Overclock.net banner

1 - 8 of 8 Posts

·
Registered
Joined
·
1,142 Posts
Discussion Starter #1
I just started folding a few days ago <img alt="" class="inlineimg" src="/images/smilies/biggrin.gif" style="border:0px solid;" title="Big Grin"> and I'm working on a 250 frame WU that will reward myself and overclock.net team 37726 - 600 points.<br><br>
Here's my question to yee Folding GODS of overclock.net: I am hesitant to set my CPU usage to 100% only because I just built my new rig about a month ago along with my new prized possession of an Athlon San Diego 4000+ (1.36 Vcore, 90nm, 1mb cache). I know that what is critical for CPU performance is temps - and right now I'm running the San Diego at 60% for Folding and my temp hovers around 35c which is acceptable.<br><br>
Jumping my CPU usage to 100% my temp jumps then to 40c which is also acceptable though I think once I install my lapped Big Typhoon VX I'm expecting a few degrees less. <img alt="" class="inlineimg" src="/images/smilies/smile.gif" style="border:0px solid;" title="Smile"><br><br>
But I'm concerned about what kind of possible wear and tear or negative effects I might have on my rig or CPU by just running the CPU 100% daily - when I'm at work and at night - i.e. keeping it going 100%, such as 24/7. What are the common working assumptions or opinions on this? Should I not be worried at all and simply go for it? Can there be some risk of reduced longevity or other possible negative side effects? Or if I simply monitor my temps carefully and make sure they are between 35c & 45c I shouldn't be concerned at all - since mobos/CPUs/RAMs are built to run 100% round the clock??<br><br>
Thanks in advance for opinions on this question - or if you can point me to answers already provided. <img alt="" class="inlineimg" src="/images/smilies/rolleyes.gif" style="border:0px solid;" title="Roll Eyes (Sarcastic)">
 

·
Registered
Joined
·
1,769 Posts
From everything that I have heard it is not running your chip at 100% that would shorten the life but the temps that will.<br><br>
So for example if the chip is running 40C idle compared to running 100% at 35C. There would be more harm being done at 40C<br><br>
Someone correct me if I am wrong but as long as your temps are fine you should not have to worry about running 2 cores 24/7
 
  • Rep+
Reactions: jamenta

·
Premium Member
Joined
·
8,095 Posts
You're fine, man...<br><br>
Temps are the only thing really, that'll wear it out any faster. Even if you fold 24/7 for 10 years, I don't think you'd kill it.
 

·
Premium Member
Joined
·
5,573 Posts
<div style="margin:20px;margin-top:5px;">
<div class="smallfont" style="margin-bottom:2px;">Quote:</div>
<table border="0" cellpadding="6" cellspacing="0" width="99%"><tr><td class="alt2" style="border:1px inset;">
<div>Originally Posted by <strong>Sideburns</strong></div>
<div style="font-style:italic;">You're fine, man...<br><br>
Temps are the only thing really, that'll wear it out any faster. Even if you fold 24/7 for 10 years, I don't think you'd kill it.</div>
</td>
</tr></table></div>
<br>
Yeah, those semiconductors are rated to withstand very high temps for a long time.<br><br>
The thing that REALLY kills computers? Turning them on and shutting them off alot. Thermal stress will eventually eat away at every component on your mainboard. Keeping it at a constant temperature is the best thing you can do.
 
  • Rep+
Reactions: jamenta

·
I remember when......
Joined
·
3,812 Posts
I recomend a detailed break-in period for CPUs, but most people laugh at me. <img alt="" class="inlineimg" src="/images/smilies/wink.gif" style="border:0px solid;" title="Wink"><br><br>
A break in period is about a week to 2 weeks of varied system load at stock settings. I start off with a couple days of mostly idle and 50% load times. Then a day of completely varied loads from idle to 100% in spikes. Then 2-3 days of straight 50% constant load. Then 8 hours of time spent shut off. Then another day of 50% load, and then 2 days of 100% load, then 8 hours of time shut off. Then 3 days of 100% load, then 8 hours of time off again, then 3 days of 100% load, then 8 hours off......are you still reading? <img alt="" class="inlineimg" src="/images/smilies/wink.gif" style="border:0px solid;" title="Wink"><img alt="" class="inlineimg" src="/images/smilies/biggrin.gif" style="border:0px solid;" title="Big Grin"><br><br>
After that, it's time to overclock.<br><br><br>
Now, technically, if we state that a CPU will only last <b>X</b> hours, then it is logical to state that a CPU running 24 hours a day will reach it's end of life sooner than one running only 8 hours a day. The difference between running it for those same hours at 60% or 100%, or even 35% load, while real, would be barely measureable IMHO. There is no way to determine exactly how long a CPU will last, it depends on a great many variables, from the wafer it started from, to the amount and quality of voltage it is given to the temperatures it is run at. As long as you keep the voltages within specs and temperatures under control, a CPU should, <span style="text-decoration:underline;">should</span> last for at least 3 years. This is the time frame that Intel warrenties their CPUs (not sure about AMD but assume it would be similar time frame) for (within listed voltage and thermal specs) so it is a good rough gauge when staying within those guidlines. Intel is not going to offer a warrenty period that a majority of their CPUs can not meet, they would go out of business. The actual time frame for most is probably a lot longer, again, depending on a great many factors.<br><br>
Exceeding the listed voltage and/or temperature guidlines (when overclocking lower temperatures than the manufacturer recomended ones are best as overclocking stresses the CPU more than stock settings) will result in shorter life span.<br><br>
This is usually not as big a concern for most overclockers as we tend to replace our equipment more often than the average user. If you plan to keep this CPU until it dies and hope that's over 8 years, you might want to think long and hard about overclocking and folding. However, if you know you will be replacing this system in 3 years or less, you really have little to worry about in terms of 24/7 operation. And of course, it can continue to fold after it is replaced if you don't sell it to fund your next system. <img alt="" class="inlineimg" src="/images/smilies/thumb.gif" style="border:0px solid;" title="Thumb">
 
