TerraRaptor

Thank you, I had results from Luumi and Noxinite as reference for 4.5GHz and higher fsb helped me a lot.

Riva should detect pci I think

In the past, overclocking was helping to save money. Nowadays you need to pay extra to access overclocking. Simple like that, good old overclocking is dead.

May be setting cpu msr registers. I remember cpumsr 0.88 was allowing to force sse support with older athlons.

I had that issue, pushed both contacts here and there softly and now it works fine. Probably the contacts inside the battery holder are shortened with battery and sometimes one needs to bend contacts a bit. Noticed today that one 16v capacitor near power button/usb headers is hot which is not normal. And that is definitely the capacitor itself as everything near it is cold. I think I will desolder it as I don't use those usb headers and that cap is probably belonging to usb as it is being hot even in standby.

Slots 2+4 are better btw..

? will truly appreciate it as I'm not able to follow you both the last pages:) Great cooperation friends!

Not sure but haven't killed any board with high vNB yet - I guess temperature control is the most essential thing when we talk about vNB as vNB affects temperature directly. Actually, x48 must be very sensitive to temperatures - while searching broken cap I was heating whole board several times and have noticed that ВС resistance of x48 changes dramatically with temperature - it is smth like 8 Ohm @25C, 4 Ohm@60C, <2 Ohm@100C. So, there could exist avalanche effect when NB starts heating rapidly dropping resistance further thus consuming more and more power. My guess that random stability issues with high FSB (i.e. same profile is stable on day1 and not stable on day2) may be caused by some hidden settings of x48 - board sets these automatically on powerup. We still can monitor/control these settings with software called BarEdit (other options may be Pmem, RWEverything etc) . Here is a good read on BarEdit>MCHBar settings compiled by Don_Dan - I think that if we collect enough dumps of registers from "better-than-usual days" and compare it to "usual days" dumps there is a chance of locating some entries that we can set manually and force stability. Just an example - DFI boards usually had much wider number of settings because Oscar Wu gave us an ability to manipulate some of those "not in public datasheets" settings.

Another step to hardware king status!

Thank you. Tried cold yesterday but no success - I was using the same profile that allowed me to post @675 at positive temparatures on cpu+nb but couldn't even get into windows with 660MHz once I've turned on my SS. Sam is right saying that REX+Wolfdales are not reliable with non-LN2 cooling at high FSB - or there is something we can discover with BarEdit to stabilize it.

Well done, nice custom WC. @ozzieI think result is not listed as usual as it gains no points (Eisbaer798 had better score with REX in the past) so we can see it only in new competition entries.

There are quite a lot options you may try - find awardflash that accepts /nab key (non-award bios - will flash any binary without questions). Other option is uniflash. Actually, I was using asrock board as backup-hotflash board when playing with bioses for p5w64, I flashed 2MB bios of C5F in REX that originally uses 1MB bios. Just find the right software and check available secret keys (may not be visible through help of the executive but still listed at many websites like rom.by).

Bet there is more in it if you make multi lower (seems x14 is available)

Well, NB should be the limit here - I was able to reproduce same clocks with e8200 today. Will try cold later this week.

Had the same issue yesterday, I can run either mem at given timings/frequency or cpu at given fsb - but if combined won't do more than 11 loops. Will try to find the solution in next two days - I remember proper dram skews helped me before.

Haha, funny thing is that we made same clocks the same day. As for screenshot, I was overclocking it with setfsb to the desired frequency, printscreen it, downclock and save it with paint.

Thank you friends. I will first try to improve on x7 multi superpi result and then will probably put nb on SS. Does anybody know if using pci video or single channel/single sided sticks make any difference with fsb?

I never play backup games. Once I have better result will submit it:) I guess AMD is still the winner here but intel was quite a fun.

Good score! Any input on fsb with air/SS for this particular chip?

Doesn't good waza completely eliminate the necessity of ramdrive? Or there is still some positive effect? I stopped using qsoftware ramdisk years ago thinking it is not necessary.

Noob question - what is eram?

Yes, 400 2:3. The board was not very stable from the very beginning with high fsb and I desoldered all futitsu caps and replaced them with tantal smds in ram and nb (nothing for cpu). Also added solid electrolyte cap for vtt. This gave me a bit more stable board but it still sucks.

I remember I used this trigger in rev5 to recover some missing points and it made me lost another 1000 hardware points. Assuming there is a bug with database I would be very careful with total recalc of my results nowadays and proceed recalculating them one by one.

Just a quick note - Biostar boards had 14 entries if I'm not mistaken:)