I have been designing a multi-level layout for a 30 x 14 ish space. I have made designs in SCARM and then built the designs on the floor as a sort of prototype and have been surprised to find a lot of variation between the floor and SCARM when I think I am following SCARM precisely.
I am using tubular track and this may be part of the issue. I have been in the hobby a long time and have a large inventory of tubular, mostly K-line but of course also Lionel. The current layout has O72 and O54. Some of the older tubular stuff does not go together perfectly tightly. So, if there is an extra 16th in one joint, a 32nd in another, they add up over 30 feet. However, this issue does not explain all the discrepancies I see. One can take a half circle of O54 and open and close it from O52 to O56 (or more or less) with little noticeable visual difference and it all works. Actual tubular curves are very squishy (technical term). Even this does not explain it all.
I am getting really bogged down with trying to reconcile SCARM to what is actually built on the floor. I have spent too much time on this, but I have instances where a piece of track on the floor is two inches shorter than SCARM says it is. One gets drawn into tying to understand where these errors are coming from. I’ve tracked down the worst cases at some time expense, but I have a couple of situations where I can’t figure out the problem. I have measured the layout many times. At some point I live with discrepancy in order to move ahead.
Building using the reconciliation process results in better track geometry on the floor. I have built dozens of floor layouts over a lifetime and looking back, many were very sloppy. Maybe this does not happen with track systems that snap together more positively.
Bill