Anyone running their AMD Sempron/Athlons at 1.6v or higher on air?

[homestarmy]

I'm running my Sempron 3000+ at 1.63v (as measured by motherboard utility) with good case airflow and an Arctic Cooling HS/F and have been doing so for a couple of months now. Idling in Florida with no A/C is as low as 38 and at full load in the low 50's, around 52. I have a better heatpipe cooler that should keep me in the 40's at load on the way.

Anyone else running this high? I read a post that stated that the higher voltage through the processor will lower the life of the processor whether the temps are high or not. Is this valid?

 

[Zap]

IIRC processors are rated at +/- 10% voltages. Also, sometimes processors of the same family have different voltage ratings, so you can do 10% above the highest rated core and still be "safe" as long as temperatures stay low. For a Sempron, this means 1.54v should be considered perfectly "safe."

Another thing is that usually you'll hit the highest "easy" overclock within that safe zone, and once you go beyond it you know you're already reaching the limits of your particular core. You may be able to drop the overclock a hair while reducing voltage considerably.

Something else to consider is that the motherboard may not report voltages accurately. I've heard there's a way to measure vcore using a multimeter on certain parts of the motherboard while it is running, but I don't know where for certain, perhaps at the coils? Anyone have more information on this?

 

[nJett]

Has anyone here ever actually cooked a CPU? I've never done it and I've tried before. Admittedly it was a Celeron I was trying with. I did run my AthlonXP with the heatsink loose so that it overheated crazy fast from the overvolting - just like the Celeron though it was fine. It's run for years after that without issue (except during summers when I had to reduce the OC on it so it wouldnt overheat and shut itself off). I ran a P4 w/o any fans going, that thing cooked damn fast and it lived, took a long time to cool off though! I wouldn't worry about running it hot, if you can drop the voltage a little I'm sure it wouldn't hurt but if the CPU gets too hot it will shut itself down, I've yet to see one that won't and I've yet to see one that has shown any ill effects from running hot. My AthlonXP literally ran for hours on end within a few degrees of the thermal shutdown limit year after year and it's still working fine to this day.

 

[o1die]

Today I cooked my barton 2600, and it defaults to a lower multiplier now. Old multiplier was 11.5, new one is only 8.5. Something melted inside. I was using an old duron heatsink until my new one arrived. Big mistake. It was just too warm today, about 90 degrees.

 

[BlueWeasel]

I've got a Epox NF3 board that is either (1) overvolting by a large amount or (2) the voltage monitoring is inaccurate.

My Sempron 3100+ is currently running at 2.5Ghz (309x8), and I've bumped the voltage up to +0.05v in the BIOS. However, the board reports the CPU voltage current running at 1.53v idle. Further, when running full load, the voltage peaks at 1.56v-1.57v. It's strange, as usually the voltage drops slightly when running at full load.

As the stock voltage for the newer Semprons is 1.40v, I should be seeing a Vcore around 1.45v, yet it the BIOS reports it as being about +0.10v higher. I've been meaning to test the voltages using a DMM, but haven't got around to it yet.

 

[homestarmy]

Thanks for the info guys. I would be really interested if someone could tell me where you could measure the vcore on the board using a multimeter.

And also, does anyone ever want to slap those people that say... "if you run your processor like that, it will shorten its lifespan..." Who cares? What is the normal lifespan of a processor? At least over 10 years! Is this going to shorten it to 8? 5? Hell, I won't have it at that time, so I'm not sure that I would really care.

Someone please tell me why I should .

Oh, and I was able to get CnQ working by dropping my FSB 3 points. So now it is running at 1.28v idle and 1.568v full load. I couldn't really tweak the voltage too much on this board, as a lot of the settings give the same voltage, and then the next setting jumps it up quite a bit.

Maybe I'll try this thing at ever higher and go for a suicide run... haha! What are the real chances of me actually cooking this thing? Like was stated above, it'll just shut itself off, no? I have an FX-60 Heatpipe cooler on it now which is completely overkill...

 

[trexpesto]

Originally posted by: o1die
Today I cooked my barton 2600, and it defaults to a lower multiplier now. Old multiplier was 11.5, new one is only 8.5. Something melted inside. I was using an old duron heatsink until my new one arrived. Big mistake. It was just too warm today, about 90 degrees.

I am so sorry. I can't even imagine the pain you must be feeling. Please accept my condolences. Did you clear cmos?