Track CPU usage by PID over time

#!/bin/bash
# --- Version history ---
#   0.4: added variable to store file path, and $2 for base file name
#        added variable to store desired reporting interval
#   0.3: added $1 to send in process ID at run time.
#   0.2: switched to $SECONDS for the loop. works.
#   0.1: didn't work well at all.
# --- Version history ---

# Usage: cputrack [PID] [filename]
#        replace [PID] with process ID #
#        replace [filename] with base file name to use (no extension)

filepath=/Users/your_user/Desktop   # modify as desired
interval=20                         # reports per minute
timelimit=6000                      # how long to run, in seconds

mydate=`date "+%H:%M:%S"`           # the timestamp
freq=$((60/$interval))              # for sleep function

while [ "$SECONDS" -le "$timelimit" ] ; do
  ps -p$1 -opid -opcpu -ocomm -c | grep $1 | sed "s/^/$mydate /" >> $filepath/$2.txt
  sleep 3
  mydate=`date "+%H:%M:%S"`
done

Hey, nice script. I have a couple of questions:

1. Could you modify this to monitor memory usage instead of CPU?
2. Could you make a version for a specific function call, so that the usage would be:
   memorytrack "myexec myarg1 myarg2" [filename]

I guess the question I really have is how to get a PID automatically from a command-line call. =)

Shouldn't the "sleep" call be for $freq and not 3?

Here's my fork: https://gist.github.com/mrnohr/ba7bf61c3f91c071f08d