The choice of which instrument to use depends on many factors such as history of use in your company (such as having a legacy of prior measurements because you can’t readily translate readings or scales from one device to another), whether the instrument is to be mill duty (such as the original Schmidt Hammer) or more sensitive but at the same time more delicate, ease of use, price and many others.
However, the number one priority should be resolution. A poor instrument or application is no better than a random number generator. So what affects the resolution of any particular device and application? A number of factors such as whether you are measuring on a parent or finished roll (surprisingly, the looser wound parent roll may sometimes give better results), roll geometries (diameter and width), who is holding the instrument (depends on care, experience and other factors). Not surprisingly, the hardness of the roll affects the goodness of the roll hardness measurement. Tissue (nonwovens, textiles etc) and board (foils etc) may at the ends of the scales of these instruments and may not result in good resolution. Also, for reasons not yet clear, results on film are not as predictive (of roll defects) as they are on paper where hardness almost always works (even better than most test labs and million dollar scanners for profile applications).
I answered this question of how to ‘test the tester’ back in 1988 and the technique is as valid then as it is today. It is simple; you compare the resolution of instruments (testers, grades etc) by taking a roll with a known upset and using the Z-test to determine which instrument (etc) can resolve a smaller value for a given confidence. A known upset is easy to make in a wound roll by simply turning up all the TNT’s tightness knobs on the winder on the bottom half of a roll and then suddenly turning them in the loose direction for the top half. Compare the standard deviation of hardness across the width two inches above and below this step change. Alternatively, you can take a roll with a known violent profile variation and compare the variation of the hardness measurements at a gage band and at a gage valley. Details and demonstration of application and results are given in the many places where you How to Measure Roll Quality has been published.
So, try before you buy, or at least test before you test. At least one of the companies, TMI, will loan or rent a device for your evaluation. This seems fair since there is no way of knowing for sure (especially for non-paper applications), whether these best of the best roll quality measurements is even any good. Even if it is not sensitive to defects on one grade, do not give up yet, you should try it on another because that grade may perform better.