That was essentially an advertizement for software. The G-tech is a FAR more powerful piece of equipment. The readings on a G-tech are highly dependant to the parameters entered into it, meaning that their numbers are off most likely because the setup is not perfect. But realistically, the function of a G-tech is not to tell your 1/4 mile time down to the hundredth (that's what the track's equipment is for) or for telling you exactly how much power you have, but to tell you variances run to run. In fact, to do the method described in this thread you need the more powerful G-tech Pro RR, which has logging and download capabilities. What they were using in that test is the SS model, which has no logging capability.