When it comes to testing, it is something that we all generally expect to be done on the things we use every day, so that the people who make the people who make the equipment know the limits of the product they are making so that the consumer in turn knows that they are using something that will not, sometimes literally, blow up in their face. But with this said, how much testing is enough testing? This is because obviously we cannot test everything any and we cannot test them endlessly. Where do we draw the line?

When it comes to testing, repetition is the key for better results. If we can test a product repeatedly over and over and if we can get the same result over and over in each and every test, then we are getting the better results. And that goes on to the logical conclusion that the more we test the greater the reliability. But how to achieve this repetition and where to draw the line so that the test does not become everything.

The answer mainly lie with robotics and with computers and equipment that can deliver the same output all the time with very little margin of error. This is where test socket or robots become very useful, because these can deliver a constant output all the time so that the item being tested will be the same standard.

The other problem with testing is knowing where to stop, the problem is that you can only test something so many times before it becomes too expensive and pointless. Other times, you cannot test a device because they are meant to work only once, and testing it means you cannot reuse it. This is when a sample is needed. Again here, the sample has to be sufficient and random enough that you can say that a batch of equipment is working fine with a certain degree of confidence.

Ultimately that decision comes from testing too. A sample size or a the number of tests can be determined by previous test results so that an educated value and limits can be set in a realistic manner so that the manufacture can be sure that the test results are satisfactory or representative of the larger batch. This magic number is the line that separates excessive costs and the effectiveness of a product. The cost of moving this line too much to either side is that you either waste money, or lose the reliability of the product where either way the business will fail. So that magic value is a very, very important value in testing and it is called the threshold.