[SteelyBob]

Tom, I'd be grateful if you could answer my question from a few posts back, repeated below for convenience.

Tom, in all the chat about rabbits and FOVs, I'm still unclear on one thing - are you standing by your statement that stars, in both real life and in Stellarium, do not move in circles around the pole stars? That their angular separation, in both real life and in Stellarium and other tools, does not remain constant?

I have to say, that picture you showed measuring pixel distances is one of the most absurd arguments I've seen on the internet for a long time. The normally straight-line silhouette features on the horizon wrapping around in an obviously distorted circle.. and then taking a linear measurement on the same picture in two different situations...and expecting to be taken seriously? Astonishing.

[JSS]

Maybe a little inherent distortion, but since lines on buildings and structures appear straight beyond anything that is easily detectable in quality rectilinear lenses, it's irrelevant.

It's not irrelevant. As I showed with my example of the blinds, you can't just find a straight line in one part of an image and declare the rest of the image is free of distortions.

If you do not know the exact specifications of the lens that took the picture, if you do not know what was done to it in post production, you can not determine what is distorted and what is not.

That is why all your examples pulled off the internet are no good, you simply can't prove they are distortion free. You can claim the stars are distorted and not the image, but you have no evidence to support that claim.

[Tom Bishop]

As I showed with my example of the blinds, you can't just find a straight line in one part of an image and declare the rest of the image is free of distortions.

It wasn't just one part of the image. There were lines all throughout the images posted.

The straight lines were not centered in the picture in the image of the farm provided. The lines on the elements of the farm were not centered. It appears that you are unable to adequately demonstrate your assertion.

Another rectilinear image. Lines are straight throughout the image:

https://fineartamerica.com/featured/mosquito-creek-star-trails-willard-sharp.html



Lines are straight:



Yet it is easy to see that the curves are not concentric circles:

Shift + Circle Tool in Paint.net:

And nor are the lines on this farmhouse centered on the image.

Here is an example to try, with some nice straight lines in the foreground.

https://www.reddit.com/r/BeAmazed/comments/50ziy3/long_exposure_of_star_trails_against_a_farmhouse/



Hold shift with the circle tool in paint.net for a symmetrical circle. The curves just aren't arcs of a circle, and nor are they concentric:

[Iceman]

Same photographer, same bridge... straight lines... still see distortion

https://www.google.ca/imgres?imgurl=https%3A%2F%2Fimages.fineartamerica.com%2Fimages%2Fartworkimages%2Fmediumlarge%2F1%2Fmosquito-creek-star-trail-2-willard-sharp.jpg&imgrefurl=https%3A%2F%2Fpixels.com%2Ffeatured%2Fmosquito-creek-star-trail-2-willard-sharp.html&tbnid=WjdDEbaWkt48bM&vet=12ahUKEwiX06Ln6cvuAhVKSawKHfHoApwQMygQegQIARB_..i&docid=_rnCygLUqSiWjM&w=900&h=600&itg=1&q=mosquito%20creek%20%20bridge%20iowa&ved=2ahUKEwiX06Ln6cvuAhVKSawKHfHoApwQMygQegQIARB_

[Tom Bishop]

Same photographer, same bridge... straight lines... still see distortion

https://www.google.ca/imgres?imgurl=https%3A%2F%2Fimages.fineartamerica.com%2Fimages%2Fartworkimages%2Fmediumlarge%2F1%2Fmosquito-creek-star-trail-2-willard-sharp.jpg&imgrefurl=https%3A%2F%2Fpixels.com%2Ffeatured%2Fmosquito-creek-star-trail-2-willard-sharp.html&tbnid=WjdDEbaWkt48bM&vet=12ahUKEwiX06Ln6cvuAhVKSawKHfHoApwQMygQegQIARB_..i&docid=_rnCygLUqSiWjM&w=900&h=600&itg=1&q=mosquito%20creek%20%20bridge%20iowa&ved=2ahUKEwiX06Ln6cvuAhVKSawKHfHoApwQMygQegQIARB_

The straight lines throughout the image show that it's taken with a rectilinear lens. If the lens was causing distortion, it would appear in the lines.

If there is distortion, it's caused by something else like resizing the image and not keeping a proper ratio, and nothing to do with the inability of a rectilinear lens to capture the shapes of images.

