Thread

1. mmooss · 1790972229 · focus · HN ↗

   From ~0:04 - 0:05, two dots at ~10 and 8 o'clock fade in and out simultaneously, with roughly the size and brightness of the planets. They peak at observation ~2016-07-06. Any idea what they are?

   Their balanced position and simultaneous changes make them seem like an artifact of the imaging.

   1. hatthew · 1790978753 · focus · HN ↗

      It looks like they're only present for one real frame (of the original 10), so very possible that it's just noise in that frame.

      1. mmooss · 1790989814 · focus · HN ↗

         They're present for multiple dates at the bottom, and fade in and out. Are those dates and the fading fabricated as some sort of intermediate state and inserted into the animation? I did briefly look for something stating how the animation was made but didn't see anything.

         Also, they are too regular in position and in their timeing (simultaneous) to be random noise, but could be an artifact of some part of the imaging and processing chain.

         1. hatthew · 1790991212 · focus · HN ↗

            Yes, they were not very clear about this but my understanding is that there are only 10 real images, i.e. less than one per (earth) year. All other frames are interpolated, which can pretty accurately animate the positions of the planets, but is presumably completely inaccurate for anything else.
         2. wthomp · 1791051948 · focus · HN ↗

            As responded above, they were persistent diffraction errors in the primary mirror phasing. Think of it like those diffraction spikes from Hubble.

            Indeed the video is almost entirely “fabricated” by motion smoothing and interpolation between images taken a year or two apart.

            1. mmooss · 1791060291 · focus · HN ↗

               Thanks for the animation, which is amazing to see, and for your responses!

               > Indeed the video is almost entirely “fabricated” by motion smoothing and interpolation between images taken a year or two apart.

               That's fine but you might say that in your post (and maybe you did and I overlooked it). There's nothing wrong with it - the animation is great, just be open. In a world of misinformation and disinformation, it's more essential than ever.

               The professional space community publishes so much artistically enhanced media that for me, it's guilty until proven innnocent. When I do trust it, like in this case (yes, I didn't really stop to think about the number of actual frames), I feel a bit deceived.

               And the vast public less experienced in these matters already has doubts about science, the space program, conspiracy theories, etc. When they learn something is 'faked', we lose them.

   2. wthomp · 1791051852 · focus · HN ↗

      Good spot! It’s an error in the primary mirror segment phasing.

      The observatory has to phase the mirror segments periodically, and it’s a bit of an art as well as a science. For a while, something was happening with their algorithm and procedure that was leaving those spots. Some of us (hi) did complain and it was fixed for the most part. It did come back a couple times though….

      The slightly incorrect phasing produces those spots. It’s conceptually like a diffraction spike.

Unofficial Hacker News client; not affiliated with Y Combinator.