- Nov 28, 2007
- 3,204
- 0
- 76
Ok, so I have my E7200 bought in May and it was overclocked to 4 ghz at 1,46 V ( bios) since June. Today, while I was playing Dead Space, I had a hard lock and lost the video signal. I said to myself that it was probably a crash because of the drivers or something like that. Played Dead Space again and everything went smooth.
After that, I've opened up one of my projects in Autocad 2009 and bang, in one minute the computer froze again. Restarted it, opened ACAD again and the same thing happened. Now, I knew my 4870 wasn't the issue anymore, since ACAD 2d hardly stresses the videocard for rendering a couple of simple lines. So I said that it has to be related with the ram /cpu.
Ran memtest and it passed it with no problems. So it remain one single test to do: linpack. Did a quick "half a memory" 5 tests stress and it passed only one out of all five. Hmmm,so I reduced my overclock to 3,9 ghz and it still wasn't able to pass more then one test. Now I'm running it at 3,8 ghz with 1,42V and it seems to do 6 out of 6 tests. The problem is that for this frequency, I "only" needed 1,37 V before, to be perfectly stable in linpack.
All this time my cpu was overclocked to 4 ghz, never went over 67 C, so I always had a good 33 degrees until T junction. In my book, that is a good temperature and yet, degradation happened.
So my advice to you, all the E7XXX overclockers, is to keep the voltage bellow 1,4 V, maybe even bellow the 1,36 V, as Intel states, if you want your cpu to live more. What happened to my chip, could be just the beginning and it might go downhill from now on, until it dies. I knew the risks when I kept it on a higher then "usual" voltage, so please do not start flaming me. This thread is meant to be a warning and you might consider it an experiment I did so you don't have to, so don't start with the "I told you so".
!!!!!UPDATED!!!!!!:
Yesterday I got a big bsod in CoH, right before the mission was loading. Had a feeling about it, and ran Linpack and the horror: it passed only one test out of ten. Tried in bios a couple of voltage increases to several components and only by upping the vcore to 1.45 V made the cpu stable again, at 3,9 ghz. So, if this isn't cpu degradation, then I don't know what it is. Temps are just about what they were before, so nothing happened there. I'm going to raise this sucker's voltage until it melts away. I will not go lower then 3,9 ghz, no matter what.:sun:
After that, I've opened up one of my projects in Autocad 2009 and bang, in one minute the computer froze again. Restarted it, opened ACAD again and the same thing happened. Now, I knew my 4870 wasn't the issue anymore, since ACAD 2d hardly stresses the videocard for rendering a couple of simple lines. So I said that it has to be related with the ram /cpu.
Ran memtest and it passed it with no problems. So it remain one single test to do: linpack. Did a quick "half a memory" 5 tests stress and it passed only one out of all five. Hmmm,so I reduced my overclock to 3,9 ghz and it still wasn't able to pass more then one test. Now I'm running it at 3,8 ghz with 1,42V and it seems to do 6 out of 6 tests. The problem is that for this frequency, I "only" needed 1,37 V before, to be perfectly stable in linpack.
All this time my cpu was overclocked to 4 ghz, never went over 67 C, so I always had a good 33 degrees until T junction. In my book, that is a good temperature and yet, degradation happened.
So my advice to you, all the E7XXX overclockers, is to keep the voltage bellow 1,4 V, maybe even bellow the 1,36 V, as Intel states, if you want your cpu to live more. What happened to my chip, could be just the beginning and it might go downhill from now on, until it dies. I knew the risks when I kept it on a higher then "usual" voltage, so please do not start flaming me. This thread is meant to be a warning and you might consider it an experiment I did so you don't have to, so don't start with the "I told you so".
!!!!!UPDATED!!!!!!:
Yesterday I got a big bsod in CoH, right before the mission was loading. Had a feeling about it, and ran Linpack and the horror: it passed only one test out of ten. Tried in bios a couple of voltage increases to several components and only by upping the vcore to 1.45 V made the cpu stable again, at 3,9 ghz. So, if this isn't cpu degradation, then I don't know what it is. Temps are just about what they were before, so nothing happened there. I'm going to raise this sucker's voltage until it melts away. I will not go lower then 3,9 ghz, no matter what.:sun:
Facebook
Twitter