As Cecil said don't get to wrapped up in the numbers from a random dyno. All you should be concerned about if you are going for the numbers is finding a dyno that is consistant and then work up from a base line(first run). The only thing a dyno is supposed to do is give a relatively close gauge as to the changes you make on any said set up. You can not get accurrate(sp?) info on whether you made a gain or loss if you use different dynos. They are all different no matter what, and there are just to many variables to try and factor in/out every little thing.