It occured to me that so far my SMTT testing is kind of matching up with the RB2 auto test. Before it started fritzing out anyway.
It usually measured the Olevia at ~40ms. It measured the Westinghouse at, well this is when it started fritzing, but initially at around ~20ms.
SMTT has measured the Olevia at ~20ms, but that's relative to the Westinghouse. So if the Westinghouse is around 20 (maybe around one frame?), that works out.
If this is correct, then the Phillips, which I rebought tonight and will test, should measure around 30-40 relative to the Westinghouse. Because it was measuring 50-60 by RB2.
If that's true, then also the Toshiba should have tested ~60ms by RB, since it tested ~40 by SMTT.
Of course I returned the RB2 guitar, and it started giving wild readings at the end anyway, when it would read at all.
Of course, that's also ignoring any scaling effects. The RB2 test was getting fed each display's native signal or close by the Xbox. Whereas in the SMTT tests sometimes I have to feed them at least slightly non native signals (1680X1050 for 1080P, etc). The Toshiba just put borders around the 1680 pic though, so I assumed it wasn't scaling. I'm guessing the Phillips will scale the image to fullscreen.