Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by hlygrail

  1. 3 of the drives are external (USB2, Firewire). 2x200GB and 1x500GB SATA and a 300GB IDE are all internal currently. I may retire the 300GB to other duties if need be, though I doubt the 12-15W it pulls at the most is a make-or-break number. I'll probably order the X2 4200+ since it's a no-brainer price ($65) and the worst that happens is I'll have to shell out for a 700W PSU if it doesn't fly. Thanks for the head-check.
  2. I would have thought that the .sig at the time of the thread posting would be a permanent record, but if you're saying that if I change my sig, it changes on every past forum post I've made... I guess I'll buy that. Just want to know whether I really need to shell out for a new PSU if I go to dual-core. My quick math (and various PSU calculators on the Interweb) says I'd be pulling about 420-440W out of this one *after* moving to the dual-=core X2 4200+. That's higher than I had planned, but the GeForce 7950GT was an impulse buy -- when it got to $139, I couldn't resist. Anyone have a ballpark on what the exact power consumption difference between a Winchester 3200+ and a Toledo X2 4200+ is?
  3. Any thoughts on the new Seasonic S12II line of PSUs (see http://www.seasonicusa.com/S12II.htm)?? There's only one PCI-Express 6+2 (8-pin) connector, but for non-SLI rigs, it won't matter. I recognize 500W is toward (but not on) the lower end, but given what I'm running (stable and solid for nearly 3 yrs) on an S12-500, I thought I'd ask. See sig below...
  4. I'm planning on upgrading an older AMD Athlon64 3200+ (Winchester 90nm) to a dual-core X2 4200+ (sadly, the highest I can go on the Socket939). I'm capable enough to change the ACPI driver to SMP mode, but someone made mention of a power supply not being capable enough or getting along with an X2, so I'd like to understand that better. My rig details are in my sig below (and yes, I'm probably still running BIOS 302 from 2005.... it's been stable, and I purposely left it the @&$# alone -- I know I'll need to revisit the 3/10/2005 BIOS for an X2). Except for the recent GeForce 7950GT addition (was running a GeForce 6800) and the addition of a 500GB drive a month ago, everything has been rock-solid and unchanged for more than 2 years, running 24x7x365 on UPS-protected power. I'm having a hard time finding any legitimate basis for the latter part of this specific post: http://www.diy-street.com/forum/showthread...6081#post616081. Why wouldn't a 500W PSU (much less a 550W) be enough to drive the above system with only changing to an X2 dual-core CPU? As much as I truly believe that this DFI mobo won't run with anything below a good, solid 500W PSU w/ solid dual-12V rails, this still sounds like a fanboy statement to me. Can anyone back this statement up, or is it just fluff?
  5. I'm out of GigE NICs at the moment, but an Intel Pro/1000 GT is on the way from Intel ($35 shipped -- look for their Evaluation program!!). I gave away all my 10/100s. Once the Pro/1000 arrives, I'll slap that in and move off the onboard permanently (the 3Com/Marvell Yukon just can't hold a candle to an Intel... and all my other NICs are Intel Pro/1000 MT, including the one in the mail server). The server is a P3/550 running Win2K Server, with a 4GB SCSI boot disk and a 4x160GB RAID5 array on an Adaptec 2400A RAID controller. No SATA on the server end. (And yes, the server is aging -- was built w/ RAID5 back when RAID5 couldn't be had onboard...)
  6. I've gone through all the settings for speed/duplex, including 10 Mbit/Half duplex. Fails no matter what it's set to. That's the first thing I checked. The fact that the first frame isn't getting to the destination (0 hops through switch) but the rest do is just freaky, and no other protocols (that I can tell) seem to fail -- FTP, BitTorrent, SMB/CIFS, etc. In all the years (going on 15) of doing IT and tech support, this is one of the strangest things I've seen. The first packet literally just didn't get sent. I guess I'll clear the CMOS and test whether it's something there, but that's a giant leap, given that this box hasn't been rebooted (other than to troubleshoot this problem once it started) since April. In other words, no changes were made, it just started failing.
  7. Sorry, but that's not it. I never installed any of the nVidia firewall stuff. I'm behind a Cisco 1721 router, so it's not needed. And to make things more fun, if I do turn off Segmentation Offload in the nForce Vitesse controller, that NIC totally stops working. Default route vanishes from "route print," and it doesn't show up in the "ipconfig /all" output. Only way to get it back was to uninstall it, (reboot optional), rescan for hardware and let XP pick it up again. It's not ActiveArmor.
