Cascade benching






i knew that 5GHz was absolutely no issue so i just set off booting into windows at 500*10=5GHz wow EEK

first run

32M SuperPi @ 5Ghz with 1.7v




next stop 3DMARK06 after the 32M run but i got a strangely low score >> dropped it down to 4.5GHz thinking it could be the CPU but still the score was strange.....PCIe at 16x ....well i gave up troubleshooting quickly as i had no time so will have to investigate that one some other time lol >> sometimes this happens if the vcore is too high but the vcore was only 1.7v so it couldn't have been the case





moving on
wPrime 32M at 5133MHz





i needed 5160MHz to finally break 10 second 1M SuperPi so stock efficiency is pretty bad but this is nothing of concern once i get time and start tweaking it down......one thing i did ask from Gigabyte is to provide strap change in bios to select 266 strap as these benchmarks are all done on 333 or worse (in any case MCH latencies are pretty loose)





at this point kayl is starting to bug me again to speed it up......oh well i stuck 1.8v into the CPU and gave 32M a go first hahahahhaha....i tightened the timings a bit to get a slightly more respectable 32M run which is still a good 40-50 seconds too slow at these clocks

5160MHz 32M SuperPi 1.79v





started a run at 5270 but it stopped on 5th loop from memory so i backed down to
5240MHz 32M SuperPi 1.8v




bump in clocks

5320MHz 1M SuperPi 1.8v still with both cores enabled



5413MHz validation
1 or 2 cores made no difference oddly enough

http://valid.x86-secret.com/show_oc.php?id=234645



i have high hopes with this CPU and my Duniek pot + LN2

obviously some more tweaking for Pi will take place and we will be rock and rolling into some serious figures heh

until next time