Finding a shop with an AWD dyno is hard enough, I suspect finding one with well trained operators would equate to finding a unicorn.
These numbers for virtual dyno have been all over the place depending on the logs. If I remove smoothing then the numbers, especially down low where all the torque hits, gets to scary territory. So for now I'm pretty skeptical of this program. I think we've got some work to do eliminating variables (weight, same pids/dmrs, etc).
I haven't kept up with where they are at with the process but SAE has proposed that all ECM's implement torque reporting that has to match a dyno reading so hopefully this will become much easier.
With our torque based ECU, it has to calculate torque accurately, the challenge is figuring out which PID is the data we are looking for.
There is also a vehicle acceleration rate pid which I think might generate some good comparisons if more of us utilized it.