  8. Ok, so I've had this rig since about May and except for some fun w/ memory that I resolved by putting good stuff in, it's been rock solid.... until this week. (apologies in advance -- this is a long post, as the problem is very finite and I've done much troubleshooting on it.) I started having problems sending e-mail. (Client is Eudora, for what it's worth.) Some e-mails would go out, some wouldn't. I run my own mail server, and everything is GigE LAN-connected. I also couldn't send the same e-mails directly to my ISP (RoadRunner) mail servers, so I knew it wasn't the mail server. Other machines on the LAN *could* send the same e-mail, so I knew it was something with this box. Troubleshooting further, I found I could generally send short e-mails, but anything over a few lines wouldn't go. Given that I have access to the mail server, I could see the logs on the other end, and see that the session was being disconnected before the e-mail was finished. (WTF??) This one had me baffled, but I was determined. I installed Ethereal on both machines and started capturing SMTP traffic across both machines. What I found baffled me even further. Text versions of the trace files, edited for privacy, are attached for anyone who wants to play along with the home game. Python is the mail server, Zoot is the mail client. In short, the mail client sends the standard SMTP commands - MAIL FROM:, RCPT TO:, DATA, etc. and starts sending the mail body. This e-mail happens to span 2 Ethernet frames (TCP segments). From the client (Zoot) side, you can see the e-mail body get transferred, but from the mail server (Python) side, you don't see the first segment of the body. The second arrives, out of sequence, and the mail server starts sending a retransmit request (an ACK to the previous segment). The client starts doing a TCP retransmit on the segment, but it never appears to get to the mail server. Interesting .... Think this is my OS? Nope... as soon as I switched to the Marvell Yukon NIC (same IP address, same GigE port, same Cat5e cable), everything is perfectly fine. Switch back to the nForce Vitesse, it goes into the crapper again. I call this a serious bug in the nForce NIC drivers, as they should be handling the TCP sequence. Anyone wanna disagree with me ??? I haven't had time to do anything else (clear CMOS, re-test) because I'd really like to get to the bottom of this and wanted to collect more failure data. However, I did upgrade from the ForceWare 6.33 drivers I was running (stock from mobo CD) to the 6.69 drivers (upgraded all - NIC, bus, IDE, etc.), with same results. Another interesting tidbit -- this machine is seeding anywhere from 1-10 torrents at any given time, so it runs a lot of traffic through it all day. If I look at the TCP stats, I see more than 2% TCP retransmits. I'd expect somewhere between 0-1%, but 2% is a bit high. May be nothing to look at. GigE switch (SMC 8508) in between has been power-cycled, all machines rebooted/power cycled. Though switch supports it, Jumbo Frames is not being used. I think this NIC (the PHY part) is hosed... anyone have any thoughts??
  9. I had this same problem -- had to install the NIC driver off the CD that came with the mobo (probably v6.33 of the ForceWare drivers). Also, FYI, v6.69 seems to work as well. Haven't tried later versions.
  10. I have read the sticky. Remember, this isn't the first DFI rig I've built. Mine (in my sig) is rock-solid and I couldn't be more happy. Further, I haven't found a more reliable PSU than the Seasonic S12 line, and neither has Tom's Hardware -- it was one of the only ones to survive their 6-month torture testing. If you're suggesting that the Seasonic S12-430 is overrated, that's anecdotal unless you can show otherwise. I've seen or read nothing to back that claim up, but have seen many reviews that show it's not (e.g. Tom's Hardware) and is, in fact, one of the more robust in its power supply claims. Go look at http://www.xtremesystems.org/forums/showth...43&page=1&pp=25 if you think the DFI boards (and not just the DFI) are blameless in this instance.
  11. Not that it's worth arguing now that the DFI board is on its way back, but by all accounts, a 430W PSU should have been fine with this board (DFI Infinity + A64 3500 + a low-end Geforce 6600LE + 1GB PC3200 memory). DFI recommends at least 400W, and I gave it that and a bit more. Also, the rails on the Seasonic are the tightest I've ever seen in a PSU (even better than the S12-500 in my existing rig). If a *good* 430W PSU is underpowered, than DFI needs to change their advertising in a big way.