It might also be that there is no radical distortion in such images and that the stars don't always appear to be as circular from one scene to the next, depending on location and time of day.

[JSS]

As I showed with my example of the blinds, you can't just find a straight line in one part of an image and declare the rest of the image is free of distortions.

It wasn't just one part of the image. There were lines all throughout the images posted.

Listen carefully to what I am saying.

You can not use a straight line in one part of an unknown image to determine the distortion in another part.

You measured a few lines, none of which even go the entirety of the image, none of which you proved were perfectly straight in reality, and then tried to claim the image is distortion free. This does not work.

You can not measure spot A and prove spot B has the same distortion, or lack of it.  Especially as you have no idea what post-processing may have been done.

Sorry, but none of those images are useful at all as proof.  That's just not how lenses work.

[Tom Bishop]

As I showed with my example of the blinds, you can't just find a straight line in one part of an image and declare the rest of the image is free of distortions.

It wasn't just one part of the image. There were lines all throughout the images posted.

Listen carefully to what I am saying.

You can not use a straight line in one part of an unknown image to determine the distortion in another part.

You measured a few lines, none of which even go the entirety of the image, none of which you proved were perfectly straight in reality, and then tried to claim the image is distortion free. This does not work.

You can not measure spot A and prove spot B has the same distortion, or lack of it.  Especially as you have no idea what post-processing may have been done.

Sorry, but none of those images are useful at all as proof.  That's just not how lenses work.

The image of the blinds you posted showed that the lines were straight only through the center of the image. All other places the lines were warped. If the a lens was warped, it should be detectable in lines in the image.

You have not shown by way of example that straight lines can be scattered throughout an image of a fish-eye lens.

[Iceman]

The straight lines throughout the image show that it's taken with a rectilinear lens. If the lens was causing distortion, it would appear in the lines.

If there is distortion, it's caused by something else like resizing the image and not keeping a proper ratio, and nothing to do with the inability of a rectilinear lens to capture the shapes of images.

It might also be that there is no radical distortion in such images and that the stars don't always appear to be as circular from one scene to the next, depending on location and time of day.

Arent you basically arguing JSS' point here?
I've provided an image where we dont know about the lens or anything that has been done to the photo in post-processing. You've now (rightly) provided a list of potential issues with the image...

[Tom Bishop]

The straight lines throughout the image show that it's taken with a rectilinear lens. If the lens was causing distortion, it would appear in the lines.

If there is distortion, it's caused by something else like resizing the image and not keeping a proper ratio, and nothing to do with the inability of a rectilinear lens to capture the shapes of images.

It might also be that there is no radical distortion in such images and that the stars don't always appear to be as circular from one scene to the next, depending on location and time of day.

Arent you basically arguing JSS' point here?
I've provided an image where we dont know about the lens or anything that has been done to the photo in post-processing. You've now (rightly) provided a list of potential issues with the image...

We've been mainly talking about lenses here. I haven't said much about possible post-production effects. Stretching the image in photoshop is possible. It's also possible that the stars in star trails which do seem circular were edited to be more circular because that is the ideal result. Showing that we have no good evidence as originally claimed:

I haven't seen much evidence that the stars move in concentric circles.

[SteelyBob]

Tom, I'd be grateful if you could answer my question from a few posts back, repeated below for convenience.

Tom, in all the chat about rabbits and FOVs, I'm still unclear on one thing - are you standing by your statement that stars, in both real life and in Stellarium, do not move in circles around the pole stars? That their angular separation, in both real life and in Stellarium and other tools, does not remain constant?

I have to say, that picture you showed measuring pixel distances is one of the most absurd arguments I've seen on the internet for a long time. The normally straight-line silhouette features on the horizon wrapping around in an obviously distorted circle.. and then taking a linear measurement on the same picture in two different situations...and expecting to be taken seriously? Astonishing.

[Tom Bishop]

I've already answered that. I said that I haven't seen much evidence that the stars make perfect concentric circles.

[SteelyBob]

I've already answered that. I said that I haven't seen much evidence that the stars make perfect concentric circles.

..and you used a picture from Stellarium where you measured the pixels to try to prove your point, saying:

Both the distances and angles between the stars change throughout the night in stellarium.

