So, today, I updated two of my oldest models, which pre-dated the per-part cost and sprued them together.
The algorythem siad they were now taking up more machine space (?), and though the part count was down, the price changed barely at all. I experiemented with not putting sprues on, and the difference in cost was minimal (like a 10-20p on a model around £20).
The lack of guidence here is not helping me, you, or our mutal customers; at the moment, today's experiment suggests that it is not WORTH me trying to sprue things together, because the pence I save isn't enough to pay for my time to actually be doing it.
So, genuine question: do you NOT want me/us to put multi-part models on sprues? Because it will certainly make my life a lot easier if I don't have to spend the time to do that. If the algorythm is set up to calculate the bounding box around each part, and not the whole model now such that you can, I don't know tessalate stuff into the gaps or whatever, I'm happy for to pay 20p more for me to do less work and you to do more - but it would be nice to have some sort of clue. Because trying to experiment to find out what configuration of any given model you would prefer me to use (as defined by your pricing) is wasting my time.
What, exactly, do you want us to do, guys and girls of Shapeways?
("Spend more than we need to" is not a valid option.)