Something not really spelled out here is that the relative measuring tools are often more precise if you’ve a) got to make something to fit something else that already exists, b) you’re creating something that has a repeated feature in it, or c) you’re creating things that only need to fit one another (but that have to fit perfectly). In all of these cases, the measurement (i.e. the number of units) is essentially irrelevant (or relevant only at the first step).
For example, when cutting baseboards for radically irregular corners (such as the allegedly ‘square’ corners that exist in my place…), it’s much more accurate to use a tool made for transferring angles to set up the mitre saw than it is to measure with a protractor and attempt to set the saw to e.g. 87 7/8° (trying to measure fractional degrees on something like a baseboard means you’ll have errors while measuring, and again while setting up the saw…)
And for repeated features such as a series of holes, you might do something like:
- scribe a line that all the holes will sit on
- measure and mark the location of the first hole
- set a pair of dividers to the required distance between the holes
- use the dividers to mark the distance on the work
Building on the preceding example, if you were making a precisely matched mating part with some repeated feature (a set of studs, slots, larger or smaller holes etc) that matched the set of holes from the last part, you could use the same, pre-set dividers to mark the locations on this part too. Using a scale risks introducing errors at each new hole/feature on each part.
The same holds not only for wood or metal working, but for things like bookbinding as well–when making books or cases for books, I tend to use a scale at the outset, but once the rudimentary dimensions are set up, the scale only gets used as a straight-edge, and my tools consist mainly of dividers, punches, knives and a square.
