I currently have HW monitor. As u said i downloaded speedfan. Wehn i opened both speed fan & HW monitor i saw these things. 1. The CPU temp in speed fan was around -40C 2. The coretemps in all 4 cores in speedfan is always 10C less than the core temps in HW monitor. Why? Here is a screenshot
I checked realtemp but cant find anything which shows The HDD, RAM, GPU, Northbridge, Southbridge temps. It only shows core temps
realtemp will be the best for reporting CPU temps, and oddly GPU temps. Just use speedfan for everything else
i cant find any label showing northbridge temps on speedfan And another prob. I usually got this prob a couple of times. When i OC i usually boot into windows. After about 5 seconds everything just freezes. The cursor wont move. Cant open taskmanager. Nothing. It just freezes. Is this caused because the CPU voltage is not enough or is it because the CPU voltage is too much? And is it good to lap the CPU heatsink? Will it reduce the temps a lot? If so by how many degrees C?
usually when something like that freezes on boot, its needing a voltage increase, more voltage only generates more heat, it doesn't cause instabilities. As for lapping, depending on the heatsink surface to begin with you can drop temps by 2-3 degrees C, and it you lap the CPU so the two mating surfaces have the same finish than a further 2-3 degree C can be dropped. lapping is really only for those who know what they are doing, and before you ask how to do it there are millions of guides out there on google.
When we increase CPU voltage when OCin, the CPU temps will increase, Right? Is this increment in temp the warning that the CPU heatsink cannot handle the overclock with increased voltage? And i got to know that modern CPU's have some protection mechanism which will shut down the PC if the temp increase. which CPU has this mechanism? And at which limit does it shut down automatically? And if it shuts down to protect the CPU y do ppl claim that they burn their CPU's by Overclocking when the temp rise?
its not built into the CPU, its a built into the motherboard, and as the diode that makes contact with the back of the CPU is only a estimate of the CPU temp, things can and do go pop. Also pushing the speed up causing the CPU to draw more current, this puts a lot of strain upon the motherboard, and sometimes the power management such as mosfets and VRM's can go pop, sending a spike into the CPU. Auto shut down usually isn't set automatically, i have always had to set it manually in the bios. Limits depend on CPU. Your Q6600 shouldn't be taken past 95 degrees C i believe, but i would never let it get above 70 degrees C under load. My Q6600 i never let get above 50, because the cooler i run it, the longer i hope it will last.
thx for the info. then what is the warning that we can see to understand that the CPU voltage is too much? Is it the increase of temp or is there another warning? If so does it mean that the only way to understand that im using too much voltage on the CPU is to check the temps? Also even if i lap my CPU heatsink i am not planning to lap my CPU because i have 2 more years of warranty left for it.
pretty much, just monitor temps, for a Q6600 i would seriously not take it over 1.6v at that voltage requirement inorder to get a stable overclock will only result in a very short lived CPU. Some guys will go higher, but they are benchmarking with some serious cooling and if it goes pop they don't care. Overclocking has way too easy tbh, i remember when i started out about a decade ago, much different then!
I tried increasing the NB voltage but i still get the internet problem. I OCed as 9x334 to get 3.0Ghz. Still have the internet problem. Then i tried 8x300 but i still get the problem. I tried increasing the Northbridge voltage from 1.100 to 1.325. But still no use. After overclocking i changed back to stock speeds. Then when i boot into windows i had the same problem. So i shut downed the PC and switched ON after about 10 min. Then it started working. How? What else can i do to correct this?
sounds like the PCI bridge isn't locking at its default speed so during an overclock it is creeping above its 100mhz speed, and causing the ethernet controller to fail. nothing you can do about that, other than changing the motherboard.
Most people have said that you need to increase the NB voltage. But at what point should you do that? How do you know when the NB needs more voltage? And when overclocking the CPU i found that the RAM is also being overclocked. On CPU-Z i found that the CPU is to DRAM ratio is set to 1 is to 2. Does the RAM need a voltage increment when OCing? If so when? When the computer crash during overclocking how do i know whether it crashed because the CPU needs more voltage, Northbridge needs more voltage, RAM needs more voltage or some other problem? I also need some help with the method of overclocking. Which method is better? Firstly increase the Vcore to a certain level, then keep increasing the FSB until it crash. Then increase the Vcore again and keep doing the same thing. Increase the Vcore little by little along with the FSB. If these are not good methods, what is the best method that most overclockers use?
everything you need to know is here, if you spend some time to read it YOU will find the information.
