f68r: eclipse from 1419?

[denty]

Summary Table of the Medieval Workflow for Calculating an Eclipse from the Toledo Tables:

Step                                               Purpose                                      Toledo Tables Used

1. Mean syzygy                                Identify candidate eclipse             Mean motions
2. Equations of center                       Correct Sun/Moon positions          Solar & lunar anomaly tables
3. True syzygy                                  Exact conjunction/opposition        Mean motions + corrections
4. Lunar latitude                               Determine if eclipse possible         Latitude tables
5. Parallax                                        Local visibility                             Parallax tables
6. Magnitude                                    Size of eclipse                             Radii tables
7. Local circumstances                       Local timing                                Longitude/time tables

Thanks!   So, from that "workflow", I would guess that they could predict quite accurately when the next eclipse would occur; but the errors in the tables would cause prediction to fail beyond a certain point because the three bodies would not be sufficiently aligned on that date and/or the path of the eclipse would be inaccessible.  

That is, the eclipse would not happen, but the non-happening would happen precisely when predicted. 

Makes sense?

All the best, --stolfi

Not necessarily precisely, as we have seen, the 1419 March solar eclipse calculation was out by 2 days too early. So I would say for a solar eclipse:
7. Local circumstances: Look at the moon, is it new? Then it is happening, (whether or not it is visible from your location) otherwise wait for the new moon. 
For a lunar eclipse, substitute full moon.