Disagree. Dave and all vendors are (and rightfully should be) completely free to overdeliver on what they promise. Such achievements are very good P.R., as they further differentiate between themselves and other companies which underdeliver (if at all). They demonstrate their ability to innovate and improve, and increases confidence.

You are proposing, however to extract punishment from (or compensate) one group or another based on later success. That's just ass-backwards. Penalizing success? exceeding goals? What exactly are you trying to motivate here?

Case in point - initial tests showed H-boards not hashing at their advertised 25GH speed. So additional H-boards were included for free to the initial customers -- and likely only them, as later batches should have tweaks made to achieve the advertise rate per board. This meant people who ordered a 25GH starter kit are now sitting on 40GH of hashpower.

Now, according to your logic, this is an unfair bonus given to one group (early orders) that later orders do not receive. Should Dave now "compensate" later orders that won't receive such extra hashpower with $$ to offset the difference? Didn't those later orders make their choices based upon calculations which weighed the cost of 25GH in august vs. 25GH in october? Have they not been wronged?? Hogwash.

Shipping october orders any earlier than the 31st would NOT cause to "harm" to earlier orders. They got what they paid for, and they worked with what knowledge was available at the time of order, like everyone else. You could even make a case that shipping Oct earlier would actually be more than fair because of the steep increase in difficulty rendering earlier ROI predictions to be way off, so the earlier delivery would be a move to compensate (as other companies have attempted various schemes to preserve estimated ROI).

Dave - ignore this guy and send everything out as soon as humanly possible.