Do you now agree that this was a flawed methodology to use, or are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate?

[Tom Bishop]

I've already answered that. I said that I haven't seen much evidence that the stars make perfect concentric circles.

..and you used a picture from Stellarium where you measured the pixels to try to prove your point, saying:

Both the distances and angles between the stars change throughout the night in stellarium.

Do you now agree that this was a flawed methodology to use, or are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate?

I haven't seen circular star trails in Stellarium, and nor would Stellarium be a real world observation enough to count as compelling evidence for perfectly circular star trails.

[SteelyBob]

I've already answered that. I said that I haven't seen much evidence that the stars make perfect concentric circles.

..and you used a picture from Stellarium where you measured the pixels to try to prove your point, saying:

Both the distances and angles between the stars change throughout the night in stellarium.

Do you now agree that this was a flawed methodology to use, or are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate?

I haven't seen circular star trails in Stellarium, and nor would Stellarium be a real world observation enough to count as compelling evidence for perfectly circular star trails.

Are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate? It's a really simple question.

[JSS]

As I showed with my example of the blinds, you can't just find a straight line in one part of an image and declare the rest of the image is free of distortions.

It wasn't just one part of the image. There were lines all throughout the images posted.

Listen carefully to what I am saying.

You can not use a straight line in one part of an unknown image to determine the distortion in another part.

You measured a few lines, none of which even go the entirety of the image, none of which you proved were perfectly straight in reality, and then tried to claim the image is distortion free. This does not work.

You can not measure spot A and prove spot B has the same distortion, or lack of it.  Especially as you have no idea what post-processing may have been done.

Sorry, but none of those images are useful at all as proof.  That's just not how lenses work.

The image of the blinds you posted showed that the lines were straight only through the center of the image. All other places the lines were warped. If the a lens was warped, it should be detectable in lines in the image.

You have not shown by way of example that straight lines can be scattered throughout an image of a fish-eye lens.

Yes, that's what an extreme fisheye does. But many lenses, in fact most have only minor distortion in the center, and get worse at the edges and especially the corners.

Your 'method' of trying to draw straight lines in a few places doesn't show that the corners aren't warped.

They also don't show large scale warping across the entire image, where small local areas won't have enough to be detectable, lines crossing the entire field will show distortion.

Like the star trails that cover nearly the entire image.

[JSS]

I haven't seen circular star trails in Stellarium, and nor would Stellarium be a real world observation enough to count as compelling evidence for perfectly circular star trails.

You should learn to correctly operate the software you use to try and prove your points.

As has been stated many times, the circular star trails produced by Stellarium are indeed compelling evidence since you can use that software to predict the position of the stars anywhere in the world, take pictures and compare them with yours. They will match. That's pretty compelling.

[Tom Bishop]

Yes, that's what an extreme fisheye does. But many lenses, in fact most have only minor distortion in the center, and get worse at the edges and especially the corners.

There were straight lines all across the entire image at various angles. You keep repeating your weak argument.

I haven't seen circular star trails in Stellarium, and nor would Stellarium be a real world observation enough to count as compelling evidence for perfectly circular star trails.

You should learn to correctly operate the software you use to try and prove your points.

As has been stated many times, the circular star trails produced by Stellarium are indeed compelling evidence since you can use that software to predict the position of the stars anywhere in the world, take pictures and compare them with yours. They will match. That's pretty compelling.

I just see multiple claims here. I haven't seen demonstration of these claims. I look forward to you demonstrating each and every one of your claims beyond dispute.

[SteelyBob]

Are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate? It's a really simple question.

[Tom Bishop]

Are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate? It's a really simple question.

I haven't seen any evidence that they do. If you think so, I look forward to you providing the compelling evidence for your topic of interest.

[SteelyBob]

Are you standing by your point that Stellarium shows that the stars don't maintain their angular separation as they rotate? It's a really simple question.

I haven't seen any evidence that they do. If you think so, I look forward to you providing the compelling evidence for your topic of interest.

That’s not what you said though, is it?

You said they didn’t, and you offered a laughably pathetic attempt at science to justify your statement.

That’s not the same thing as saying you haven’t seen evidence that they do, is it?

So, again, are you standing by your statement, or are you admitting you were wrong?