Originally Posted by
Spoonrobot
Casing stiffness has a huge effect on drum roller resistance because the tire deforms so much around the roller both on a lateral axis and a horizontal axis. I suspect it would have less effect on the road but would be very difficult to measure.
If it's significant, it could possibly be tested by comparing differently-designed tires on a tiny drum and a huge drum, and seeing if their relative performance changes.
Intuitively, it seems to me that the drums would underemphasize radial suppleness relative to tread/axial suppleness. The sidewalls smush down and outward whether a tire is on the road or on a drum, but tread rubber sits a lot flatter fore-aft on a road than it does on a small drum.
I'm also unconvinced of the efficacy of testing tires with any tread that protrudes more than mere file tread on rollers.
I can see several potential issues once you throw knobs and off-pavement riding into the mix.
First, the firmness of the riding surface will affect how much knobs actually flex/deflect/whatever off of it. A knob that bends against pavement may just sink into a soft or loose surface, which probably has implications for the amount of energy wasted.
Second, large knobs at the centerline can probably have "suspension"-like issues on firm surfaces. I mean, if you ride really aggressive MTB tires on pavement, you can feel the bicycle rumbling from the tires. This effect, if significant, is probably one of those things that demands field testing to characterize.
Third, on loose surfaces, the power required to plow through the surface might matter.