I read the thread and heard the theory that chuffing was due to friction with the walls of the slot.
Thus it makes sense to look at the ratio of area of the opening to its perimeter, for a constant length, and normalized to a height of 1
This is another way of saying cross-sectional area to surface area
For a circle, I get pi r squared/ 2pi r = 1/2
for a square, I get r squared / 4 r = 1/4
For a slot, I get N/(2N+2), where N is the aspect ratio
for N = 1, 2, 3, 4 ... I get:
.25, .33, .375, .4 heading towards .5 as an asymptote
That a narrow slot tends towards the same ratio of perimeter to area of a circle is counter-intuitive until/unless you recall the derivation of the formulas for area and perimeters of a circle. These involve dividing the circle into wedges, flipping every other one over end for end, laying them out side by side and then taking the limit as the number of wedges approaches infinity.
If chuffing results from some sidewall related effect, than slot aspect ratio matters a little, for low aspect ratios, but not much.
I would expect things to get progressively worse as the aspect ratio increased and that isn't the case for this metric so I don't think this ratio is a good indicator of slot performance.
Perhaps slot aspect ratio doesn't matter until the slot gets so narrow it starts to show some resistance effects. This would be highly non-linear, no doubt.