Is my core speed actually jumping from 1000 to 4600 or are the readings inaccurate?

[Fornever]

I'm new to overclocking and have managed to OC my 6600k to a comfortable 4.6 GHZ stable. In Windows I have my power plan set to Balanced so that it's not running at max speed when it doesn't need to and only amps up when I play a game. I also have Intel Speedstep enabled in the BIOS respectively and all voltages except for the CPU Vcore set to Auto, with the Vcore set at around 1.3v.

However, when I see the MHZ fluctuating between 4600, 1000, 3500 and so on, is the CPU actually doing this or are monitoring programs generally inaccurate? I would like to think that it's staying at a constant 1000 or similar frequency, like it would stay at a constant 4600 if you set the power profile to High Performance in Control Panel. I even see the temperatures constantly changing from 20c to 37c and so forth - does this really happen or is it just the programs I'm using?

I've used a handful of programs although my favorite is CoreTemp. They're all pretty similar though so there's that.

 

[Ecky]

Your CPU is actually doing that. Frequency is adjusted in real time at sub-millisecond intervals based on demand, clocking up only as high as is necessary so as to save power and prolong the CPU's lifespan.

 

[Fornever]

Your CPU is actually doing that. Frequency is adjusted in real time at sub-millisecond intervals based on demand, clocking up only as high as is necessary so as to save power and prolong the CPU's lifespan.

Huh, then I wonder why it would burst occasionally at 4.6 when nothing demanding is being ran on it. Wouldn't a constantly changing frequency be taxing on the CPU itself? I could just change the power plan or even change the BIOS settings when I'm about to play a game and back, but that would be impractical.

 

[SinxarKnights]

Your CPU is actually doing that. Frequency is adjusted in real time at sub-millisecond intervals based on demand, clocking up only as high as is necessary so as to save power and prolong the CPU's lifespan.

Huh, then I wonder why it would burst occasionally at 4.6 when nothing demanding is being ran on it. Wouldn't a constantly changing frequency be taxing on the CPU itself? I could just change the power plan or even change the BIOS settings when I'm about to play a game and back, but that would be impractical.

No its designed for it, i've never heard of a CPU failing because of EST or Cool 'n' Quiet. I mean if you think about it, there are transistors switching billions of times per second and it can do that for decades on end without failing. Why would a few transistors switching only a few hundred times a second be taxing? It will boost to 4.6 to get things done in a timely manner so you have more idle time available.

There is some algorithm involved that controls it. I would guess it has a lot to do with timeslicing and prediction on a single core to calculate how much time it would take to complete a calculation and if increasing the speed would help save power.

 

[Fornever]

No its designed for it, i've never heard of a CPU failing because of EST or Cool 'n' Quiet. I mean if you think about it, there are transistors switching billions of times per second and it can do that for decades on end without failing. Why would a few transistors switching only a few hundred times a second be taxing? It will boost to 4.6 to get things done in a timely manner so you have more idle time available.

There is some algorithm involved that controls it. I would guess it has a lot to do with timeslicing and prediction on a single core to calculate how much time it would take to complete a calculation and if increasing the speed would help save power.

I see, that puts things in a much better perspective. Always good to know the science behind these kind of things. Thanks.

 

[SinxarKnights]

Well i'm not saying it is right. I'm just guessing based on what I know about CPUs in general.