You are reading a single comment by @Numerist and its replies.
Click here to read the full conversation.
-
Interesting! My first experiment with 210 ms/hr ended after 16 hours. The watch gained ⅓ to ½ a second, so about 600 ms/24 hrs. I don't know how linear that was. Success, regardless. While awaiting your further developments, I'll try a slightly high number, going towards your 0.68 figure. Then I may try halving it for every 30 minutes, although of course that gets close to 100 ms.
Numerist
Actually I just got an idea to make delay+setTime corrections more accurate. Main problem with delay+setTime is that delays aren't accurate, but what if I check current time again after the delay to see how long the delay actually was? I'll need to test this a bit before sharing any code.
ps .While more accurate
setTimewould be nice, I'm not sure if that is actually needed.