I agree with you that an interchange of ideas and analysis is what these boards are for and can be quite interesting so long as there is an insightful analysis to support the conclusions reached.
You seem to have a need to take the amount of powder produced by APC and extrapolate it to a conclusion re how many arcam printers will be neeed. For the reasons I've posted, I find this exercise flawed and meaningless. One cannot take the amount of powder produced, guess on the amount of powder each arcam machine will use (which you have done in numerous posts and I've explained why I disagree) and then infer how many arcam printers will be needed. I've explained my reasoning and you have admitted there most if not all of your variables are merely guesses and as such suspect.
But I've given my analysis and you've given yours and it will be up to readers to determine if your guesses have any merit. To me, i'm not persuaded and I've explained why. Especially since you admit all your data points are guesses and not based in fact and that many variables also change with time.
We both have given our reasoning and analysis and readers can decide. But I agree with you, the process is interesting and useful as an exercise to try to determine how many arcam printers will be needed. BUt the real key, IMHO, will be how soon and to what extent EBM volume manufacturing proceeds and we must await that playing out. To me all these posts attempting to determine how much powder is used by ebm printers, how many will be needed to produce a certain number of parts, even the amount of parts to be produced are unknown facts and thus any conclusions based on such guesses is flawed. We don't really even know how effective and prevalent stacking is or how different companies differ in their efficiency of producton.