Interesting to see one of my posts pop up here. (Someone posted a link to this on TGR).
It's very easy to check torsional rigidity yourself with Noodler's method. The results might not be measurably accurate, but you can get a good relative ranking if you do the tests all at the same time.
There are two main things that affect torsional rigidity in a ski binding. The first is play in the boot/binding or binding/ski interface; the second is elasticity in the release mechanism itself.
"Freeflex" binding systems, like Atomic, Marker Motion, and Tyrolia Railflex all have some play in the system that allows the binding to slide up and down the rails. Even regular heelpieces also have some play, because they are made of two distinct pieces: the track and the heelpiece itself. There can also be play if your toe height is not properly adjusted.
Play is bad, but generally not as bad as flex unless there is a lot of it -- because once the play is taken up by pressure, the boot won't move any farther.
Flex, on the other hand, comes from the release mechanism or the binding itself bending under pressure, which means the harder you edge, the more your ski will do something other than you intended. My post above explains how a Salomon Spheric binding flexes under pressure (with the ugly side effect of making your toe blow out more easily).
Tyrolias have some play, as Noodler observed, because the diagonal release toe can flex upward to allow, well, diagonal release. As you'd expect, this flex is proportional to DIN, because it's part of the release mechanism, not the binding itself flexing. (Some Atomic bindings allow you to lock this feature out.)
Looks have diagonal release on the toe, but the resistance curve is much steeper: they are much stiffer in diagonal release compared to Tyrolia. Their toepieces have almost no flex at all. Unfortunately, the heelpiece has some play and some flex: the heel track gets slightly bent over time, increasing play, and the pivot itself has some flex to it. This is why the 2006 bindings no longer have the pivot heel. Lots of people mourn the demise of the Pivot heel, but I am likely to buy a pair and sell my Pivots -- if they work like the design says they should.
It's all reasonably common-sense stuff that you can test, as Noodler did, by putting a boot in and wiggling it around. And it gets more important the fatter your skis are: a 66mm slalom ski is basically straight under your boot, and doesn't see nearly the torsional force that a 98mm powder ski does.
I'm convinced that a lot of people's thinking that "powder skis suck on hardpack" has mostly to do with the poor torsional rigidity of their bindings. I have no skis narrower than 92mm in the waist, and I have no problems holding an edge on icy days -- but I run Look Pivots, nearly the stiffest bindings available (the Salomon 916/920 may be stiffer, but I don't have one to test.) On the opposite end, I've skied a pair of powder boards with demo track Salomon S900s, and they are truly dangerous -- I have no edge control on anything harder than afternoon crud on a powder day. If I put Salomon S912s on my Nordicas, I'd think they sucked on hardpack too!