of me remember when or why I added that to my cart but I'll get it. Well it's on my CPU and the Panaflo is on the shelf! I like this fan because it has a slot thing with a knob - and of course knobs are good. You can put it in the little PCI-E slot that gets blocked by the heat pipe of the XP120 - a slot I don't think anyone will ever use. Then you have a knob on the back to adjust CPU fan speed. You can adjust from about 1250RPM - 2700RPM - I have mine fairly low but this P180 case has awesome airflow so theres a draft through my PC all the time anyway. Turned up it is loud but I assume it's moving some air too - specs on Newegg say 93.7CFM max and I don't doubt it - sounds a little like a leaf blower is in my case when it's full speed. At low speed, it blends in with the other stuff or maybe I can't even hear it, don't know...
  10. 2700 is about where I got to - beyond 2700, even 2708 or so, would eventually show problems - reboot, lock, bluescreen - after a long prime run - once I think 6+ hours til a hard lock running prime95. I went as high at 1.7v and found no luck at 2750 so I'm at 2600 @ 1.5v and quite happy, stable and cool (45c typical load temp.)
  11. Very good point, I wasn't thinking of old Cyrix - though I think they did wind up going out of business and AMD is alive and well - heck they even make money now! Apparently AMD's execution of the idea was a little better. There isn't any denying Intel has more server business than AMD but Sun, IBM (Lenovo get their servers too?) and HP all make AMD servers - I wouldn't call that "hardly any." Granted Dell may ship more servers than those guys - I dunno. I doubt Dell will ever use AMD - I have a whole conspiracy theory about that lol. If nothing else AMD has lit a fire under Intel and kept us from having Itanium rammed down our throats - so far.
  12. Some time ago, AMD and Intel each had their own ideas about how to improve performance. Intel decided faster clock speed was the way to increase performance. AMD had other ideas - I can't really describe how AMD did it other than to say they figured out clock speed did not have to be connected 1:1 to performance. Turns out AMD was right - and Intel never hit the clock speeds I think they had envisioned for Prescott cores. Intel's new strategy is to do a Pentium M type thing for desktops. If you've used a Pentium M laptop, you'll know that even though it may have a low clockspeed (I had one with 1.3ghz) they work great and the low clock has nothing to do with how well they perform. Moral of the story: Don't worry so much about clock speed. When AMD first did their "clock speed isn't important" thing, I didn't beleive it. I was wrong - at least I had good company (Intel) hehe... Edit: Should also mention, just because it entertains me, my 4400 even running stock speeds, smokes my friends P4 3.2 Northwood. Running at 2.6 it just pisses him off badly. lol
  13. If you look at my post, you'll see I did that - I used the same settings others used and in fact what I mentioned, what you quoted me on, ARE those settings. My question, which I think I answered myself anyway, was why with the default settings I got such insane read figures. Cache is what I am assuming. Anyhoo, thanks!
  14. 4400 CCBWE 0537 stable at 2600mhz (260 x 10 @ 1.5v - BIOS shows 1.45v) - idle temp 28-30, load temps 44-47 (47 I've seen once for about a second - 44/45 is normal) I can get 2680mhz but to do so, I have to go to 1.6v and my load temps go up to 55-56. I can't seem to get anywhere significantly above 2700mhz even going as high as 1.7v. I'm very happy with 2600 @ 1.5v @ 45c.
  15. I think at least part, if not much of the problem with the 3114 on the mobo is that it has no cache. Any decent RAID SATA add-in card will have some cache and should perform much better than the 3114. Even a good add-in PCI card might not compare to the performance of SATA2 RAID0 on the NVidia onboard ports because the NVidia onboard ports don't have to go over the PCI bus, the 3114 and any add-in PCI card is on PCI. If you get a good PCI-E RAID card, then RAID5 performance could be ok - maybe even come close to the NVidia onboard SATA2. Now I'm curious about this, I may have to see if I can find such a controller! RAID5 isn't inherently bad for performance. I had use of a Dell server with RAID5 in it some time ago - it was VERY fast. Granted they were SCSI320 15k RPM drives but still, with the RAID5 it was very very very fast (never ran any benches unfortunately.) When one of the drives failed, the performance got terrible but once the drive was replaced and array rebuilt, was fast again. Edit: Ok...for $400 you can have reasonable RAID5 - maybe: http://www.pc-pitstop.com/sata_raid_controllers/arc1210.asp This card will do x8 PCI-E, so you'd flip the SLI jumpers, put the RAID card in one slot and your video in the other. Not even sure if that'd work - someone drop $400 and let us know?
  16. By switching the jumpers, both PCI-E slots become x8 slots. If you don't switch, the top slot is x16, the other one is x2. On the new NVidia NForce4 chipset that Dell is selling in their XPS line right now (with Intel chips...jackasses that they are), both slots are x16 or at least can be x16 in SLI mode.
  17. Actually I have an Antec Tricool in my duct - I took the duct itself apart while I had it outside the case, placed the fan INSIDE the duct and put it back together. I left the speed adjustment inside the duct so I can't reach it (set to medium speed), and ran the power wire out the main opening where the duct plugs into the back part on the case - if you flatten the wire down and plug the duct into the back part, the wire can fit through there. Granted it's not sucking as much air in as it could because the back of the fan is about 1/4 inch from the back wall of the duct - but some sucking is better than no sucking at all.
