In his defense, he was shooting a star trail. @10s per exposure, 6 per minute, becomes 360 per hour.

I believe 30s might be more appropriate for star trails which makes satellite trails oh so much more obvious. But that means a 2-3 hour session of 2 exposures per minute.

I also used to take hundreds of photos in meteor season - and having to diff between meteors and satellites was quite time consuming.

in photos, meteors and satellites do not look the same. satellite trails are consistent lines. meteors start small, flare in the middle, and then ends small again. they are easy for humans to distinguish, so i'd assume some software could do it as well

These photos are taken in pitch-black darkness.

You need to take a lot of exposures in order to get the data necessary to even see anything.

This is a single 10sec exposure in the Aussie outback on old consumer gear ( Sony a7iii )

https://www.instagram.com/p/CersLuLBfCz

You don’t need multiples, and you don’t need an overly long exposure.

It sounds like you know what you're talking about until one realizes the earth is spinning. Wide field photos can be shot up to thirty seconds depending on the back and lens.

Anything more zoomy than 50mm uncropped you're getting streaks in < dozen seconds. There's a rule of thumb but I don't remember it.

Best course of action is to take a video and let a stacking program deal with it, especially if you use a real telescope.

Also the Sony a7r have like "150,000 ISO " and iirc cost like $3500 with a kit lens. That's a bit above consumer, but I may have mixed up models.

My camera is not an a7r. I paid $1200 for it used.

The exposure was 10 seconds, so by your own explanation, the spinning of the earth is not a problem ( as you can clearly see in the photo I linked )

An astronomy photo can commonly require hours or even tens of hours of exposure time.

The noise from a single image of that duration would be unworkable. Even if there were no planes or satellites, nobody would take that as a single exposure.

Signal scales with exposure time, noise scales with the root of exposure time. So although the noise increases with time, the signal-noise ratio increases faster. When dealing with astronomy photos, you want to maximise SNR.

(This is a simplification, there are many types of noise including shot noise and thermal noise, but in general this is the rule of thumb you use)

your method of shooting the stars with a single long exposure is called a "wide-field" image. They're cool, but if you want to get any details of anything other than the milky way you either have to track or you have to stack. that's all there is to it.

I was going off memory, and with 50mm, it looks like 10 seconds is about when you start to get streaks, assuming no cropping. I have cropped sensors so even with a 10mm lens i can only do 12-15 seconds. I have a barn door tracker, and i know how to use https://www.autostakkert.com/

you need to have black frames for it to work, and ideally a lot of them (iirc at least a half dozen if not more) - because this is how the stacker knows which pixels are noisier than other ones, and also what the thermal noise of the sensor "looks like", at least the smarter stackers use it that way. you take a bunch of pictures with the lens cap on in a dark room at around the temperature you're gunna use it at.

real amateur astrophotography is done with CCD cameras, usually monochrome with filters (or a filter wheel), done in video mode, and the frames are processed after the fact. It takes hours with clear skies to make a decent deep field astrophotograph.

Tens of seconds is about right. It's something like 500/f(mm) in seconds, but you get a feel for what will blur and what won't. Here's an example I shot with a Nikkor 50mm f/1.4 @ 20s: https://www.instagram.com/p/C-mU6iIp0re/?igsh=cm16bWx1cGp3OG...

Rule of 500. Divide your lens focal length into 500. That's the longest exposure to avoid trails. 500 / 24mm = ~21s so your example would be 500 / 50mm = 10s.

Some random search result:

https://astrobackyard.com/the-500-rule/

No, you need to take a long exposure. Multiple exposures may improve the quality but isn’t necessary at all.

If the camera is stationary you must stack multiple short exposures to avoid star trails. You can only take long exposures with an equatorial mount.

Not if your “long” exposure is just 10s long, which is enough to get those annoying starlink trails.

Sure, 10s is certainly long enough to capture Starlink trails, though 10s exposure is not especially long for an astro photo and you would typically want to take longer exposures or stack multiple shorter ones.

For example, my Seestar S50 takes many 10-second photos to form one exposure. I would normally expect to take 1500 10-second exposures or more for a good picture, perhaps over mutiple nights (although this is very different to a wide-field image like this). A Starlink trail would be very visible in any individual frame, often brighter than stars, and I would expect to capture many individual frames containing starlink trails.

This is not true. Image in article is taken (and stack from many images) during sunset/sunrise when satellites are lighted by sun while photographer is in dark. It's not that visible at night.