A thread over on Wildsnow got me digging back into the Off Piste data set. While in the past, I’d looked at weight per unit length vs girth, it’s also possible to plot against surface area (or surface density).
The ski surface area is computed with a formula which is both atrocious and practical. If M is a ski’s mass, S is the width of the ski shovel, W is the waist, T is the tail, and L is the length of a ski, then the density D plotted on the horizontal axis is
Be certain that you’ve used the right units (grams and centimeters, even for the width dimensions).
This treats the ski as two trapezoids, joined at the waist. This will tend to over-estimate the area of a ski, and hence underestimate its density. Relative comparisons of skis should be more accurate than the absolute density determined by this method. With densities determined this way, it would appear that anything in the 0.7-0.8 g/cm^2 realm is light by modern standards.
The Dynafit Cho Oyu’s claimed weight/dimensions would put it at 0.60 in these units (note that the pintail will pull the “trapezoidal density” down. Prototype skis from Ski Lab would land in the 0.64 realm.