help me OC i3-2100 bus speed

[intel's fan]

I wanna know how limit of i3-2100 bus speed when

1.air-cooled

2.no voltage adjusted

 

Could someone give the data & graphic

 

Although I had a i3-2100 but I scared destroyed it :P

[iskout]

It's rough, because so much is dependent on the bus speed. Without being able to adjust the multiplier, getting even a few hundred MHz out of it can become very dangerous to the longevity of the system.

[intel's fan]

It's rough, because so much is dependent on the bus speed. Without being able to adjust the multiplier, getting even a few hundred MHz out of it can become very dangerous to the longevity of the system.

 

oh oh I see ! thank your replly !

[The Smith]

There is no danger if you don't increase the voltages. So basically, just up the bus one MHz at a time, testing for stability in between.

 

Increased voltage is what damages transistors, much more than heat. There was a good article about this in the IEEE Spectrum a few months ago. They were talking about CPU odometers made to find out the overall wear of the transistors. With such a thing, you could buy a CPU at 4GHz, and then as time passes its frequency would be going down, according to the transistor wear. This would be much better than playing safe and selling it at say 3.4GHz just to make sure it will last what's it's supposed to.

 

That's why many people begin to experience stability problems even though their computer has been stable at that particular overclock for a quite long period of time; the transistors have degraded a bit and now they can't switch fast enough at that particular voltage and frequency. They then need to decrease the frequency a bit or increase the voltage a notch more to compensate for the wear.

[iskout]

There is no danger if you don't increase the voltages. So basically, just up the bus one MHz at a time, testing for stability in between.

 

Increased voltage is what damages transistors, much more than heat. There was a good article about this in the IEEE Spectrum a few months ago. They were talking about CPU odometers made to find out the overall wear of the transistors. With such a thing, you could buy a CPU at 4GHz, and then as time passes its frequency would be going down, according to the transistor wear. This would be much better than playing safe and selling it at say 3.4GHz just to make sure it will last what's it's supposed to.

 

That's why many people begin to experience stability problems even though their computer has been stable at that particular overclock for a quite long period of time; the transistors have degraded a bit and now they can't switch fast enough at that particular voltage and frequency. They then need to decrease the frequency a bit or increase the voltage a notch more to compensate for the wear.

There's also the issue of PCIE being linked to the BCLK, which can have some serious issues if increased beyond what it's normally stable at.

[intel's fan]

 

Increased voltage is what damages transistors, much more than heat. There was a good article about this in the IEEE Spectrum a few months ago.

 

Can you give me the website about this article?

[intel's fan]

There's also the issue of PCIE being linked to the BCLK, which can have some serious issues if increased beyond what it's normally stable at.

 

So,increase the BCLK could affect(positive or native ) the PCIE device such as video card?

[ccokeman]

With the core i3 2100 you are limited to about a 105 bclock on that chip. The 2100 is not a K-Sku chip so the clock multiplier is locked so you cannot overclock by raising the multiplier. So no big overclocking is possible. I just reviewed one of these type of CPUs so getting about 200MHz more out of it is all you will be able to get if your chip will run a 105+bclock. Many wont from what I have read. You can do all the overclocking you can without a voltage increase on this chip so no fear of killing it with voltage. If you really want to clock a Sandybridge platform you need to get a K-Sku chip.

 

 

 

 

Good luck!

[The Smith]

Can you give me the website about this article?

I originally read it on paper, but I found a copy here.