I've become increasingly interested in watch movements, in understanding the attributes of a good movement and in understanding how watch movements can be compared.

I can intuitively appreciate that attributes of a good movement likely include robustness across a range of conditions, parsimony in mechanical design, integrity of materials and construction as well as efficiency. This basic hypothesis has led me to dig around to better understand how the efficiency in particular of a movement is calculated/demonstrated. I appreciate from other areas of my life that the efficiency of an energy transformation is often assessed on an input-to-output basis (e.g. x units of energy input is transformed to y units of useful energy output + z units of non-useful energy such as losses). I happened upon the article linked below which I will read soon - I hope it will give me a better theoretical understanding: http://www.nawcc-index.net/Articles/Headrick-EscMechanics.pdf.

In the meantime, I was wondering whether any of you might have ever measured the efficiency of your own movements and how you went about doing it (i.e. in a non-lab setting). I suppose power reserve can be a practical approximation for the denominator (i.e. for useful energy output), but how to approximate energy input, especially on a non-manual wind movement?

Please feel free to describe your own experiences in measuring the performance of your own movements and/or please feel free to share your working criteria for what makes a "good" movement and how different movements could be compared objectively. Thanks much for listening.