I was advised here to draw in full size and before printing to make a copy, scale it down and then email to the printers without touching it any further, advice which I found quite good and solved a number of problems. The intent is to create masters for later 3d printing in smaller scale.
The subject of my screen captures in green is a Messerschmitt Me109, drawn full size. Is there any way to prevent or at least minimize this? The consequence is that if I now wanted to draw an edge using this point as an End, the system might pick any point in the region, and it’s a major irritant having to go back and trying to instill some discipline on a drawing every so often, while being unsure whether this redrawing work will be definitive or not, likely not. So as I add further geometry using the inferences, or using geometry that is just passing close by, the distortion increases. The original edges were that fantail on the right all nicely gathered on a single point, which I took as the origin for a number of further geometry, using the knot as an End inference. The horizontal edge was a single line originally but it became split and dislocated, and a lot of secondary edges were added up by the system. Here is the detail view of the knot at bottom. In the image below the near horizontal edge is a pretty single line, right? Also the near vertical edge looks a bit murky. I made my model from scratch entirely, no imports.
