[Kstars-devel] Some HTM progress

[Jason Harris]

Hi,

James Bowlin wrote:
> So the rule is simple.  If we give ourselves a safety margin of 10 
> arc-minutes then we need to re-index a moving object before it moves
> 10 arc-minutes compared to the fixed stars.
> 
> Please correct me if I am wrong.
> 
Sounds right.  We may have to come up with something to deal with large 
timesteps, such as offloading the solar system objects out of the HTM in 
that case.  Again, this is something we can deal with later.

> This raises the question of what format to use for triangles in the
> HIP files.  Their cc_name2ID() code looks very fast.  Therefore I think
> we should store the triangle ID as an "S01230123" string, perhaps with a
> few more digits than we are planning to use.  Yes, this is the way to go.
> It will make things much easier if we want to use more than one level at
> the same time.
> 
Agreed.

> It sounds like we have a general game plan now for the fixed stars.
> We have yet to determine:
> 
>  1) whether to use circles or rectangles for getting the visible triangles,
>  2) exactly what lowest level to use, and
>  3) whether it will be worthwhile to use more than one level.
> 
> All of these we can nail down with some simple experiments.
> 
Great.

> BTW: They expect declination and right ascension in degrees.
> 
That's not a problem at all; the RA and Dec are read into dms objects, 
which are stored internally in degrees.

regards,
Jason