I will not defend the approach that is taken in regards to yield monitoring but I will not condemn it either. I could not condemn the approach unless I were serious enough about it that I were willing to switch crews because of it. I could also not condemn it unless I were first willing to spend a lot of time in the cab emphasizing how important it is to me that this is right. I also cannot speak highly enough about this crew in regards to their integrity, how they set their machines, the care they take and the lack of supervision required. We have a long standing relationship with them and they are excellent. They also do not charge anything in addition to their normal harvest fee for yield maps.

In regards to having each machine calibrated so that it reads exactly the same as the next one, I wonder how many people have actually accomplished that? When you have 6 machines in the same field I wonder if it might be a bit unrealistic to expect perfection in terms of yield maps?

This is what has been explained to me and it makes sense:

If you have two combines cutting side by side through the field and one combine says that pass in the field was making 250/acre on average and the combine next to it says it was making 150 per acre on average through the field, you can suspect something is wrong with the yield data from one or perhaps both of the machines. Now let's go a step further and say that you notice this pattern between the two machines no matter where you look in the field. How do we resolve this?

Pick a machine and call that machine the correct machine. Post process the data from the wrong machine so that it matches up with the correct machine. If those were the only two machines in the field, then compare the total bushels that came off the field with what the scale tickets say. Now adjust both machines equally to reflect the actual total bushels that came off the field.

With 6 machines this would be more of a chore than with 2, but the process would still be the same. I don't think it would be all that complex and I would think it would yield fairly dependable results. The only thing left after that would be removing areas where a combine was logging data even though it was not harvesting anything because another combine had already been there.

I really don't see why this wouldn't work and I don't see why it wouldn't produce pretty accurate results despite poor calibration procedures. I just don't know what tools I need to accomplish this task. When I get some time during normal business hours I will call Mapshots about it and maybe they'll be able to point me in the right direction, but I figured I would ask on here since there are some pretty smart people here as well.