# A question from Computer Organization on Peak Clock Frequency

Given below are 3 different pipelined processors:

$P_1:\ 4\ stages\ with\ delays\ \ \ \ 0.6_{ms}\ \ 0.8_{ms}\ \ 0.6_{ms}\ \ 1.1_{ms}\\ P_2:\ 4\ stages\ with\ delays\ \ \ \ 2.0_{ms}\ \ 1.8_{ms}\ \ 2.0_{ms}\ \ 1.0_{ms}\\ P_3:\ 5\ stages\ with\ delays\ \ \ \ 1.0_{ms}\ \ 0.8_{ms}\ \ 1.0_{ms}\ \ 1.5_{ms}\ \ 1.5_{ms}$

One has to find the Peak Clock Frequencies of each $\mathbf{P_i}$. Pipeline buffer register latency is Zero.

I don't know how to solve this, but I've come up with an intuitive formula for calculating peak frequency of clock for processor $P_i$ as $$\mathcal{C_{p.f}}=\frac{1}{max(d_i)}$$ where $d_i$ is individual stage delay for processor $P_i$. For example for $P_2$ it's $\frac{1}{2}KHz = 0.5 KHz$.

Now I don't know whether it's the correct way to calculate peak frequency.
If it's not, can anyone tell me what is it?

P.S: I also don't know whether this is the right place to ask this question.
But I didn't have any idea about where to post it otherwise.

## 1 Answer

The peak possible frequency is, no doubt, equal to the inverse of maximum time supported by each stage. The stage consuming maximum time is the limiting condition and your anwer is correct.