On Mar 13, 2010, at 6:48 AM, Adrian Crum wrote: > > Understanding the difference between wall clock time and elapsed time is > especially important in manufacturing. They are completely different, and > manufacturing depends on both. > > The problem with the test is it advances the wall clock, then compares > elapsed time. They are not the same - the test is comparing apples to oranges. >
Adrian, I would like a suggestion for fixing the test. Here are some assumptions and details of the test: 1) the availability/capacity (calendar) of the fixed asset is the same every week: 8hours per day from 8am to 4pm from Monday to Friday 2) the goal of the test is to verify if the same exact task (i.e. a constant amount of time of usage for the fixed asset) when is started "tomorrow" (e.g. Friday this week) takes the same time to complete of when it is started the same day of the next week (e.g. Friday of the next week) 3) in order to determine the two dates (tomorrow and the same day next week), since I didn't want to hardcode the date, I used the trick of adding 1 day to "now" (this is the first date) and adding 8 days to "now" (this is the second date) 4) the above two dates are used as starting dates for the tasks and the ending dates are determined; the test asks that the differences between the ending date and starting date of first period and the ending date and starting date of the same period is the same Is this error caused when the DST is happening inside one of the time ranges computed in #4? This is what I understand... Any suggestion for fixing this (apart from hardcoding the dates)? Thanks! Jacopo
