Improving the Effectiveness of Repetitive Testing
Nobody likes to repeat themselves. I know I don’t. Repetition is BORING!!! And when we do repetitive testing, we tend to zone-out.
We stop paying attention to the very details that we are seeking to understand… …this is a problem.
So how do we increase the effectiveness of our repetitive testing?
The answer is through automation!
You’re probably thinking:
“Automation?!? That takes a lot of effort and is expensive, isn’t it?”
Automation doesn’t need to be expensive, nor does it need to be complicated, though you can make it as complex and expensive as your business case warrants.
We try to keep our automation simple and cheap. After-all, I don’t want to spend money on something unnecessarily, and I imagine nor do you.
Instead, we use tools like the DataDyno, and simple test harnesses, to automate our repetitive tests “just enough” to derive the benefits of automated repetitive testing while keeping the infrastructure simple.
To do this, we need:
- A DataDyno
- Breakout boards and wire
- Sample hardware
Yup, that’s it! No special software, no specialized hardware. Just those few things enable us to run many of the repetitive test cases we need to in order to test such things as:
- Repeated power-on/off cycles
- Repeated reconnection/disconnection events
- Repeated button presses
And the list goes on…
When planning such tests, however, you need to bear a few things in-mind:
- If timing is critical to your application, then make sure you vary the timing parameters during each iteration. You can do this by sweeping the time from either end. With more sophisticated systems, you can even use a random function to vary the timing (though this is not available with the DataDyno yet).
- If voltage is critical to your application, then make sure you vary the voltage.
- Understand your dependent and independent variables BEFORE creating your scripts.
- Remember the value of statistical significance. In other words, don’t just run something 5 or 10 times once you’ve automated it. Instead, consider running the script or repetitive test 30 or more times.
If you bear these principles in mind, your repetitive testing will not only yield more accurate results, you’ll also be able to defend your conclusions.