- Jan 29, 2005
- 5,202
- 216
- 106
Ah !
After a couple of months of tries, I finally found that stable spot we're always looking and testing for to make our systems stable at OC'ed speeds on a 24/7 usage basis. A 4400+ from 2.2Ghz to 2.53Ghz (2531, more exactly), isn't that lovely ? Humble perhaps, but hey it's on stock cooling ! It always refused to get past the 220FBS/2.42Ghz mark without crashing or even refusing to POST at all, or just never finished the CPU test in 3DMark06, or never loaded up CPUz without crashing either ... in other words it was never stable.
Well I think I found that little spot, stable, finally:
http://img153.imageshack.us/img153/4370/oc1yx3.png
http://img440.imageshack.us/img440/2816/oc2qw3.png
As you can see, I used a Memory divider, and at such a lower CPU multiplier I could keep my Memory at around the same speeds it was before (in my sig). I only lost 10Mhz on the Memory while I gained around 110Mhz on the CPU speed. Good deal isn't it ? Well, so far, Prime95 was stable (Blend and FTT's) for 18 hours without errors. I also had a few hours of gaming sessions with Oblivion and Lost Planet (both using two Cores, especially Lost Planet) without any issues whatsoever. Along with the inevitable 3DMark05 and 06.
I'm posting all of this for two reasons, because, first of all, I'm proud of that over-clock for such a lemon of a CPU, and also because, secondly, I wanted to have your opinions about the vCore setting ... vCore is set at 1.35v if I remember well, at default speeds, and is now 0.1v higher, at 1.45v, for a speed of 2.53Ghz. Is that alright for an X2 ? Right now it idles at 47ºC, and, during the Prime95 FFT's test (that test runs both Cores at 100% activity non-stop) it never went above 59ºC (according to my ASUS Probe logs). And, as I mentioned above, I'm using the default air cooling solution from AMD, however it happens to be the model with the two heat-pipes on each sides (I know that a few X2's were shipped with a non-piped cooling model), which I guess might help my case a little.
One last thing, and it concerns the first CPUz screen-shot above ...
As you guys can see CPUz indicates that my vCore runs at 1.504v, but in reality it constantly fluctuates between 1.495 to 1.580 when CPUz is open and I leave it running. Is it a known issue to CPUz ? I hope so ... because my vCore is set at 1.45v within the BIOS and I wouldn't want to break my CPU without being aware of it in the first place. Well I do know that OC'ing is supposed to reduce the life-span of any electronics, especially CPU's since they are supposedly more sensible to that kind of stuff, but what I mean is that I wouldn't want to exaggeratively increase the vCore to a point where it'd just be dumb and useless for the desired settings. I've tried a lower vCore anyway, at 1.425, and it wasn't stable for more than 64 minutes in Prime95, after which one of the Cores failed and the other kept on running the test, which is why I had to come up with the 1.45v final setting, which made the CPU's stable.
Well, anyways, what do you guys think about all this ?
Oh, and I scored 9041 in 3DMark06 with those settings (rest of specs in sig) using ForeWare 160.02 WHQL, I will post the compare link of it if requested.
After a couple of months of tries, I finally found that stable spot we're always looking and testing for to make our systems stable at OC'ed speeds on a 24/7 usage basis. A 4400+ from 2.2Ghz to 2.53Ghz (2531, more exactly), isn't that lovely ? Humble perhaps, but hey it's on stock cooling ! It always refused to get past the 220FBS/2.42Ghz mark without crashing or even refusing to POST at all, or just never finished the CPU test in 3DMark06, or never loaded up CPUz without crashing either ... in other words it was never stable.
Well I think I found that little spot, stable, finally:
http://img153.imageshack.us/img153/4370/oc1yx3.png
http://img440.imageshack.us/img440/2816/oc2qw3.png
As you can see, I used a Memory divider, and at such a lower CPU multiplier I could keep my Memory at around the same speeds it was before (in my sig). I only lost 10Mhz on the Memory while I gained around 110Mhz on the CPU speed. Good deal isn't it ? Well, so far, Prime95 was stable (Blend and FTT's) for 18 hours without errors. I also had a few hours of gaming sessions with Oblivion and Lost Planet (both using two Cores, especially Lost Planet) without any issues whatsoever. Along with the inevitable 3DMark05 and 06.
I'm posting all of this for two reasons, because, first of all, I'm proud of that over-clock for such a lemon of a CPU, and also because, secondly, I wanted to have your opinions about the vCore setting ... vCore is set at 1.35v if I remember well, at default speeds, and is now 0.1v higher, at 1.45v, for a speed of 2.53Ghz. Is that alright for an X2 ? Right now it idles at 47ºC, and, during the Prime95 FFT's test (that test runs both Cores at 100% activity non-stop) it never went above 59ºC (according to my ASUS Probe logs). And, as I mentioned above, I'm using the default air cooling solution from AMD, however it happens to be the model with the two heat-pipes on each sides (I know that a few X2's were shipped with a non-piped cooling model), which I guess might help my case a little.
One last thing, and it concerns the first CPUz screen-shot above ...
As you guys can see CPUz indicates that my vCore runs at 1.504v, but in reality it constantly fluctuates between 1.495 to 1.580 when CPUz is open and I leave it running. Is it a known issue to CPUz ? I hope so ... because my vCore is set at 1.45v within the BIOS and I wouldn't want to break my CPU without being aware of it in the first place. Well I do know that OC'ing is supposed to reduce the life-span of any electronics, especially CPU's since they are supposedly more sensible to that kind of stuff, but what I mean is that I wouldn't want to exaggeratively increase the vCore to a point where it'd just be dumb and useless for the desired settings. I've tried a lower vCore anyway, at 1.425, and it wasn't stable for more than 64 minutes in Prime95, after which one of the Cores failed and the other kept on running the test, which is why I had to come up with the 1.45v final setting, which made the CPU's stable.
Well, anyways, what do you guys think about all this ?
Oh, and I scored 9041 in 3DMark06 with those settings (rest of specs in sig) using ForeWare 160.02 WHQL, I will post the compare link of it if requested.