3000+ Venice and DFI weird and annoying prob

JhontaJhonta
edited July 2005 in Hardware
my spec is this
3000+ Venice
DFI LanParty Ultra D
2X512 Twinmos Speed Primiuem
Enermax 420W 24 Pin

my system is up and running for a few days now so i thought giving OC a go

i started with getting the FSB up to 220 with no V change and the memory working on 2.9 VDIMM(2-2-2-5)
everything went well

but whenever i tried to get the CPU higher than that (230*9, 260*8) i would get a post but when the windows started loading, i got a sudden reset
i tried setting the vcore up to 1.45 but still, the prob remains
when i worked with the 250 FSB i changed the HTT (which i saw was under a diffrent name) to X4 but as i said, nothing helped

am i missing something here??
what about the chipset and the HTT V ? need a change ?
maybe a bad confg in the bios.. ? but what can cause that ?

hope to get an answer soon.. :(

BTW
about the board power connectors,
i didn't connect the floppy power cord.. but the molex, the 24PIN and the 4PIN i did... maybe that's the prob ?? :confused:

Comments

  • Mt_GoatMt_Goat Head Cheezy Knob Pflugerville (north of Austin) Icrontian
    edited July 2005
    If you have at least the molex on the board connected you should be OK. You will need to loosen the memory timings as you increase the HTT speed as well as increse the dimm voltage. For starters try setting your memory to c3-4-4-7 and the dimm volts to 2.8.
  • TheLostSwedeTheLostSwede Trondheim, Norway Icrontian
    edited July 2005
    Memory voltage...at 250, you need at least 3.3 volts for stability. The connectors are fine. I use them both even though one should be ok. The key is to fiddle with the timings.
  • JhontaJhonta
    edited July 2005
    but the thing is that i didn't change the memory speed (i played with the ratio and set the memory at around 200-210)
    and i didn't change the htt as well... 250*4 = 1000...

    and y change the timing to 3-4-4-7 ?? that's really bad... isn't it ?

    and one last thing,
    i'm really afraid to set the VDIMM above 3.0 V...
    by what i've heared, it can be exremly dangerous for the memory and the V generator or some like that.. something about setting a fan right above it...
    is that true ??
  • TheLostSwedeTheLostSwede Trondheim, Norway Icrontian
    edited July 2005
    If you have a fan over the dimms, 3.4 volts isn't a problem...at least not with our memory as we actually guarantee it. I dont know with Twinmos though.
  • muddocktormuddocktor
    edited July 2005
    I've run 3.1-3.2v vdimm in my OCZ VX, which is using basically the same chips as your TwinMos SP with just decent airflow in the case and no fans directly over them and no problems. I also have some value VX, but I presently have it in a rig that can only supply 2.8v vdimm (damn Epox, not enough vdimm and vcore without mods). The Value VX will go to around 235 MHz at 2-2-2-5 timings (tested on a DFI NF3-250 Gb) with 3.2 v vdimm applied though and it should be a direct comparison to your SP's as it's just untested UTT on a brainpower pcb. Give your ram some volts. :thumbsup:
  • JhontaJhonta
    edited July 2005
    ok guys, the problam remains...
    i just got my xp90+tt silentcat today (and no major change in temp)
    and i gave OC another go

    i set the vcore up to 1.55 and it booted at 256*8 but from there on, the same prob, it just restarts it self when it gets to windows logo at startup
    now my xp acts funny and i have to format it again

    but i really want to solve this annoying prob first !
    y can't i get to decent oc performence with my CPU ??
    bad stepping ??
    plz guys help here... :bawling:
  • muddocktormuddocktor
    edited July 2005
    Crank the volts up on your ram and loosen your timings or run with the 166 mem divider. You need to isolate whether your ram or the proc is crapping out, but I'm betting it's the ram. Also try running memtest86 at your highest speed you can boot at and see what the ram does.

    UTT chip ddr needs a lot of volts to run high fsb speed with tight timings.
  • TheBaronTheBaron Austin, TX
    edited July 2005
    I thought he said he was using a divider already, so the ram wasn't being OC'd? maybe I misread
Sign In or Register to comment.