Hooo. The problem with swaging is it, too, needs to be done right. Does your local wire rope shop have a proper roller swager? Some West Marines have one, but I sometimes wonder how well-trained the people operating the machine are. The thing about swages, of course, is that they only as strong as the initial crimp. That is, they don't get tighter as more force is applied -- unlike a rope splice or compression fitting. Actually, they get looser, as the wire inside the swage elongates and thins, a process called 'necking.' So a swage has to be absolutely within tolerance to achieve its intended strength. Hammer or crimp swaging is not as strong as roller swaging.
(People get real upset that Hayn doesn't even supply torque values for their compression fittings; instructions just say "tighten until snug." Hayn replies that initial torque is not critical, because the wedge-like geometry of the cone ensures the fitting will find its own proper compression.)
Ja, good idea with the spreadsheet. Find out how much each swage will cost, then put that against compressions. Nice thing about the SJ21 is that it doesn't have that many fittings to begin with.
You may find that subtracting the materials and labor of swaged ends makes DIY compressions cost-equivalent. You could also think about swages at the hounds, and compression fittings down low. Doing your own riveting is easy. Remember to goop on a galvanic inhibitor if using SS rivets. And I like laying down a piece of electrical tape under any SS item I bolt, screw, or rivet to the stick. Pls do share your decisions and the cost of refitting a SJ21. I'm eager to know what this will ding us in March.
You going with different turnbuckles?
ETA: You should talk to Rig-Rite. They have a reputation for good prices, service, and quality; and they can advise you on swage vs. compression fittings. Kinda grumpy some mornings, but aren't we all.
)
http://www.rigrite.com/
Defender also does rigging at good prices. Many shops like you to mail in your old shrouds and stays, to prevent measuring error.