Jump to content

canadianpsycho

Members
  • Content Count

    31
  • Joined

  • Last visited

About canadianpsycho

  • Rank
    New Member
  1. Thanks for this... I'm going to try out an increase in the voltages then. I'm currently stable at 2850 at around 1.45vcore... Can't get any higher with more vcore. Maybe this will help. Will be a pleasant surprise if it does.
  2. Thanks for the advice guys, I am seriously considering going back to 623. It just seems weird to me.
  3. I'm pretty positive 3.0GHz on air is an exception rather then the rule. I've got the same stepping chip and it was stable at 2.9 with BIOS 623 and is now testing at 2.85 with 704. You pretty much have to have phenominal cooling to consider 3.0 GHz.
  4. This PC is giving me fits. I managed to Prime95 8 hours overnight without issues using the 9:10 divider (2.8 GHZ, DDR504) . Still cannot use the 1:2 divider without BSODs/crashes irregardeless of BIOS settings/core or mem voltages/clearing CMOS. I'm not sure if its the 704 BIOS or the Opteron itself being really picky. I could use the 1:2 divider up until 270ish.
  5. I'm convinced the Opterons are divider finicky. I can't use a 1:2 divider at all once I take my CPU past 280 HTT. However I primed 8 hours overnight last night using the 9:10 divider. 1:1 doesn't work, but DDR560 isn't guaranteed with my RAM anyways even though a couple reviews I read with RAM using UCCC chips have gotten higher (I'm thinking its the CPU's memory controller anyways). I'm finding Opteron OCing much more frustrating then Venice OCing. Honestly, I don't know what to say... I'm using the same BIOS and had to back off my previous stable OC from 2.9 to 2.8 (testing 2.85 now) and as previously mentioned the dividers are a nightmare - no matter how many times I reset the CMOS, and even using OCZ's recommended timings for my SLI-DR (non Expert) I cannot get certain dividers to load or stay stable in XP.
  6. Alright. So my 146 was stable at 2.9 GHz with the 623-3 official BIOS. I picked up a 2GB Kit of OCZ Gold and one of the sticks needed an RMA. When I switched them up I decided to try out the 704 BIOS and see how it would go. I'm about 3 hours prime stable running a 9:10 divider for DDR527... I tried dropping the timings from 3-4-3-8 to 3-4-4-8 and then tried upping the Vdimm aftar a failed Prime run. Well that didn't work. So I figure maybe my 146 doesn't want to run 2.9 anymore. I decide to drop the divider to 1:2 so I can stress simply the CPU - but now I can't even stay in XP without a PFN_LIST_CORRUPT BSOD or getting the "Windows has recovered from a serious error" message. So what gives? 3 hours stable at DDR527 but can't even load windows at DDR290? Am I doing something wrong?
  7. To save yourself some time you can run SuperPI 16M or 32M test... Takes around 20-40 mins and if you get an error you know your OC isn't good enough. Saves you from priming every 5 MHz of the HTT.
  8. My 146 I'm testing atm does around 2.9 GHz with a XP-120 and Panaflo (I think Low) at 36° idle 47° load. This is with 1.52BIOS/1.50MBM+CPU-Z Vcore.
  9. I guess I should have elaborated further: Does anyone know first hand, or have any info supporting one side of the argument or the other? I am aware of the fact that increased voltages typically raise temps.
  10. Just make sure it's the right chips and you should be fine. OCZ has great tech support, you can email them or hit their tech support forums to get a definite answer. Also, at 3.2V you may want to consider actively cooling the sticks if they are the VX chips. At those timings, you shouldn't even need 3.2... That high Vdimm is ususally reserved for running 2-2-2-5 - extremely tight timings. You're not even close to those timings atm. EDIT: Also personally I think 55°C is a little high. My 146 with a XP-120 and panaflo hits 47°C load.
  11. Woah, 3.2V Vdimm is high unless running VX or UTT RAM. If you are indeed running generic value RAM I'd drop it to 2.8-2.9 or you'll burn those sticks out.
  12. So I've read the guide, it implies upping the LDT voltage and Chipset voltage shouldn't do much for your OC. Yet I've read a couple threads, and a few people suggest upping the LDT and Chipset voltage at least 0.1V. Does this accomplish anything? What are the benefits/hazards or any other effects of upping it to certain levels? Thanks.
  13. I'm pretty sure I might be in the same boat... I was fine at 2.9 GHz with my Opty 146 with 1.225 * 126% for about 1.51V in BIOS using the 623-3 BIOS Used the same settings with a newly flashed 704 mod BIOS and I get a bunch of PFN_LIST_CORRUPT messages when trying to get into windows.
  14. Just to summarize the things I've tried so far (after a fresh XP install): CPU + RAM at normal OC - crash CPU + RAM OC lowered by both 50 MHz and 100MHz - crash CPU at OC, RAM using a divider - crash CPU + RAM OC in Memtest - no errors CPU + RAM stock - stable So basically I'm thinking the Venice just doesn't like the volts anymore. I still haven't inspected the board or reapplied the AS5 yet - I'm really not looking forward to removing the mobo just to do this - I have a small case. I will probably do it today or tomorrow.
  15. Back from the weekend trip... the TCCD passed memtest fine at DDR622 for about 10 hours. I'm going to reapply the AS5 and check the board and chip. Probably tomorrow, I'm exhausted.
×
×
  • Create New...