![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Hi
I have an unsual problem I believe. First the spec:
- E8200 3.12Ghz (FSB390), vcore 1.05
- 2x1G Corsair DDR2-800 PC6400-CL4 (4-4-4-12, 2.0v)
- MSI P35 NEO2-FR (a renamed "P35 Platinum", very quality/money ratio)
- Gigabyte 8800GT 512Mb GV-NX88T512HP (type B, the one with copper heat pipes)
Now, before you start crying about vcore, let me tell you that by default this mobo uses 1.0375 vcore in setup (and real reported by programs is around 1.02v). I have a stable system at FSB390 as stressed by Orthos and general usage of the system with that default voltage. I had increased it a little to 1.05 because when I do want to get to FSB400 I need that little increase. Once at FSB400 with 1.05vcore the system is again very stable (by Orthos testing and general usage).
However, at FSB400 something strange happens when I try to stress the video card (either by trying to play STALKER or running "Video Card Stress Test" program), the video driver crashes (or sometimes worse, a blue screen shows for a milisecond and then the system reboots). I repeat, there seems to be no problems with the system with any other usage of stressing other than video stressing. One more thing to note about this issue is that when the system is stable with the Video Card Stress tool, the globe that rotates bellow (the program displays 2 globes that rotate) is rotating smoothly, almost without any interruption but when I get to FSB400 and before the driver crashes, the globe bellow rotates then interrupts for a little time then again rotates and so on until it blocks completely and the driver crashes. Does this means there is some kind of desyncrhonization happening?
Obviously the first thing to check out is the PCI-E frequency, but this motherboard makes it very easy/safe to overclock the FSB because the PCI-E frequency setting is separate and is at 100 (and does not increase if I increase the FSB, at least it does not increase in the setup menu). So theoretically the PCI-E frequency should be 100 so everything should be fine but for some reason it isn't when I get to FSB400, it works fine (3d video stress and all) at FSB390.
And no, increasing the ram voltage to 2.1v (which is the EPP spec for the 4-4-4-12 timings) or increasing the vcore to about 1.12 didn't helped either (I didn't wanted to increase it more because the setup program starts showing the vcore values with "red" colors when I get over 1.12). It all seems to be dependent only on the FSB here.
Is it possible that the motherboard does increase the PCI-E frequency although not told so? How can I then measure it with some program?
Or do you have any idea about any other voltage that might be needed to be increased? (although if the PCI-E frequency is not increased then I don't understand why other voltage would need to be increased to make the video work stable).
Please let me know if you have any idea about this problem or any other info you might need.
I have an unsual problem I believe. First the spec:
- E8200 3.12Ghz (FSB390), vcore 1.05
- 2x1G Corsair DDR2-800 PC6400-CL4 (4-4-4-12, 2.0v)
- MSI P35 NEO2-FR (a renamed "P35 Platinum", very quality/money ratio)
- Gigabyte 8800GT 512Mb GV-NX88T512HP (type B, the one with copper heat pipes)
Now, before you start crying about vcore, let me tell you that by default this mobo uses 1.0375 vcore in setup (and real reported by programs is around 1.02v). I have a stable system at FSB390 as stressed by Orthos and general usage of the system with that default voltage. I had increased it a little to 1.05 because when I do want to get to FSB400 I need that little increase. Once at FSB400 with 1.05vcore the system is again very stable (by Orthos testing and general usage).
However, at FSB400 something strange happens when I try to stress the video card (either by trying to play STALKER or running "Video Card Stress Test" program), the video driver crashes (or sometimes worse, a blue screen shows for a milisecond and then the system reboots). I repeat, there seems to be no problems with the system with any other usage of stressing other than video stressing. One more thing to note about this issue is that when the system is stable with the Video Card Stress tool, the globe that rotates bellow (the program displays 2 globes that rotate) is rotating smoothly, almost without any interruption but when I get to FSB400 and before the driver crashes, the globe bellow rotates then interrupts for a little time then again rotates and so on until it blocks completely and the driver crashes. Does this means there is some kind of desyncrhonization happening?
Obviously the first thing to check out is the PCI-E frequency, but this motherboard makes it very easy/safe to overclock the FSB because the PCI-E frequency setting is separate and is at 100 (and does not increase if I increase the FSB, at least it does not increase in the setup menu). So theoretically the PCI-E frequency should be 100 so everything should be fine but for some reason it isn't when I get to FSB400, it works fine (3d video stress and all) at FSB390.
And no, increasing the ram voltage to 2.1v (which is the EPP spec for the 4-4-4-12 timings) or increasing the vcore to about 1.12 didn't helped either (I didn't wanted to increase it more because the setup program starts showing the vcore values with "red" colors when I get over 1.12). It all seems to be dependent only on the FSB here.
Is it possible that the motherboard does increase the PCI-E frequency although not told so? How can I then measure it with some program?
Or do you have any idea about any other voltage that might be needed to be increased? (although if the PCI-E frequency is not increased then I don't understand why other voltage would need to be increased to make the video work stable).
Please let me know if you have any idea about this problem or any other info you might need.
