The Second LN² session
The Second LN² session.
The new mainboard is insulated and ready to use. Lets hope it runs better than the first mainboard.
The other components I used were exactly the same, the only difference was that this time I had also installed a probe to measure the temperature of the power management. During the whole session the temperature of the power management was between -30 and -40 degrees, so that was ok.
Again, the main focus was to bench 3D, but I also had serious problems this time reaching high speeds with 4 cores.
It went a little bit better, but still above 5,7GHz was a no go when it came to benching with 4 cores.
I also had some strange issues benching 3DMark05, even though I benched with only 2 cores. A system like this should easily be able to bench this benchmark above 6GHz, but no matter what I did, it just would not run the benchmark.
After about 2 hours I stopped and continued to bench 2D again. This was more to see the maximum speed I could realize with one core using Superpi 1m and Pifast.
Going strong and fast though, below you see the best results I managed to reached after about four hours of benching, yes, at this stage we reached nearly 6500 MHz (!), but we let go of the 3DMark series due to the aforementioned reasons.
Superpi 1M @ 6438 MHz:
Time : 10,764 sec
I had one faster score of Superpi 1M (10.733 sec) but during the print screen and saving of the picture the system rebooted, so this was really the absolute maximum of the CPU. A multiplier of 29 was used in combo with a 222 MHz bus speed.
Pifast @ 6264 Mhz :
Time : 17,64 seconds
And here a result for PiFast, slightly slower at 6264 MHz at a 216 MHz bus and a multiplier of 29.