Both tails are not cones, rather thin paraboloids, with the nucleus in
focus. The gas tail (bluish) should point straight away from the sun, the
dust tail lags behind a bit, so you need to find not only position but
speed (vector), and use this to push the tail from the direct antisolar
direction. Anti-tails should be visible then just from observer/viewing
geometry. The textures for those parabolas may be animated to move away
from the coma and thin out. also, the ice/dust/gas "particles" coming from
the vents should be pushed away from the sun.
On Di, 1.05.2012, 03:30, Reaves, Timothy wrote:
> As a general rule, as accurate as possible.
> As you zoom in, the tail would still be visible. For the boy itself, there
> are some OBJ files available for some of the more major comets. For ones
> where these are not available, perhaps there is a library of OBJ files
> could be sleected from based on composition, or some-such. As for seeing
> venting, that would definitely be nice.
> For the other questions, 'yes'. :)
> On Mon, Apr 30, 2012 at 2:28 PM, Kristen Aw <jiayuaw@...> wrote:
>> I was just wondering about the level of detail in comet rendering that
>> Stellarium wants. Currently, from afar, the comets are balls of light,
>> when zoomed into, are rotating spheres with some crater-map. I
>> that the plasma tail comes out cone-shaped, and appears at 1.5 AU, the
>> tail appears at about 3 AU, and the coma is visible when it passes
>> (5 AU). And that the brightness of the coma/ length of tail increases as
>> the comet approaches the Sun.
>> So, from afar, a comet would have a coma and tails. But what about when
>> they are zoomed into? I guess it would have an irregular potato shape
>> Hartley, but as it releases gas/dust, should we be able the material
>> out from vents, as the nucleus rotates...?
>> And also, should we be able to calculate if the observer sees an
>> (due to parallax), and whether the dust tail is viewed head on (it would
>> seen as a fan on a plane) or directly from the side (it would be seen as
>> straight line)?
>> I'm not sure how feasible the above is, so I'm just asking.