by MarkBour » Thu Aug 09, 2018 6:06 pm
Chris Peterson wrote: ↑Wed Aug 08, 2018 10:13 pm
MarkBour wrote: ↑Wed Aug 08, 2018 8:52 pm
I love this animation. It either "stood on the shoulders of giants", or took a lot of work to produce, I think. If I am not mistaken, every represented object in the foreground (except for the Sun, probably) is following a fine Keplerian orbit. And these are no doubt carefully built on current science data (This is
Sparta! APOD!). Perhaps there is massive calculation going on to render those orbit positions.
There are thousands of meteor cameras operating every night around the world. When you have a meteor on any pair of them (or more), the orbit is automatically calculated. The CAMS system that collected this data is particularly high resolution, since it utilizes multiple cameras. But the technique is the same. It's not a terribly complex calculation to derive an orbit from a meteor track, and there are now libraries with millions of calculated orbits.
Wow, I didn't know they number in the thousands, nor that the number of calculated orbits would now be in the millions.
I know that you contribute to this with meteor cams and computing that you have at cloudbait and elsewhere.
Chris Peterson wrote: ↑Wed Aug 08, 2018 10:13 pm
That data goes into another fairly straightforward routine that renders a visual orbit from a set of orbital elements. There's a lot of data, which means a lot of calculations are going on, but that's all.
I guess you're right to say: "fairly straightforward routine", and I suppose to animators this is all in a day's work. I don't know if there needs to be any specialization of rendering software to handle distances of millions of kilometers versus a room or a city block. Perhaps not. But it is still marvelous to me, for sure.
One thing I assume the animation software did *not* do was to compute the time delay for light to travel from each object to the observer. That is one thing that would differ in an animation on this scale versus a scene in a room. I assume we would rather see them where we think they "really are" and not add that complication to the rendering.
Chris Peterson wrote: ↑Wed Aug 08, 2018 10:13 pm
BTW, there is some confusing language in today's caption. This is not the entire debris stream, but rather, particles which have already been destroyed by their collision with Earth's atmosphere- only the tiniest fraction of the entire debris stream. In reality, none of these particles should be seen surviving their encounter with Earth and coming out the other side. We're seeing their orbits as if the Earth were not there.
Also interesting that these are all ghosts! The Earth is at work, cleaning out Perseids, Leonids, etc. (doing its job as a planet). I don't suppose one could detect any "cleaned up" regions in the streams ... there is probably a lot of randomization filling back in.
[quote="Chris Peterson" post_id=284833 time=1533766425 user_id=117706]
[quote=MarkBour post_id=284832 time=1533761544 user_id=141361]
I love this animation. It either "stood on the shoulders of giants", or took a lot of work to produce, I think. If I am not mistaken, every represented object in the foreground (except for the Sun, probably) is following a fine Keplerian orbit. And these are no doubt carefully built on current science data (This is [s]Sparta![/s] APOD!). Perhaps there is massive calculation going on to render those orbit positions.
[/quote]
There are thousands of meteor cameras operating every night around the world. When you have a meteor on any pair of them (or more), the orbit is automatically calculated. The CAMS system that collected this data is particularly high resolution, since it utilizes multiple cameras. But the technique is the same. It's not a terribly complex calculation to derive an orbit from a meteor track, and there are now libraries with millions of calculated orbits.
[/quote]
Wow, I didn't know they number in the thousands, nor that the number of calculated orbits would now be in the millions.
I know that you contribute to this with meteor cams and computing that you have at cloudbait and elsewhere.
[quote="Chris Peterson" post_id=284833 time=1533766425 user_id=117706]
That data goes into another fairly straightforward routine that renders a visual orbit from a set of orbital elements. There's a lot of data, which means a lot of calculations are going on, but that's all.
[/quote]
I guess you're right to say: "fairly straightforward routine", and I suppose to animators this is all in a day's work. I don't know if there needs to be any specialization of rendering software to handle distances of millions of kilometers versus a room or a city block. Perhaps not. But it is still marvelous to me, for sure.
One thing I assume the animation software did *not* do was to compute the time delay for light to travel from each object to the observer. That is one thing that would differ in an animation on this scale versus a scene in a room. I assume we would rather see them where we think they "really are" and not add that complication to the rendering.
[quote="Chris Peterson" post_id=284833 time=1533766425 user_id=117706]
BTW, there is some confusing language in today's caption. This is not the entire debris stream, but rather, particles which have already been destroyed by their collision with Earth's atmosphere- only the tiniest fraction of the entire debris stream. In reality, none of these particles should be seen surviving their encounter with Earth and coming out the other side. We're seeing their orbits as if the Earth were not there.
[/quote]
Also interesting that these are all ghosts! The Earth is at work, cleaning out Perseids, Leonids, etc. (doing its job as a planet). I don't suppose one could detect any "cleaned up" regions in the streams ... there is probably a lot of randomization filling back in.