-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Truth about CPU degradation.
- Thread starter bluezero5
- Start date
Interesting and varying opinions here. My Q6600 based system won't hold an OC after 5years but I suspect it is the motherboard (VRMs?) rather than the chip. Ran into a similiar problem with the Q9450 and replacing the MB restored the OCing and corrected the problem. So perhaps MB stress needs to included in this discussion.
Iindeed it is an interesting thread.
To me is kinda unspoken. Heat and voltage are the cause of degradation. Motherboard and CPU are Siamese twins. GPU is not as important to me. Oversimplified yes but if you keep the voltage and heat at reasonable levels everything lives.
To me is kinda unspoken. Heat and voltage are the cause of degradation. Motherboard and CPU are Siamese twins. GPU is not as important to me. Oversimplified yes but if you keep the voltage and heat at reasonable levels everything lives.
- Joined
- Oct 29, 2012
This. I have had to conceive some interesting cooling solutions to keep my mosfets from hitting egg-frying temps on my 3930k
Also why I keep the voltage as low as humanly possible
This. I have had to conceive some interesting cooling solutions to keep my mosfets from hitting egg-frying temps on my 3930k. Also why I keep the voltage as low as humanly possible
I realize on the x79 board, the mosfets can be burning hot too. reaching above 85'c (measured with probe). I was worried at a point untill I read that their operational temperature is up to 140'C, so I thought it was ok.. is there something to worry here that i do not know of?
hehe, when I started to OC, I know what I signed up for, to be honest, 11 months with such minor degradation, I should actually PRAISE this CPU for doing all the hard work for me. For me, it was worth it, as my work is calculation heavy (I am a physicists, so lots of models to run.) and I will say my computer has saved me WEEKS of time with its cal speed, when this chip dies, I will Frame it up.thank you 3930k, then it will be HELLO 3970X lol.
You would save a lot more time if you were starting to learn programming Cuda or OpenCL, or use apps which can use these technologies. For my work, I use a dual xeon E5-2687W with a tesla C2075 card. I can get over 500GFlops on this card, which, I'm sure is a lot more than you can get on your overcloked cpu. At home, I use a rig similar to yours, which is more oriented toward gaming.
- Joined
- Oct 29, 2012
Nah- I hear that these x79 boards die fast if you let em get above that 87~ mark, I just offer them extra cooling and they stay happy haha
I didn't know they were rated for 140 though, that's nuts!
I asked that in OCN previously, and that's what they told me judging on the model of the Voltage molecute on the ASUS rampage board. I was also told that older chips use to run at even higher voltages, and these mosfets are artifacts from those generations. like Pentium IV use to run at 1.75V...
http://en.wikipedia.org/wiki/List_of_Intel_Pentium_4_microprocessors
and cause of that, these days the voltage moledules easily can do lower voltages.
- Joined
- Jul 26, 2004
I ran a e6420 @ 1.65v for 2 years and a e8400 @ 1.52v for another two years (both were over the limit that ppl said was safe for 24/7 and they had no problems still the day I sold them
- Joined
- May 10, 2009
The MOSFETs used on all the SB-E boards I've seen didn't exist in the P4 era at all. They're self contained MOSFET/MOSFET/Driver ICs. While older chips did run much higher voltages, the MOSFETs used then and the MOSFETs used now share little in common.
- Joined
- Feb 13, 2010
- Location
- Missouri
The FETS on P4C/P4P and IC7's (P4 boards all) had a bad habit of going POP!
- Thread Starter
- #55
de-railing from the post a little, but I need to answer this one here.
You are right, and so I was told and I got 2 students looking into that as a 'next step' solution already. Currently our lab models and programs are all self-built and the software limits us from using Tesla calculations. We are testing this now only, but if after our test sample comes back with say a 20%+ improvement in calculation speed then it might be worth it to put more effort into reprogramming our software here for OpenCL.
That being said, our program now can use all 12 threads in our hex core, so my preliminary belief is that we can't be 'that off' the performance peak we were seeking.
Particle physics is ALL about math, and I keep saying we are approaching the ability 'to see the future' with our mathematics in quantum mechanics.
- Joined
- Nov 3, 2008
de-railing from the post a little, but I need to answer this one here.
You are right, and so I was told and I got 2 students looking into that as a 'next step' solution already. Currently our lab models and programs are all self-built and the software limits us from using Tesla calculations. We are testing this now only, but if after our test sample comes back with say a 20%+ improvement in calculation speed then it might be worth it to put more effort into reprogramming our software here for OpenCL.
That being said, our program now can use all 12 threads in our hex core, so my preliminary belief is that we can't be 'that off' the performance peak we were seeking.
Particle physics is ALL about math, and I keep saying we are approaching the ability 'to see the future' with our mathematics in quantum mechanics.
The God equation! Is it real?
By the way I am not talking about that crap that went around about proving God I am talking about perfect prediction formula. To see every possibility and then remove the BS.
- Joined
- Oct 29, 2012
Haha, shoot, -I- would, there's some crazy knowledgable people on these forums. I'm confident with what I know, what I can bench, etc, and I still feel like a 5 year old when I come read things on these forums sometimes
saw this article: just linking to an interesting read: http://www.extremetech.com/gaming/1...-pc-failures-results-shatter-enthusiast-myths
probably doesn't study many enthusiast rigs, however and crashing is different from permanent degradation- i imagine expert builds would likely have lower crash rates...but again, not siding.
probably doesn't study many enthusiast rigs, however and crashing is different from permanent degradation- i imagine expert builds would likely have lower crash rates...but again, not siding.
I skimmed it and reads like advertising and media hype. No references to data, etc.