  18. After 6 days of building, burning in, tweaking and benchmarking my new system, I'm stable! My system is working well too. Right now I'm running a RAID0 array of 3 x Hitachi T7K250 (250gb SATAII drives). Default cluster size/stripe size - just booted from SP2 CD and installed XP SP2, then updates, drivers and assorted other benchmarking stuff. No defrag needed as I have yet to delete anything lol. My HDTach looks good: But in the ATTO bench, I don't understand the results - if I run it with default settings, I get this result: If I change to 32mb "Total Length" and "Neither" then I get this: I do think my 2nd result, where I change from the defaults, is more "real" because I don't beleive I'm getting 650mb/sec reads. Any idea why ATTO would show me that on the default settings? Ahh, actually I may have my own answer...Caching - 32mb "Total Length" gets me results where caching doesn't skew things maybe?
  19. If I understand correctly, OCZ is offering to swap the memory - and you have some tiling and grouting to do? If so, I'd say if you see a HINT of any further problems - without going to 3.0v - RMA it. Actually even if you don't see problems but don't get the chance to test much due to grouting, RMA it. RMA it while you have their attention and they're offering, don't delay more than a couple days on that if you can get a memory swap. If you get the memory swap and go back with same settings and see same behavior, well you eliminated the memory as a problem anyway, right? And I'm sure OCZ can test that memory with some fairly exhastive test to see if it's not in spec or something.
  20. Hey another thing to try, based on your description of the memory problem. http://www.hcidesign.com/memtest/ this is a Windows memory tester. I say try this because I passed hours of BIOS memtest86 at 282FSB on my memory - but failed in minutes on this thing. I had problems. Way I set mine up was to run 2 instances of the app once installed, set each to test half the memory, click ok a few times and in < 30min it can test 1800mb of memory (I set each to test 900mb.) I was able to find that I had problems at 282 but did NOT have problems at 260 with this test, pretty quickly. I then prime95'd for 21 hours, no problem. You might try this test with just one stick in, then swap and same test same setup with the other stick. If it fails on one but not the other, bad stick of memory. If it fails on both, memory timings or both sticks bad (I can't see that) or something else MB related maybe... In any case, this is a fairly short test 20min a stick or less I think.
  21. You *may* have to kick the OS to get it to load the multi-proc HAL - else it may only notice one CPU instead of two. Installing and removing the AMD ACPI driver does this. If you do install the AMD ACPI thing, I do recommend you subsequently remove it as it's really not needed unless you want the OS to be able to slow your processor down to save power and such. I swear there is some Microsoft KB someplace about MP versus SP HALs but I only find stuff pertaining to Windows 2000. Maybe I'm confused. Some folks on here have said they had to install the "AMD X2 driver" which is the AMD ACPI thing that I mention. If you do need it, you should at least know why you need it and that you don't have to keep it - you just need whatever it causes to happen, to happen.
  22. Same board & eVGA 7800 GTX as you, my chipset temp bounces around 48-50s all the time. The GTX card basically laying across the NB fan really hinders airflow. As long as it stays in a range and doesn't cause the rest of your case to start cooking - in other words, as long as you have airflow pulling that heat off someplace - I think you're fine. In my case, it has a blower thing that goes in over the video card with a spot for a fan. I placed an 80mm fan in there - so it sucks in fresh air and blows it right across the GPU and the NB chip too. That did help a lot. If you can get some direct air blowing past it'll help yours...
  23. I read someplace that the way OCZ rates their PSUs, 520 is probably like 700 from some cheap brand. That may be the case with all the A-list PSUs, dunno. I am quite happy with the OCZ 600w so far - in my case I think I could have gone for the 520w and been very safe, but I wanted to make sure I had as much expansion potential as possible. I may go SLI, add HDD, etc. I have hooked an HP benchtop DMM to my OCZ that reads to 4 decimal places - I found that the BIOS way overstates the voltages. My PSU is outputting (per the DMM) 12.0076v, 5.0051v and 3.35 (couldn't get a probe stuck in the back of the ATX connector so I could adjust while watching the meter lol.) The BIOS shows 11.64v, 4.91v, 3.28v and they jump around on the BIOS/Everest display. But with the meter attached I can see that the OCZ is rock stable, fluctuating less than .01v no matter what I do. Impressive! If you do invest in a nice PSU with adjustable rails, DO adjust them with a DMM - at least the 12/5v rails - they're super easy to do, just stick a problem in a free molex, adjust close as you can.
  24. Reason for that driver causing XP to recognize X2 would be that it must kick the OS into re-recognizing the CPU and getting the MP (multi-processor) HAL in place. There is a Microsoft KB article someplace about how to do that manually, may involve device manager or something like that. But apparently installing the ACPI stuff kicks the OS too, good to know I guess. 99% chance you remove that ACPI driver - and XP will still recognize you have 2 processors...
  25. The "AMD dual core driver" that you guys are talking about, is that the one here: http://www.amd.com/us-en/assets/content_ty...ties/amdcpu.exe Because if so, this shouldn't be required for anything, ever. It's an ACPI driver that lets the OS monkey with the power requirements of the CPU, shut down a core maybe, etc. Unless you're plugged into a battery or you have horrible heat problems, I can't imagine why you'd want ACPI even enabled on any non-laptop. I paid full price for this CPU and damnit, I want it working full time! This driver would have nothing to do with Windows XP seeing 2 CPUs and shouldn't have anything to do with stability unless it just happens that when ACPI kicks in and slows your processor down it becomes stable at the slower speed. If there is another driver y'all are talking about, please do correct me!
×
×
  • Create New...