Reply to thread

Message: <blockquote data-quote="Anusha" data-source="post: 1467006" data-attributes="member: 828">There is a limit to the clock speed, and it is dependant on the following factors.1. Temperature = Higher the temperature, more the transistor leak is, and the correct clock levels (high and low) cannot to be achieved.2. Voltage = There is a defined voltage level for high/low values of a clock. For example, 5V for high and 0V low maybe theoretical, but actually the circuits identify high as 3.3V+ and low as 1.5V-. (These maybe not be the values used in a CPU) And There is something called a rise time, as in the clocks do not rise really sharply...They increase kinda exponentially (like a charging capacitor), not as a brick wall. So if the high/low values cannot be reached within the half cycle, things go haywire. This is the worst case, you always have to keep a safety zone so that it never happens. And also the clock shown on the pic is exxagerated to make you folks understand it.<img src="http://i177.photobucket.com/albums/w224/anushand/Untitled-11.jpg" alt="" class="fr-fic fr-dii fr-draggable " style="" />3. Process tech: 65nm or 45nm etc. I'm not entirely sure how this works though. Probably has something to do with tunneling (as in quantum physics; 25nm tech is supposed to give phenomenal yields...as I read somewhere) This is also the major factor in making higher clocked CPUs without endangering the components within the CPU.</blockquote>

Top Bottom