# FSB and actual clock speed?

Tags:
Last response: in CPUs
Share

I know this is a dumb question and I apologise about that, I just need to know the difference between these two. People say that when they overclock, lets say their E6300, they overclock the clock speed, which is at 266 (I assume?). So the FSB of that is 266 X 4 = 1066? And the actual speed of the processor is 266 X 7 = 1862 Mhz? So why is the multiplier of the FSB 4? Why did Intel implement FSB in their processors? And why does it have a multiplier?

Ty

More about : fsb actual clock speed

The 266x4 is the total FSB speed. (cpu -> memory wise)

To figure out the CPU speed at stock:

266x (multiplier) 7 = 1.8 Ghz = E6300

But to overclock you change that 266.. so if you bump it to 280:

280x(multiplier) 7 = 1.960 Ghz = changes the speed of the CPU.

Since the Memory is quad pumped... 280x4= 1120 mhz rated FSB.

It is not really a multiplier... the bus runs at 266 MHz and not faster. The bus does double data rate transfers, meaning that it sends data on both rising and falling edges of the clock signal. Also the memory can be arranged in "dual channel" mode, meaning the bus will be twice as wide. Thats where the 4x comes from... 2x from DDR and thats multiplied by 2x from the wider bus.
Related resources

Quote:
It is not really a multiplier... the bus runs at 266 MHz and not faster. The bus does double data rate transfers, meaning that it sends data on both rising and falling edges of the clock signal. Also the memory can be arranged in "dual channel" mode, meaning the bus will be twice as wide. Thats where the 4x comes from... 2x from DDR and thats multiplied by 2x from the wider bus.

This is true --- this is also why you will see, in more technical literature the reference to bandwidth not as MHz but as MT/sec or megatransactions per second, as data is carried on both the rising and falling edge of the clock. But for the most part, errantly in the strictest sense, it is often refered to as quad pumped or multiplied by 4.

Jack

So if the memory is not running in dual channel mode...it'll be multiplied by 2 instead of 4? That part is still a bit confusing to me. So basically, if you need to overclock a cpu and it's multiplier is locked, all you do is increase the clock speed and it'll increase the FSB of the CPU and it's speed? Or do you have to change them separately? I know this is an overclocking question but it'll help in explaining this to me.

Quote:

So if the memory is not running in dual channel mode...it'll be multiplied by 2 instead of 4?

So in a way, yes, the FSB will still run at its clock speed (266 MHz), that wont change, but in theory you can get twice the data through with dual channel. The "4", or "2" for that matter, should not be treated as clock multiplier though since the bus speed does not change.

Quote:

That part is still a bit confusing to me. So basically, if you need to overclock a cpu and it's multiplier is locked, all you do is increase the clock speed and it'll increase the FSB of the CPU and it's speed? Or do you have to change them separately?

If the cpu's multiplier is locked you can probably increase the system clock to which both the CPU and FSB are tied to. So both will be overclocked at the same time.
Overclocking if not done right might lead to permanent damage on overclocked components. Even if the system is stable, the speed at which the components "degrade" will likely be increased, so be careful.

Actually I agree that nowadays overclocking is quite safe  , I've done it myself on a few systems, I just wanted to scare him a little so he would be careful  .

Can I throw a monkey wrench into the whole works and discuss the effect that dropping the multiplier on a 965 chipset has on the northbridge fsb?

Sorry.....couldn't help myself

Nah... I'll leave the mind-f**king info for a later date. Let him wrap his head around the intro-info first. :twisted:
Related resources:
!