94
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 01 Sep 2024
94 points (89.2% liked)
science
29054 readers
126 users here now
A community to post scientific articles, news, and civil discussion.
dart board;; science bs
rule #1: be kind
founded 3 years ago
MODERATORS
Not how it works. They are not taking pictures with cameras. Resolution has more meaning then pixels.
What I mean is a higher resolution than the linked site provides. NASA has an “original” image of the last time they made an image like this but I haven’t found one for this new image yet. That original is 4000x2330 but the one linked in the OP site is only 1200x729.
wdym? how was this taken
Radio telescopes. While I don't know the complete process of how an image is created, it's likely a composite of hundreds of thousands of points where radio wave strength was measured.
A very basic explanation is that each radio antenna likely takes a reading of some kind for each equivalent pixel in the resulting image. Over time, you can build an image.
Again, I don't know the full details of how the full image is recreated. It seems super complex reassembling millions of data points from antennas that are located on a rotating earth that is also rotating around a sun. The position of the earth probably has a huge impact on radio signal strength at any given time.
They could be very well using the earth's orbit around the sun to get better resolution - two data points from opposite sides of the orbit. What I know is that the largest "virtual" radiotelescope is literally the size of earth. The data points are synced with atomic clocks (or better), and a container of harddrives gets shipped into a datacenter to be ingested. Thats hundreds of streams (one per antenna) of data to be just synced up, before the actual analysis even can begin. (I'm just guessing after this) At this point, you have those hundreds (basically .wav files) lined up at timepoints they were sampled (one sample, one timepoint column). So row by row, so you can begin to sort out signal phase differences between the source rows.
I.e to put it shortly: an image is not taken, it is inferred and computed. Not that you even could in the first place, it's a blackhole after all.