  12. Shenanigans is called. Another nForce4 board (Abit AN8) has no problem with this memory, I'm sad to say. Chalk up another DFI swing and miss. As much as I LOVE *my* rig (LanParty Ultra-D w/ PC4400 TwinX), I'll be hard-pressed to pick up another one until the memory issue is resolved. This much trouble with selecting memory just doesn't warrant the hassle. Thanks for the help, though.
  13. The PC4400 in my sig is my primary box (and is working FLAWLESSLY). I'm asking about memory for a new box I'm building using a DFI Infinity Ultra. Had I bought the TwinX1024-4400C25 for this box as well, I'm sure it would be working fine, but I thought the TwinX1024-3200PT was high enough on the "doesn't suck" charts to work with the Ultra Infinity board. Any other suggestions besides OCZ for the DFI Infinity Ultra mobo (not the Lanparty Ultra-D in my sig)?? .
  14. Went to 3.0v on the memory with a moderate (.1v) bump on CPU. Also tried hard-coding the 3-3-3-8 setting with everything else default (as suggested by the Corsair folks). Still no love... So, looks like the 3rd time I'll be RMA'ing memory on a DFI board. What CAN I use that's in the <$175 category for a matched 1GB pair (2x512MB) ?? 2.5-2-2-n would be nice, but as you can see, 3-3-3-8 will be fine for what I'll be doing with this box. Just needs to be in the <$175 range, with $150 even better (these were $140). Thanks!
  15. Searching for "corsair" will be like finding a needle in a haystack. Is there a specific thread? I tried bumping the memory voltage up to 2.8v with no improvement. Will try going to 3.0v tonight and also bump the CPU voltage as well. Also posted on Corsair's support page. Thanks for the help!
  16. Building a second DFI board (my sig is my box, this one will be for someone else): DFI Infinity nF4 Ultra (http://www.newegg.com/Product/Product.asp?...N82E16813136163) Corsair TWINX1024-3200PT (http://www.newegg.com/Product/Product.asp?...N82E16820145492) Seasonic S12-430, 430W PSU Flashed to newest BIOS (8/11/05 - CK8UD811.zip). Had some blue-screens while trying to install WinXP Pro+SP2. I'd planned on running memtest86+ anyway after XP, but this pushed me to reverse that. During memtest, I get errors in test #8: Failing Address Good Bad 0001c417f20 - 452.4MB 00000000 00a20000 0001da35c00 - 474.3MB ffffffff fffdffff 00032395c00 - 803.3MB ffffffff ffdfffff 000224d7f20 - 548.4MB 00000000 00a20000 ... etc. I recall getting similar errors on my DFI LanParty nF4 Ultra-D until I went *to* Corsair TwinX (TCCD in my case) from some Mushkin value memory. Did some research and found an interesting problem with the VTT circuit - bore you with the details here: http://www.xtremesystems.org/forums/showth...43&page=1&pp=25. Essentially, test 8 and 11 especially draw more current than the VTT can dish out on the DFI boards. Unsure if this also applies to my DFI Infinity, but I would assume it does. Should I really be concerned about these errors in test 8 ?? Will upping the memory voltage help ??
  17. What multiplier? My quick test at 2.5-4-3-8 at 280 x 9.0 didn't go so well. Rebooted at test 4 the first time, started logging errors the 2nd time.
  18. I swapped out my Mushkin 2x512 (non-value, but also not the highest end performance label) for a pair of Corsair TwinX1024-4400C25 memory (DDR550). All I can say is: Holy mother of ^&@#^%#@@# this stuff rocks! I mean, [email protected]^%%$ !!!! What a difference. I knew from testing with the Mushkin memory that this board was severely memory-bound (limited by the memory), so much so that I couldn't even MAKE the CPU go above 100°F, so I decided to trade way up to something I knew had been tested and proven by some of the folks here. As a new DFI customer, I'm impressed as he!! with this system (knowing what I'm coming up from on a primary system would only scare you, the reader, but let's just say it's several generations back). There were a few hurdles and quirks, but for the most part, I'm very satisfied. I've been running Memtest86+ now for just under 23 hours at 260x9, 2.5-4-3-8 on a 3200+ Winchester with not a single error. I know it still isn't even struggling to keep up, and I can only imagine how much farther it can really go. Speaking of that, how far can I really push this thing with these sticks?? I'm still running the memory at 2.5V even...(Oh, and I'm sure I'm on the 3/10 BIOS now as well.)
