It's about damn time. For about 13 years I've been thrashing around trying to make an accurate drawing of Fort Pickens. I could have bought reproductions of some of the plans--but where's the fun in that? No, you don't get to explore the place if you're just copying someone else's drawing.
I spent a lot of time at the fort with a tape measure and some other improvised measuring tools, sketching and scrawling and crawling all over the place inside and out. For something like individual rooms (casemates), drawing the layout is simple: kind of hard to mess up a rectangular room.
But when all five of the curtain walls and all five bastions are brought into it, things get tricky. At this scale, I'm dealing with 25 brick walls, each at a non-square angle to its neighbors. Because they're brick, you can't really get a true reading on those angles because no given brick is aligned precisely. Measure at different heights in a corner and the difference can be several degrees--and even a small error can become several feet when you're drawing a 350-foot long wall.
I made several attempts over the years and each one came out wrong, so I just put it all aside and stopped messing with it for awhile.
A week ago, I shelled out for an inexpensive 2D CAD program called QCad, finally ready to throw down and get the Fort Pickens project going again, if only for something to do other than reading news blogs and sleeping.
The first draft came out almost perfect! I changed my approach, using a blown-up printout of an 1831 sketch to get those aggravating angles and just plugging in wall lengths from the sketches...
...and the resulting CAD drawing came out only 2'9" off (the corner lines at top above didn't quite meet). So damn close, and so much better than any previous attempt. After a few more hours of troubleshooting the measurements and tweaking the drawing, it's close enough to perfect that I'm just going to stop polishing and get on to adding the innards.
This is pretty much the key to getting all the other views of the fort, from sectionals and cutaways to simple elevations, because everything else depends on this top plan being right. As a bonus, once I've gotten some of the main layers drawn (top plan, horizontal section) I can run it over to Blender for 3D modeling and scenery building.