The serpent in my blade was done by twisting, a tight series of left and right 90° twists which I then re-squared on the corner. This gives you the sine wave on two faces, and opposing “c” figures on the other two. GEzell gets the closest-approximation prize!
Since the dark age smiths relied on twisting for most of their patterning, and did a lot of alternate twisting of bars, it fits with the processes in use at the time and wastes no more material than a solid twist. The pattern could have been done by some of those other methods, but they seem less likely to have been used "back in the day" IMHO.I did a sword back in 2002 with a central bar of back-and-forth 180° twists, thinking that was the answer, then worked out how to REALLY do it on some practice bars, one of which I still have kicking around – I’ll take a photo so you can see what the bar looks like as you twist it.
