Ethan:
We use the Bonello criteria which is purely Axial based, and also consider oblique and tangential modes when doing simulations. Typically (in the texts) tangential modes are weighed at 50% significance (I believe) compared to Axial and oblique at 25%....
In practice, we have found that once you get past the first two axial modes, the build up of "other" modes becomes very helpful in filling in the energy. They certainly are significant in the transitional zone above 125Hz or so and below 300....
that's how we look at it..
Chris:
ALL rooms have modes, boxes, trapezoids, eight sided, curved, etc.. They all have modes. They are just much more difficult to predict in non-rectangular environments which is one of the reasons we often start with rectangular shells and build the treatments "inside" the box to look more like studio geometry.....
Of course, we also design many hard shells with non-parallel geometry when it best serves the design requirements such as layout, sight lines or arrangment of adjacent spaces... or when the roof is angled in the host structure (and all of the above).. then we do more intensive simulations studying min, max and average dimensions....I've also developed a bit of a feel for these non-parallel designs through hundreds of "experiments" ... oops I mean projects... so I know what has worked and can kinda reverse engineer why. Yes, you still have modes and these rooms still need (maybe even more) broadband bass traps... okay?