  19. I have the Seasonic S12-500 in my NF4 Ultra-D system now. I will tell you it is almost SILENT, as well as being more efficient than any of the others (except the S12-430) you listed above. Now, if Newegg is now selling it for $150, that's a crying shame. I got mine in late February for $129ish. (Oops... I lied. Just checked, and I did indeed shell out $139 for it.) Actually, I see that it's $152 with $.99 shipping now. Still a bit on the expensive side... they probably say the higher demand for it and went with it. Even though I own 3 different Seasonic PSUs now on various PIII, P4 and AMD64 systems (and they're all extremely quiet and right-on for voltage rails), I'd have second thoughts about hitting Add to Cart at anything over $140. http://www.newegg.com/app/viewProductDesc....-151-024&depa=0 FWIW, I bought the S12-500 because Newegg wasn't carrying the S12-430 then, but they are now @ $102.99 + $.99 shipping. Maybe you can drop back -- I know I will never draw 500W in this system, even with 2 HDDs, 2x512 RAM, GeForce 6800 (and a plan for a second).
  20. EXTREMELY quiet (usually silent), very efficient, lots of extras.
  21. They didn't have it a few weeks ago when I went shopping, but for $99, this is a much better deal than the Enermax PSUs you're looking at, both in noise and efficiency: http://www.newegg.com/app/viewProductDesc....-151-023&depa=1
  22. You mentioned you were locked into Enermax (don't know why you would be), but if you're open to others, I would suggest the Seasonic S12-500 (they also have a 430W model, but can't find it anywhere online in the US), for just a few bucks more. The major advantage is that the S12-500 is more efficient than the Enermax (costs less to run over time), and also has a thermally-controlled 120mm fan in it, so it's almost dead silent most of the time. I have 3 Seasonic PSUs now, and will never go back to 80mm PSU fans. Ever. http://www.newegg.com/app/viewProductDesc....-151-024&depa=1
  23. For the benefit of everyone -- where is the MAC address that should be entered? I couldn't find a sticker anywhere on the mobo, so I left it blank, and I definitely had issues with LAN1 port (nF4/Vitesse) until I uninstalled the 6.39 Ethernet drivers and installed only the 6.37 Ethernet drivers.
  24. The Marvell chipset sucks. There, I said it... It's the same one (newer flavor) that 3Com uses in their onboard GigE chipsets, and they suck. Find someone with an Asus P4P800-family board, and ask them if they ever see "Delayed Write Failed..." errors on their builtin GigE ports. Happens all the time, inexplicably. (It's a driver issue mainly, not just hardware, but it's aggravating. None of my Intel Pro 1000's ever do that, on the same switch.) I tested both the LAN1 (NF4 controlled) and LAN2 ports for speed, and the nForce4/Vitesse chip is definitely faster. I did not look at CPU utilization in any way. With this much overhead to spare (my board is not even a week old, and I can't BELIEVE how much overhead I still have to go in overclocking -- rock stable at Athlon 64 4000+ speeds on a A64 3200+), who cares about CPU utilization for GigE? You're talking probably a 1-5% difference, if that. Despite my evil slant towards the Marvell chips, my original thought was: "why should I have 2 sets of drivers installed, when just having the nForce drivers gives me GigE by itself?" That being the case, unless you have serious need to use 2 GigE ports on this mobo, I'd say definitely use the LAN1 (nForce/Vitesse) port. You'll be happier, regardless of whether you use the nForce firewall stuff.
  25. I think this is somewhat expected with some chipsets/memory configurations. Memtest86+ even mentions that on their site (sorry, I didn't save the link while digging for same issue last night) -- test 5 and 8 may generate errors, but doesn't necessarily mean the memory is bad. Google for something like "memtest86+ test 5 error" and you'll see Run Prime95 for extended periods (6-24 hours) in "use lots of RAM mode" (sorry, don't know what the radio option says off hand), and test your RAM+CPU stability to be certain. Prime95 will tell you of any errors, and you don't want ANY.
  • Create New...