Jump to content
HWBOT Community Forums

GRIFF

Members
  • Posts

    984
  • Joined

  • Last visited

  • Days Won

    11

Everything posted by GRIFF

  1. The strange thing that with the same profile an E8500 reaches 656 of fsb in spi 32M with air cooled NB. The cpu on the GA-EP45-UD3P has validated 700mhz fsb with the ram in single channel. in dual channel instead the limit is 622 of fsb, exactly like on REX in dual channel and divider 5:8 with strap 333. I still have to try other straps and dividers. I started with this which was already tested on the E8500.
  2. I Need to solve issue with fsb Wall at 622 fsb with dual channel and 5:8 divider and 637fsb at single channel and 5:8 divider. 636*9 Is not limit for this CPU at cascade cooling)))))
  3. I don't think so, I don't have the possibility of finding nitrogen for the moment.
  4. Hello to all. I wanted to tell a little about this beautiful card. Doing the tests for the CUP with the E8500 arrived at 650 * 9.5 and 1.82V doing mathbench on a restart the card has stopped starting. After verification I find the VRM mosfets in short. I look for datasheets and find that the original mosfets are painful. Composition of a single phase is made as follows: 20A + 2 * 29A. Soldered the new mosfets in combination 75A + 2 * 81A. Result: cold mosfets even with 1.9V on the cpu and under stress. mathbench closed at 650 * 9.5 at 1.78V.
  5. For me is the combination of the frequency + voltage. At lowest frequency but at the 3.6-3.7V all right. It also depends on luck
  6. Vcpu-1.67V Vdimm-3.6V Another CPU. Rip my best bank. later I make a 256mb bank from this
×
×
  • Create New...