  • Rep+
Reactions: jamenta

·
Premium Member
Joined
·
5,573 Posts
<div style="margin:20px;margin-top:5px;">
<div class="smallfont" style="margin-bottom:2px;">Quote:</div>
<table border="0" cellpadding="6" cellspacing="0" width="99%"><tr><td class="alt2" style="border:1px inset;">
<div>Originally Posted by <strong>Thumper</strong></div>
<div style="font-style:italic;">I recomend a detailed break-in period for CPUs, but most people laugh at me. <img alt="" class="inlineimg" src="/images/smilies/wink.gif" style="border:0px solid;" title="Wink"><br><br>
A break in period is about a week to 2 weeks of varied system load at stock settings. I start off with a couple days of mostly idle and 50% load times. Then a day of completely varied loads from idle to 100% in spikes. Then 2-3 days of straight 50% constant load. Then 8 hours of time spent shut off. Then another day of 50% load, and then 2 days of 100% load, then 8 hours of time shut off. Then 3 days of 100% load, then 8 hours of time off again, then 3 days of 100% load, then 8 hours off......are you still reading? <img alt="" class="inlineimg" src="/images/smilies/wink.gif" style="border:0px solid;" title="Wink"><img alt="" class="inlineimg" src="/images/smilies/biggrin.gif" style="border:0px solid;" title="Big Grin"><br><br>
After that, it's time to overclock.<br><br><br>
Now, technically, if we state that a CPU will only last <b>X</b> hours, then it is logical to state that a CPU running 24 hours a day will reach it's end of life sooner than one running only 8 hours a day. The difference between running it for those same hours at 60% or 100%, or even 35% load, while real, would be barely measureable IMHO. There is no way to determine exactly how long a CPU will last, it depends on a great many variables, from the wafer it started from, to the amount and quality of voltage it is given to the temperatures it is run at. As long as you keep the voltages within specs and temperatures under control, a CPU should, <span style="text-decoration:underline;">should</span> last for at least 3 years. This is the time frame that Intel warrenties their CPUs (not sure about AMD but assume it would be similar time frame) for (within listed voltage and thermal specs) so it is a good rough gauge when staying within those guidlines. Intel is not going to offer a warrenty period that a majority of their CPUs can not meet, they would go out of business. The actual time frame for most is probably a lot longer, again, depending on a great many factors.<br><br>
Exceeding the listed voltage and/or temperature guidlines (when overclocking lower temperatures than the manufacturer recomended ones are best as overclocking stresses the CPU more than stock settings) will result in shorter life span.<br><br>
This is usually not as big a concern for most overclockers as we tend to replace our equipment more often than the average user. If you plan to keep this CPU until it dies and hope that's over 8 years, you might want to think long and hard about overclocking and folding. However, if you know you will be replacing this system in 3 years or less, you really have little to worry about in terms of 24/7 operation. And of course, it can continue to fold after it is replaced if you don't sell it to fund your next system. <img alt="" class="inlineimg" src="/images/smilies/thumb.gif" style="border:0px solid;" title="Thumb"></div>
</td>
</tr></table></div>
It wouldn't be too difficult to turn this into a program that runs the cpu at X% for X hourr, reboots, does the next step...etc etc...until the unit is "burned in". That would be a neat program. It would: A) Seat all the thermal compounds and B) seriously test the integrity of the hardware.<br><br>
Not a bad idea.
 

·
Registered
Joined
·
1,214 Posts
yeah don't worry about the 100%. Just keep your temps under 50.<br><br>
Honestly, if a cpu lasts 10 years, isn't that more than enough? Who would expect a computer to have a ten year life span these days?
 

·
Registered
Joined
·
1,142 Posts
Discussion Starter #8
George Bush?
 
1 - 8 of 8 Posts
Top