madpete

I was hoping DFI would include some universal mounting holes for water blocks....looks like I will need a new CPU block to use this board.

I break it down into Memory and CPU seperately rather than both at the same time, I find that gives me a better picture as to what is holding me back or can be pushed more. For CPU I use OCCT, if it passes that then next is Prime95 torture test (Small FFT's) overnight, if it passes that too I consider my CPU stable enough. For memory I use memtest+ v1.60 for inital testing or burnin again overnight, then memtest for windows for a more thorough test. I've known memtest for windows pick up errors within minutes, that memtest+ did not find after hours. Good luck with your overclocking

Bov, to enable T2, go into your bios and set CPC to disable.

My experience of this has been that the benchmarks will come out better with the higher mhz, however in the real world your programs and apps prefer the lower timings. I would probably say the 290mhz @2.5-3-3 but like the lads say run some tests, try something like 3dmark 2001se too as this will give you a better feel than sandra etc. If I remember correctly, test 2 is sensitive to memory performance.

Hey guys, a while back I started this thread after my second lot of ram died on me (first lot was OCZ VX). Due to two different manufactures of ram, the rumours floating around, and the fact that I had stuck to the warranty voltage and provided plenty of cooling I blamed the DFI for the death of my memory. 3 or 4 weeks ago I received of replacement Redline modules from Mushkin, determined that this time the memory would not be killed by dodgy volt control I brought a OCZ DDR Booster and an OCZ PowerStream to replace my old Enermax PSU. With these installed I set about installing the Redline, which I had set at the following: 260mhz / 1.5-2-2-6 / 3.46v (using booster). The ram handled this no problem, with no instability and no errors in any memtest I ran. All was fine, and I was sure that I was safe this time round, until yesterday it happened again. Exactly the same thing, one stick dead, one stick ok…system wouldn’t post. I would like to say now that given what’s happened, I no longer believe the mobo was at fault for the death of any of my ram. I now agree 100% with Pershoot’s comments that CH-5/UTT cannot handle high volts for periods of time, I have no experience of BH-5 yet so I can’t comment on that. If you have to get / use CH-5/UTT I would keep it under 3.35v no matter what the cooling, and buy from OCZ or Mushkin both of which have given me excellent support when I needed it. The point here today anyway is to set the record straight on what I say earlier in the initial post, hopefully I’ve done that. Pete.

FX's always run hotter than standard A64's, I have no idea why but they just do. Angry, how the hell you have got your temps that low on air I do not know, do you live in the artic? Or maybe in a fridge?

Sounds very similar to whats happened to my board....I spent ages trying differant bioses and settings, tweaking the rails on my psu etc but the only thing thats helped is installing the OCZ Booster. It seems this has taken over the voltage regulation from the mobo and now I am stable again, I still say these boards are wobbly when it comes to memory....waiting on ATI to release crossfire so I can upgrade.

Lowering it will only help on UTT type memory (VX/ Redline etc), TCCD does not like this being lowered too much. I can't give you a technical answer on why this is, only know from hours of testing and having both types of ram to play with.

Worked fine for 2 days... then the system would randomly refuse to post at first, a reboot would cure it....just gradually got worse until it would not post at all. Found out that one stick would post, and the other wouldn't. This was on two differant PC's, Mushkin have confirmed that they could not get the one stick to post. Replacements should be shipping to me around now hopefully.

I was running them off the 5v line @3.5v with lots of cooling....you tell me. Actually it was just the one stick that died, but Mushkin wanted to replace both for me, so they can test them in dual channel which is fine by me...just waiting to get them back now. I brought a OCZ booster now, to tide me over till the ATI boards come out. Great memory before it died by the way, I had it upto 270mhz @1.5-2-2-6 using above volts.

Try changing the following: DRAM Drive Strength = Lvl3 DRAM Data Drive Strength = Lvl1 Use 3.3v to start off Trc 10 or 12 (I found it didn't hurt performance much to go higher, but really hurt stabillty to go lower - if your running at 7 that will be why you get errors). My Redline XP4000 REALLY liked those settings before they died on me, good luck.

If using Venice or San Diego and high FSB with this bios and you are having instability, try setting your LDT multi to AUTO in the bios

Same thing happened with my old VX, anything above 3.3v and errors no matter what, I was told by the OCZ guys that it needed burning in which I did and it was slowly getting better over time. After a few days of burnin I was getting less and less errors, and eventually I could use 3.4v with no errors...unfortuanatly thats as far as it went because the VX died while on a overnight memtest burnin (yes I was using plenty of cooling). This was before anyone knew about the "alleged " 5V problem. If your running of the 3.3v rail then in thoery your safe, so it might just need some more burnin...upto you. You might also wanna go speak to the extremely helpfull OCZ chappies over at bleeding edge.

I noticed that a couple of times, but I think its just temp control on the boards part. I monitor my temps quite closely and I have never seen them go dangerously high. One thing I have noticed tho is, a differant between the orange and yellow slots. Say you put your ram in the orange slots, and say set volts to 2.8v in the bios...smart guardian will ALWAYS reports the voltage @2.77v for me (which might wobble ever so slightly). If I put that same ram in the yellow slots, and set volts to 2.8v in bios, smart guardian will report voltage at exactly 2.8v and does not seem to wobble. I know smart guardian might not be the most reliable tool, but the results are constant. I think that if I was going to start again with high volts I would use the yellow slots this time....but thats just my instict.

Does anyone have anymore info on this??? I really need to know if thats the case