The startTmr() command returns the number of seconds that passed since 00:00 GMT of January 1, 1997 — as far as the calculator knows, anyway, since this value is adjusted every time the current time and date changes with setTime() or setDate(). If the clock is running, this number is also updated every second, which is how the calculator keeps track of time.
Together with checkTmr(), startTmr() can be used to measure off a time interval (in seconds) while a program is running (make sure to use ClockOn first). the name of the commands reflects their use: you can think of a startTmr() call as creating and starting a timer:
The checkTmr() command will then return the number of seconds that have elapsed on the timer (without stopping it):
:Disp "Seconds elapsed:",checkTmr(timer)
This is a good abstraction and you don't need to know the details of how startTmr() and checkTmr() work to use them. In reality, checkTmr(x) returns startTmr()-x, so using it on the result of startTmr() gives a time difference.
Because both startTmr() and checkTmr() deal with whole numbers of seconds, the resulting difference in time could be off by up to a second in either direction. That is, if checkTmr() gives 15 seconds as the time, you know the time that actually passed is between 14 and 16 seconds.
The startTmr() and checkTmr() commands can be used to figure out how much time a command or routine takes with much greater precisions by running it multiple times. For example:
:startTmr()→t :For i,1,1000 : somecmd() :EndFor :Disp checkTmr(t)
Suppose that the result displayed was 100 seconds. This is accurate to 1 second, so the actual time was between 99 and 101 seconds. However, this actual time is for 1000 repetitions of somecmd() (we assume that the time the code to increment i takes is negligible, although that, too, may be taken into account). So somecmd() repeated only once takes between 99/1000 and 101/1000 seconds, so the actual time is 100 ms, measured to within 1 millisecond error.
See Code Timings for more information on this, as well as the results of some common comparisons.